    Ŀ Ŀ     Ŀ Ŀ Ŀ Ŀ Ŀ Ŀ Ŀ
    Ŀ   Ŀ        Ŀ       Ŀ     Ŀ 
                 Ŀ                        
     Ŀ           Ŀ                         
                               
                   

                              Quick Documentation



1. Shortcuts and hotkeys
2. IPAS routines
3. Batch rendering
4. 3DS.SET file
5. List of advanced tips and techniques
6. Boolean Operations hints
7. Review of 3DS 2.0 - Lists speicifications and features.
8. Tutorial on the use of the Keyframer for animation
9. Article on doing an animated walkthrough
10. Beveled text Tutorial
11. Quick Tutorial on 2D Editor, 3D Lofter, and 3D Editor  
12. Presentation of animation and stills for Airport Proposal




1. S H O R T C U T S  A N D  H O T K E Y S

3D Editor Module

(Other modules have virtually identical hotkeys)

Keyboard shortcuts:

?       Current Status
*       Configure
`       Redraw selected viewport
~       Redraw all viewports
[       Save current to hold buffer
{       Restore saved from hold buffer
A       Angle snap
B       Switch selected viewport to Bottom view
C       Switch selected viewport to Camera view
D       Disable viewport
E       Grid extents
F       Switch viewport to Front view
G       Grid on/off in viewport
H       "Hit" -  select an object with a list (as opposed to picking it 
        with the mouse). 
K       Switch selected viewport to Back view
L       Switch selected viewport to Left view
N       New
Q       Quit
R       Switch selected viewport to Right side view
S       Snap on/off in viewport
T       Switch selected viewport to Top view
U       Switch selected viewport to User view
V       Fastview mode toggle in selected viewport

F1      2D Shaper Module
F2      3D Lofter Module
F3      3D Editor Module
F4      Keyframer Module
F5      Materials Editor Module
F6      Box PXP
F7      Grids PXP
F8      Waves PXP
F9      Ripple effect PXP
F10     Dos shell

Alt-F1  /   Optional
Alt-F2  |   Add-on IPAS PXP
Alt-F3  |   Modules 
.       |   show
.       |   up
Alt-F10 \   here.

Alt-A   Select All
Alt-B   View in block/normal mode
Alt-C   Hide/Unhide camera
Alt-E   Safe Frame
Alt-G   See Background in view
Alt-L   Hide/Unhide Lights
Alt-N   Select None
Alt-P   Hide/Unhide construction plane
Alt-R   Render selected view
Alt-T   Hide/Unhide ruler
Alt-V   Vertex snap
Alt-X   Hide/Unhide axis
Alt-Z   Zoom (same as icon)

Alt-box icon        -- Zoom all viewports to fit all objects (called zoom
                       extents)
Alt-zoom out icon   -- Zoom all viewports out
Alt-zoom in icon    -- Zoom all viewports in

Ctrl-A  Drawing Aids
Ctrl-D  Delete a file
Ctrl-J  Load a project
Ctrl-L  Load a file
Ctrl-M  Merge
Ctrl-P  Save a project
Ctrl-R  Replace mesh
Ctrl-S  Save 
Ctrl-U  Units setup
Ctrl-V  Viewports setup
Ctrl-Z  Zoom to fit all objects (zoom extents)

Icons:




2. I P A S  R O U T I N E S

External IPAS routines:

IXP     Image Processing external process (like Stars). Accessed in 
-       Keyframer Video Post. Acts on images after rendering. These can
        be thought of as intelligent 2-D processes. They act like a paint
        program that knows what to paint and where, after the frame has
        been rendered. Glowing rockets and blinding lens flares are 
        animated using IXPs. 
        
PXP     Modeling External program, (like Silicon Garden). Installed by 
-       modifying 3DS.SET and accessed by top menu pulldown or assigned 
        F key. This external system allows third parties to write programs
        that make objects or object modifications externally and then put 
        the results back into 3DS. Try using Ripple or Wave to get the 
        idea of the seamless integration. Version 2.01 works much more
        quickly with PXPs than 2.0.  
        
AXP     Animated Stand-In Object external process (like Tornado). Accessed 
-       in Modify Object Attributes. These are quite unique, in that they
        allow a dummy object (like a cube) to be used for the animation of
        a more complex thing, usually a particle system. Tornados, rain,
        snow, fire and other tough-to-animate things can be tackled this
        way (after you purchase the third-party AXP to do them). 
        
SXP     Surface procedural routine (like Checker). Accessed in Materials 
-       editor in Texture Map or Bump Map filename. They exist because 
        sometimes mapping a 2D image over a 3D object looks unacceptable 
        (lots of streakiness over areas perpendicular to the mapping normal). 
        SXP's are procedural textures. When invoked, the surface color or
        surface normal is created with formula that is truly 3 dimensional,
        rather than just wrapping an object with a 2D picture. They use the 
        3D coordinates and a mathematical algorithm to procedurally create 
        the color or "bumpiness" of the object they are applied to. Often,
        a much more realistic image can result from effective use of SXPs.

Note that a set of developmental tools for the creation of IPAS routines is
being made available by the Yost Group. Because these tools are becoming
widely available, one can expect to see many new IPAS routines from third 
party vendors. Several public domain routines are already being distributed.
Eventually most of us will have to choose which 10 PXPs to have resident in 
our 3DS.SET!


3. B A T C H  R E N D E R I N G

Batch rendering does not require the hardware key to be present.

Command line batch mode:

Usage:
        3DS RENDER FILENAME[.3DS] [Options]
or      3DS RENDER @OPTFILE.EXT

Options:
   /A<float>       -- Aspect ratio override
   /B<filename>    -- Background bitmap override
   /C<op><value>   -- Cast shadow ops: A=Area S=Samples
                       L=Lowbias H=Highbias N=No Shadows
   /D<name>        -- Device driver name
   /E<option>      -- Mapping options: N=None
   /F<type><comp>  -- Pic file type: G=GIF T=Targa C=Ctif M=Mtif F=Flic
                      comp: C=Compressed U=Uncompressed
   /H<int>         -- Device height override
   /I<filename>    -- Filename to save generated image to
   /L<filename>    -- Save last image to specified file
   /M              -- Make hidden objects visible
   /N              -- Render FIELDS
   /O<type>        -- Oversampling: N=None L=Low M=Med H=High
                       modifiers: F=Filter, T=Turbo, e.g. /OMFT
   /P<int>         -- Antialiasing threshold
   /Q<float>       -- Texture blur
   /R              -- Resize background (default=Tile)
   /S<mode>        -- Shading mode: W=Wire F=Flat G=Gouraud P=Phong
   /T<command>     -- Tape commands: F=File H=Heads I=Inpoint
                        L=DTV Loops N=Nframes T=Tails
   /V<filename>    -- .VUE file name
   /W<int>         -- Device width override
   /Z<float>       -- Device zoom override (none=1.0)
   /8<type>[file]  -- 8-bit palette type: L=Low, M=Medium,
                       H=High, C=Custom palette file
   /#<s>,<e>,<stp> -- Frame start, end, step


4. T H E  3 D S . S E T  F I L E 

3DS.SET setup initialization file:

; 3D Studio system set-up parameter file:  Release 2
;
; This file contains information used by the 3D Studio system (3DS.EXE).
; It provides information on where key files are located as well as user
;  customization parameters.

; Note that the most-often-used commands in this file are also selectable
;  from inside the 3DSr2 user interface.

; For detailed information, consult the 3D Studio Release 2 
;  Installation Guide.

; Disk path defaults.
; All of these paths will default to the 3DS origin subdirectory if they 
;  are not specified.

; Default path for geometry.
SHAPE-PATH = "shapes"
LOFT-PATH = "lofts"
MESH-PATH = "meshes"
FONT-PATH = "fonts"
PROJECT-PATH = "projects"

; Default path for ADI drivers (with the .exp extension).
DRIVER-PATH = "drivers"

; Default path for the View Flic option to look into.
FLIC-PATH = "flics"

; Default path for the keyframer to write preview flics to.
PREVIEW-PATH = "flics"

; Default path for all images and animations to be output by the system.
IMAGE-PATH = "images"

; Default path for storing material libraries in.
MAT-PATH = "matlibs"

; Default material library.  Must be in your MAT-PATH.
;  (note, this is NOT a pathname... it is a filename)
MATERIAL-LIBRARY = "3ds.mli"

; Default path for all maps and backgrounds to be used by the system.
;  Note that you can define up to 9 MAP-PATHS anywhere on your
;  hard-disk.  
MAP-PATH1 = "maps"
;MAP-PATH2 = "maps2"
;MAP-PATH3 = "maps3"
;MAP-PATH4 = "maps4"
;MAP-PATH5 = "maps5"
;MAP-PATH6 = "maps6"
;MAP-PATH7 = "maps7"
;MAP-PATH8 = "maps8"
;MAP-PATH9 = "maps9"
; Note: We recommend that you keep .SXP files in your MAP-PATH1 to avoid 
;  having to change map-paths while selecting them in the materials editor.  

; This variable sets the DOS pathname for the IPAS external
;  processes. (.IXP, .PXP, .AXP, and .SXP files, although it's "legal"
;  and more convenient to store .sxp files in your primary map-path)
; Note that this is also the directory to store .BAT and .EXE files that
;  are used by Video Post.
PROCESS-PATH = "process"

; Default path for temp files
TEMP-PATH = "temp"
; Note that temp files do not include the Phar Lap virtual memory swap
;  file.  That file always "lives" in your 3DS partition unless you
;  redirect it with the CFIG3DS -swapdir command (explained in the
;  CFIG386.EXE section of the Installation Guide).

; The BACKUP-FILE switch turns on a function which creates a .BAK
;  file for any file saved within the system.  We highly recommend that
;  you leave this ON always.
; This is system selectable in the Info/System Options dialog. 
BACKUP-FILE = ON

; External programs.
; These appear in the "Programs" menu. 
; You can have up to 14 user programs, and they can be a mix
;  of .EXE, .BAT, and .PXP files.  (note that .PXP files don't
;  need the specific _W or _I reference because the program
;  automatically loads the .PXP that matches the version of 3DS.EXE.
; 3D Studio must be invoked by 3DSHELL.COM for external programs 
;  to have enough memory to run.
; Also note that programs need to have explicit path names
;  including the .exe, .pxp, or .bat extender for 3dshell.com to be able
;  to find it.  Note that COMMAND.COM must be in your 3DS 
;  directory or you must be using the COMSPEC= environment variable
;  for .bat files to run.
; Finally, note that you can pass arguments to .exe and .bat
;  files.  For example:
; USER-PROG9 = "c:\norton\lp.exe d:\froon.txt","LP Froon"
;  will execute Norton Utility's LP program, causing it
;  to print the file d:\froon.txt.
;
USER-PROG1 = "process\box.pxp","BOX"
USER-PROG2 = "process\grids.pxp","GRIDS"
USER-PROG3 = "process\waves.pxp","WAVES"
USER-PROG4 = "process\ripple.pxp","RIPPLE"
USER-PROG5 = "process\3dsurf.pxp","3D Surfaces"
USER-PROG6 = "process\fract.pxp","Fractals"
USER-PROG7 = "process\sg.pxp","Sil Garden"
USER-PROG8 = "process\place.pxp","Place"

;USER-PROG5 = "c:\aa\aa.exe","Animator"
;USER-PROG6 = "c:\calab\ca.exe","CA Lab"
;USER-PROG7 = "c:\asketch\sketch.exe","AutoSketch"
;USER-PROG8 = "c:\autocad\acad.exe","AutoCAD"

; Note, if you want to shell out to another Phar Lap application, such
;  as Animator Pro, read the CFIG386 section of the Installation and
;  Performance Guide.  It explains how to reserve 2Mb of memory so the
;  second application can run.  (note that this will permanently take memory
;  away from 3D Studio, so you may have to go out and spend $100 on
;  RAM just for this purpose.  Such is life in DOS-extender-land.

; Axis labeling -- height, width and depth axes for the Lofter and Editor
;  (change the H and D axes to get the right-hand-rule)
; SystemDefault is H-LABEL="Z" and D-LABEL="Y" (right-hand rule), but we
;  use the left-hand-rule in the documentation, as set below.
H-LABEL = "Y"
W-LABEL = "X"
D-LABEL = "Z"

; Default construction plane location for the Lofter and Editor
CONST-H = 0.0
CONST-W = 0.0
CONST-D = 0.0

; Pointing device parameters.

; This sets the default input device to either the built-in
;  MS Mouse driver or the external DGPADI driver.
;  The default is MOUSE.  For ADI use DGPADI.  For built-in Summagraphics
;  support (a la 3DSr1) use SUMMA.  
INPUT-DEVICE = MOUSE

; This command sets the COM port for the MOUSE and SUMMA pointing
;  devices (not the DGPADI device).  You can set it to ports 1 through 4.
; This is system selectable in the Info/Configure dialog.
;COM-PORT = 1

; Mouse scaling/speed factor for the entire system (default=4).
; This is system selectable in the Info/Configure dialog.
MOUSE-SPEED = 2

; Interface control parameters

; This command sets the size of the cursor "pick box".    You can set it
;  to size 5(pixels) through 16(pixels).
; This is system selectable in the Info/Configure dialog.
;PICK-BOX = 5

; The SELECTED-RESET switch will keep the SELECTED button from turning
;  off until you choose a command that doesn't make sense to use it
;  with.  It's currently set to ON, to make it the same as the default
;  shipping 3DS R1.  Experiment with turning it OFF...  you'll probably
;  like it better that way.
; This is system selectable in the Info/System Options dialog. 
SELECTED-RESET = OFF

; This makes the TrackInfo and KeyInfo buttons inside the Keyframer
;  modal.  When set to ON, they don't turn off until you select another
;  command or right-click.  Default is OFF (a la 3DSr1).
; This is system selectable in the Info/System Options dialog. 
MODAL-KFBUTTONS = OFF

; The REGION-TOGGLE switch, when OFF, will cause the Region selection tools
;  in the Shaper and Editor to be additive, instead of toggling.
;  This can be a tremendous modeling aid if you need to create
;  irregular selection sets.  This is currently set to toggle only
;  because that is the same method used by 3DSr1.   
;  It's recommended that you change it to OFF so it doesn't toggle 
;  for general use.
; This is system selectable in the Info/System Options dialog. 
REGION-TOGGLE = OFF

; Parameters affecting geometry.
; These parameters set up basic modeler functions and memory usage limits.

; This command determines whether or not closed DXF polyline entities get
;  filled with 3D faces when imported into 3D Studio.
; The default is YES.
FILL-PLINES = YES

; Maximum number of rubber spline segments in the Shaper.
; Range= 1-9999, Default= 10
SPLINE-MAX = 1000

; Maximum vertices in the Shaper.  
; (NOTE -- USES A MEMORY BUFFER)
; Range= 1-9999, Default= 500
SHAPE-MAX = 5000 

; Maximum vertices in the Lofter's deformation grid splines.
;  (NOTE -- USES A MEMORY BUFFER)
; Range= 1-9999, Default= 20
DEFORM-MAX = 500

; Maximum vertices on a path in the Lofter.
;  (NOTE -- USES A MEMORY BUFFER)
; Range= 1-9999, Default= 50
PATH-MAX = 5000

; Maximum shapes in memory in the Lofter.
;  (NOTE -- USES A MEMORY BUFFER)
; Range= 1-500, Default= 32
MOD-MAX = 80

; Number of interpolations between shape vertices in Shaper and Lofter.
; Range= 0-10, Default= 5
SHAPE-STEPS = 5

; Number of interpolations between path vertices in Lofter.
; Range= 0-10, Default= 5
PATH-STEPS = 1

; Threshold for defining coincident vertices to be used by
;  the Modify/Vertex/Weld (Selected) function in the 3D Editor
;  and the DXF-loading function.  (in units)  If objects don't
;  Unify correctly, this setting is probably too low.
; Range= 0-9999, Default= 0.0001
WELD-THRESHOLD = 0.0001

; Display, rendering and hardcopy parameters.

; Default rendering driver to use...
; Built-in output drivers: VGA320X200, Vision16 (512x486), 
;   NULL (any-resolution), VistaL (512x486), VistaH (756x486)
; ADI system rendering display driver: RDPADI or RCPADI
; Note:  Must have SET RDPADI or SET RCPADI statements
;  in your AUTOEXEC.BAT for ADI drivers to function.
; Defaults to VGA320x200.
RENDER-DISPLAY = RCPADI

;For Vista card... converts VistaH to 738x576 and VistaL to 492x576
;  when set to PAL (requires a PAL-compatible monitor to view).
VISTA-MODE = NTSC

; This command sets up the default number of frames in the Keyframer to 25
;  (for 1 second of PAL video) instead of the default 30 (for NTSC video).
;  Default is NO.
PAL-VIDEO = NO

; Built-in hardcopy devices: None
; ADI system rendering hardcopy driver: RHPADI
;  Note:  Must have SET RHPADI statement in your AUTOEXEC.BAT
;  for this to function.
; Defaults to RHPADI.
;DEFAULT-HARDCOPY = RHPADI

; Display to use for the main interface (VGA/RCPADI) if you have
;   the proper RCPADI (combined rendering/display) protected mode
;   ADI driver.  (the sample RCPVESA.EXP driver is for VESA-compatible
;   VGA cards.) 
; If this is not set, it defaults to 640x480 VGA.
MAIN-DISPLAY = RCPADI

; Display to use for Materials Editor (VGA/RCPADI) if you have
;   the proper RCPADI (combined rendering/display) protected mode
;   ADI driver. (the sample RCPVESA.EXP driver is for the VESA-compatible
;   VGA cards.)
; If this is not set, it defaults to 320x200 VGA.  Note that using
;   the RCPADI interface for the Materials Editor is basically pointless,
;   and will only slow you down without actually using the extra 
;   "real estate."
MATERIAL-DISPLAY = RCPADI

; Renderer file output parameters.

; Turn last image viewing on (it defaults to NO)
; This is system selectable in the Info/System Options dialog. 
SAVE-LAST-IMAGE = YES

; Last rendered image hold file.  Note that this variable must be the
;  complete, explicit filename including the entire path.
;  It's not just the pathname.
; Defaults to "lastrend.img" in the 3DS IMAGE-PATH directory.
LAST-IMAGE-FILE = "lastrend.img"

; Default output image type in the 3D Editor 
;  (GIF, TARGA, CTIF, MTIF, Default=TARGA)
IMAGE-TYPE = TARGA

; The TGA-DEPTH switch controls whether or not Targa (.tga) files
;  are written to disk as dithered 16-bit (15+1) or 24-bit truecolor files.
; Default is 24-bit.   Valid parameters are either 16 or 24.
; This is system selectable in the Info/System Options dialog. 
TGA-DEPTH=24

; Output image compression settings for TGA and TIF files
;  (ON/OFF, Default=ON)
IMAGE-COMPRESSION = ON

; Default output animation file type in the Keyframer 
;  (FLIC or TARGA, Default=FLIC)
ANIM-TYPE = FLIC

; This turns field-rendering on when it's YES and off when it's NO.
; Default is NO.
; This is system selectable in the Info/System Options dialog. 
RENDER-FIELDS = NO

; This sets the field-rendering order.  It controls which scanlines are
;  rendered on which field.  Different frame-buffers require different
;  field ordering.
; If you get flashing images when rendering in field-mode, then switch 
;  this to 0.   (Can be 1 or 0)
;FIELD-ORDER = 1

; This option allows you to customize the rendering resolution buttons in the
;  Render/Setup/Configure dialog for your own sizes.  The format is:
RES1 = 320,200,0.82
RES2 = 320,240,1.00
RES3 = 640,480,1.00
RES4 = 800,600,1.00
;  Where the first button is width, second is height, and third is
;  aspect ratio.  You can replace any of the four buttons by using
;  the appropriate number.  The buttons default to their settings
;  in version 1.0.
;RES2 = 256,243,1.23
; This RES2 statement is a very useful setting for previewing Targa
;  images because it's an exact 1/4 screen version of the Targa+
;  24-bit display.  If you have a Targa or Vista card, you should
;  modify it accordingly so the horizontal and vertical resolution
;  is 1/2 full size on each axis.  Very convenient!

; This command sets the default resolution and aspect ratio for the 
;  NULL output device.  (The default is 320,200.0.82)
DEFAULT-NULLRES = 320,200,0.82

; Renderer image parameters.

; This option determines whether or not the frame buffer is cleared
;  before each frame is rendered in the keyframer.  The default
;  is set to NO.  
CLEAR-BUFFER = NO

; Alpha-channel variable will tack on an extra 8 bits of transparency
;  information to your image files if it is set to YES -- creating
;  a full 32-bit file.
;  (It defaults to NO.)
; This is system selectable in the Info/System Options dialog. 
USE-ALPHA = NO

; When USE-ALPHA is set to YES, this command splits the 8-bit alpha
;  information off into a separate 8-bit .tga file that starts with
;  A_ to denote that it's alpha information only.
ALPHA-SPLIT = NO

; The shading limit for the renderer: WIRE, FLAT, GOURAUD, PHONG
;  (system default = FLAT but the shipping program default = PHONG)
SHADING-MODE = PHONG

; This command turns TURBO mode on or off.  When ON, there is no extra
;  antialiasing done on texture-mapped surfaces no matter what AA level
;  has been set.  (The texture filtering handles the job on its own and
;  does very well at it.  The major penalty is that reflections and bump
;  maps don't get filtered, so always use some reflection-blur on your
;  reflection maps.  Bump maps look fine in flics, but may have
;  to be rendered with Turbo mode OFF for true-color video animation.)
; This is system selectable in the Antialiasing Parameters dialog.
TURBO = ON

; The level of antialiasing for the renderer: NONE, LOW, MEDIUM, HIGH
;  (default = NONE)
ANTIALIASING = LOW

; This command sets the threshold for adaptive over-sampling.
;  It defaults to 20.  Lower values will slow down the renderer
;  ONLY if Turbo mode is OFF.  Higher values will speed it up if Turbo is
;  OFF, but the antialiasing won't be as complete.  
;  If you always render with Turbo mode on, you can set this to 0 and
;  get only a 2-5% performance decrease. 
; Note that when rendering with FIELD's ON, this ALWAYS ought to be set
;  at 0 (won't slow you down as long as Turbo mode is ON).
; This is system selectable in the Antialiasing Parameters dialog.
ANTIALIASING-THRESHOLD = 20

; The following three commands can be used to customize the oversampling
;  amount (2x2, 3x3, 4x4, 5x5, 6x6, 7x7, or 8x8) for antialiasing.
; This is usually COMPLETELY UNNECESSARY to touch unless you want
;  ridiculously-high quality antialiasing for still images.  It hardly
;  makes any difference at all for animations.  The only time when it
;  it may be a good idea is if you're rendering wireframes and need to
;  antialiase them extra-thoroughly.
; Note that if you see "ropy edges" when rendering scenes, you need to
;  use the VBLUR.IXP program with a vblur value of 0.6-0.9 to reduce 
;  the ropiness INSTEAD of turning up the antialiasing level above 4.
; LOW defaults to 2 (a la 3DSr1), range is 2-6
; MEDIUM defaults to 3 (a la 3DSr1), range is 3-7
; HIGH defaults to 4 (a la 3DSr1), range is 4-8  
;ANTIALIASING-LOW = 2
;ANTIALIASING-MEDIUM = 3
;ANTIALIASING-HIGH = 4

; This command turns texture filtering (Moltum-In-Parvum "mip" mapping) 
;  on or off.  When on, it uses 30% more memory for texture maps but
;  doesn't take much extra rendering time.
; Note that it should ALWAYS be on... ESPECIALLY when TURBO mode is on!
; This is system selectable in the Antialiasing Parameters dialog.
FILTER-MAPS = ON

; This command sets the blur factor for texture map filtering.
; Increase it if you see scintillating pixels in your texture 
;  maps as you move around them.  For true-color Targa or Vista video 
;  animation, you may find it necessary to push it up to between
;  1.5 and 2.0.
;  0.30 is crisp -- texture pixels will scintillate
;  1.00 is slightly blurry -- generally useful for flic animation,
;   although to keep heavily-detailed textures from scintillating, 
;   you may have to bump it up to 1.5-2.0.
; The default is 0.60 -- minimal amount for flic animation.
; This is system selectable in the Antialiasing Parameters dialog.
TEXTURE-BLUR = 0.60

; Dither 256-color output?  (default = ON)
; This is system selectable in the Info/System Options dialog. 
DITHER-256 = OFF

; These two dithering commands give the 3DS user a choice about what 
;  method to use for the 8-bit VGA dithering color compression.
; Floyd-Steinberg dithering is much more accurate, and perfect for
;  use with still images (GIF and 8-bit greyscale TIFF formats), but when
;  you use it in an animation, it has temporal artifacts that may be
;  objectionable.  The default is for the F-S method to be used for stills,
;  and not to be used for animations.
FLOYD-DITHER-STILLS = YES
FLOYD-DITHER-FLICS = NO

; Set to YES to flip the reflection map left-to-right. (default = NO)
; This is system selectable in the Info/System Options dialog. 
REFLECT-FLIP = NO

; This command controls the number of inter-object reflections in NON-FLAT
;  automatic reflection maps.  It controls the number of times each
;  auto-reflect-map object is rendered -- watch out because it can 
;  EXPONENTIALLY increase the amount of time it takes to render reflections.
; Default is 1.  Range is 1-10.
NONFLAT-AUTOREFLECT-LEVELS = 1

; This command affects the reflection of objects which intersect or
;  surround flat-mirror reflection objects.  It has no effect on other
;  automatic reflections.  When set to NO (the default), only those faces
;  in an object that completely fall below the surface of a flat-mirror
;  surface are clipped (not reflected).  When set to YES, any faces in an
;  object that even partially fall below the surface of the flat-mirror
;  surface are clipped.  (see the readme.doc for more details)
MIRROR-CLIP-ANY = NO

; Hither camera clipping plane in units from the camera.
;  (default = 1.0)
; This is system selectable in the System Options dialog.
Z-CLIP-NEAR = 1.0

; Shadow parameters.

; Spotlight/shadow parameter defaults...  These can be reset from within the
;   program from the Renderer/Configure/Shadows branch or individually
;   for each spotlight.
; Note that these parameters get saved with the geometry in the .3DS
;  mesh file.

; The size of the spotlight shadow bitmap in pixels-squared. 
; Range= 10-4096, Default= 300
SHADOW-MAP-SIZE = 300

; Width of the area (NxN) in which the shadow-map is averaged (smoothed).
; Range= 1-10, Default= 5  
SHADOW-SAMPLE-RANGE = 5

; Shadow bias (offset) from the base of the shadow to the base
;  of the object being shadowed.  (in units)
; Range= any positive floating point number.  Default is 3.0 units
;  but it's currently set to 2.0 for general use.
SHADOW-BIAS = 2.0

; This command is for switching between the INTERNAL Diaquest
;  VTR control code and the new VTPADI control code.  Contact the
;  manufacturer of your VTR controller for the appropriate VTPADI
;  driver.
; The default is DIAQUEST.  For ADI use VTPADI.
VTR-DEVICE = DIAQUEST

; This command is for turning off requests for user input from the renderer 
;  while rendering from the command line in batch mode.  If it is set
;  to YES, the renderer will proceed with rendering the scene even if
;  it can't find files it needs to correctly render the scene.
; The default is YES.
;BATCH-RENDER-PAUSE = YES
; Note that if you set this to NO, you ought to redirect the screen
;  output to a file by adding a DOS redirection command to your
;  3DS RENDER command sequence, like this:
; 3DS RENDER TEST /SP >>RENDER.LOG


;COLOR REGISTER SET COMMANDS
;---------------------------

;Palette-1 is the entire background. Include area between 
; viewports, lower right icon borders and some dialogue text.
;Default is black
;PALETTE-1 = 0,0,0

;Palette-2 is the background of the command column and
; the cameras.
;New color is blue with some teal mixed in.
PALETTE-2 = 0,114,162
;Old default was pure blue
;PALETTE-2 = 0,0,168

;Palette-3 is the tape measure.
;Default is green
;PALETTE-3 = 0,168,0

;Palette-4 is the command column text and the background color
; under the command line along the bottom.
;New color is desaturated cyan
PALETTE-4 = 129,187,202
;Old default was pure cyan
;PALETTE-4 = 0,168,168

;Palette-5 is the "selected" color, also appears in
; selector buttons in dialogues.
;Default is red
;PALETTE-5 = 168,0,0

;Palette-6 is the highlight on upper left side of icons
; and dialogue boxes.
;Default is light gray
;PALETTE-6 = 232,232,232

;Palette-7 is the dialogue box background color.
;New color is slightly-lighter dark gray
PALETTE-7 = 115,115,115
;Old default was dark gray
;PALETTE-7 = 100,100,100

;Palette-8 is shared by the viewports background color and Materials
; Editor background color.
;New color is slightly lighter gray
PALETTE-8 = 141,141,141
;Old default was gray
;PALETTE-8 = 128,128,128

;Palette-9 is the viewport outline color, dragging color
; and cursor color.
;Default is medium/light gray
;PALETTE-9 = 192,192,192

;Palette-10 is the "blue" single face highliting color
; in the 3D Editor.
;Default is light blue
;PALETTE-10 = 84,84,255

;Palette-11 is undefined.
;Default is light green
;PALETTE-11 = 84,255,84

;Palette-12 is the highlighted command which moves with 
; the mouse.
;Default is light cyan
;PALETTE-12 = 84,255,255

;Palette-13 is the red highlighted text in dialogue boxes.
;Default is light red
;PALETTE-13 = 255,84,84

;Palette-14 is the color of the interpolated ticks in
; the deformation grids in the Lofter.
;Default is orange
;PALETTE-14 = 192,148,0

;Palette-15 is the third (active) column highlighted color.
;Default is yellow
;PALETTE-15 = 255,255,0

;Palette-16 is objects color, viewport outline, viewport name, 
; icons, dialogue text, mouse.
;Default is white
;PALETTE-16 = 255,255,255

END


5. T I P S  A N D  T E C H N I Q U E S

ADVANCED 3 D Studio Tips and Techniques

Fine-Tuning the 3DS.SET File

The 3DS.SET file contains default settings and parameters.
Attempt to edit this file only if the text editor supports
reading and writing text or ASCII files. If you use a word
processing program such as WordPerfect or Microsoft Word, make
sure that you save the file as an ASCII-format file rather than
in document format. The 3DS.SET file must be in textonly format
for 3DS to read it correctly on start-up.

The following paragraphs cover some useful configuration options
contained in the 3DS.SET file. You must change the SET file to
use high resolution video adapters. Some useful defaults exist
that affect the quality of the renderings.

Display and Rendering Devices

3D Studio Version 2 requires an ADI 4.1 driver (compatible with
3DS) installed and loaded into your system. To use these drivers
to take advantage of high resolution available in the video
adapter, you must adjust the 3DS.SET file.

ADI drivers come in three models: rendering display drivers,
display drivers, and combined rendering/display drivers. 3DS can
look to DOS environment variables for the full DOS disk, path,
and file names for the EXP ADI drivers. This information is
located in the Renderer setup information section of the 3DS.SET
file. In 3DS, you can configure three sections to an ADI driver:
the rendering display, the main display, and the materials
display. With a standard installation, both the rendering and
materials displays default to VGA 320x200 resolution with 256
colors. The main display defaults to VGA 640x480 resolution at 16
colors.

3DS includes video adapter drivers other than standard VGA. ADI
drivers for VESA-compatible cards (Super VGA cards that conform
to specifications set by the Video Electronics Standards
Association) and Targa+ cards are supplied with 3DS. Your video
adapter probably comes with an AutoCAD ADI 4.1 driver. If this
solution fails, you can consult Panacea, which makes ADI drivers
for video adapters. The easiest driver to use is a combined
rendering/display driver.

Installing the driver may take a few steps. Usually, you must
first configure the driver by running the installation procedure
that comes with the driver. The installation asks questions about
the colors and resolutions you want to use in 3DS. After you
answer these questions, the driver is configured.
                                      
At times, the installer updates the AUTOEXEC.BAT file to include
the correct DOS SET statements, or the installer may just create
an ADI.BAT file for you to run to set the DOS environment
variables. Either way, be sure that you enter these SET
statements in the AUTOEXEC.BAT file to load each time you start
the computer. The variables are RDPADI for a display driver and
RCPADI for a combination driver.

After this installation, reboot the computer to make sure that
the variables are set correctly. Next, edit the 3DS.SET file and
change the corresponding RENDER-DISPLAY, MAIN-DISPLAY, and
MATERIAL-DISPLAY lines to equal the RCPADI or RDPADI variable.
After changing these settings, start 3DS. For more information
about editing the 3DS.SET to set up an external video driver, see
the Installation and Pefforrnance Guide.

The first time you start 3DS after installing new ADI drivers,
the program prompts you about resolution, colors, and number of
monitors you want to use in 3DS. The answers to these questions
are saved in the file 3DS.CFG. If you need to adjust these
settings, delete the current CFG file and restart 3DS; the
program prompts you and creates a new CFG file.

A sample VGA VESA driver loads; this driver works with most Super
VGA cards that offer VESA compatibility. If you cannot locate the
correct ADI driver, try the RCPVESA.EXP driver. If the video
adapter doesn't have VESA built into the hardware, look for a
VESA TSR (terminate-and-stay-resident driver) that makes the
video card VESA compatible. A TSR program usually is included on
the Utilities/Drivers disk that comes with video cards. You also
can check with the manufacturer or on-line services such as
CompuServe, which often provide support for the VESA standard.

Targa File Depth

Another important setting is the Targa File Depth. Even if you
aren't using a Targa+ or Targa video adapter, the TGA-DEPTH
setting may affect the quality of the output. This setting
determines whether Targa (TGA) format files are written to disk
as dithered 16 bit (15+1) or 24-bit true-color files. The default
Targa file depth is 24 bits. For best quality dithering to 8-bit
colors, use a TGA-DEPTH of 24 bits. The disadvantage is that
creating FLI animations with a high palette setting stores each
frame as a TGA file until all frames are rendered. Storing 200
24-bit Targa files on the hard disk can quickly use up disk
space.
                                      
Last Image Save

Last Image Save is useful because it saves a copy of the last
image you rendered in the 3D Editor. If this setting is set to
Yes, 3DS saves on disk a copy of the image with the name
LASTREND.IMG. Each new image you render replaces the preceding
file. You use this feature when adjusting texture maps and
repeatedly referring to the last rendered image. You can call up
the LASTREND.IMG file with one command column option sequence:
Renderer, View, and Last.

The disadvantage of this option is that all images you render to
disk are saved under the file name and the LASTREND.IMG file
name. Small, lowresolution images present no problem. If you are
rendering high-color or high-resolution images such as 2000x2000
24-bit color images, however, the disk space the last image file
uses becomes huge, even many megabytes. You may want to turn off
this option for these projects.

Dither Settings

Size and speed requirements determine the settings for the Dither
option. If you are outputting to an FLI file to be played back on
VGA PCs, note that dithered FLI animations can take up to twice
as much storage space and may play back very slowly. Turning off
dithering helps with storage and playback speed. You pay the
price with banding, a sharp transition from one color to another
due to a color shortage in the VGA palette. If you have disk
space, always use dithering when creating renderings with 256
colors.

Null Render Size

Another setting in the 3DS.SET file is Null Render Size. If you
are creating images for printed output, you probably render these
images at high resolution, lK by lK or higher. Because few ADI            
drivers support resolutions this high, you probably render these
final images to a file. 3DS can never render an image at a
resolution higher than the current video driver. You need to
choose Renderer, Setup, and then Configure to specify the NULL
setting as the display device. You can preset the default
resolution and aspect ratio of the NULL display device in the
3DS.SET file. This change saves time when you are ready to render
the final image because you don't have to enter the resolution
and aspect ratio every time you render to the NULL device.
                                      
2D Shaper

The 2D Shaper is the foundation by which most complex 3D objects
are built. Learning how to control deranged spline curves can
assist you in creating base shapes quickly with a minimal number
of vertices. Using the GRID and SNAP functions enables you to
create precise curves and polygons, as well as to position text
correctly.

For the 2D Shaper, the next sections cover how to create large
amounts of text, how to trace existing images, and a few
additional tips and shortcuts.

Creating Large Amounts of Text

If you need to create more text than can fit in the Enter Text
dialog box, press Ctrl when you use the Snap option. The Ctrl key
guarantees the correct aspect ratio of the text so that you don't
squeeze or stretch the text. The Snap option enables you to
create multiple text lines of the same height.

If you are creating multiple lines of text and need to use upper-
and lowercase, make sure that each line contains at least one
capital letter. Although you use Ctrl for the correct aspect
ratio, the size of the letters currently being placed determines
the size of the selection box. If you place the words in and
three using four grid snaps for the height, for example, the size
of the two words differs. The word in is slightly larger because
no ascender, such as the letter h in the word three, is contained
in the word in. To solve this problem with multiple lines of
text, type a capital letter at the beginning of the text and then
delete the capital before lofting the text.

Tracing Bit Maps

You can trace bit maps in the 2D Shaper by first loading them as
the background image in the 3D Editor or the Keyframer. From the
Views menu in the 2D Shaper, select See Backgrnd. If the
background image isn't clear, adjust the image with the Adj
Backgrnd command. Adj Backgrnd gives you control over the
background images contrast.

Shortcuts for 2D Shaper

When adjusting splines in the 2D Shaper, holding down Ctrl
enables you to reposition the vertex being adjusted. Holding down
Alt while adjusting a vertex enables you to adjust the tangent
point between the vectors that create the spline.
                     
T I P 

Always try to create as few vertices as possible. Fewer
vertices make the resulting 3Dmesh file smaller and cut the time
needed to render the file.

When loading PostScript fonts, use the Shape and then the Check
command before lofting. PostScript fonts may have overlapping
polygons that are rejected by the 3D Lofter. By first checking in
the 2D Shaper, you can edit the polygons before attempting to
loft them.

3D Lofter

The 3D Lofter is a powerful tool that creates complicated objects
that twist, teeter, bend, and otherwise not conform to common
geometric shapes. Among the most efficient uses of the 3D Lofter
is to create morphable objects that have the same number of
vertices and work well for morphing from one to the other.
Morphing is where one 3D object transforms itself into another 3D
object--a kind of metamorphosis.

Using Multipie Shapes

Each level along the path can have a different shape. This
capability enables you to loft an object like a sword that starts
out with a square cross-section with four sides (where the blade
attaches to the handle, then changes to a diamond-shaped
cross-section for the length of the blade, which finally closes
to a sharp tip). When lofting through different shapes, follow
these important rules:

 Each shape must be on a different level.

 All the shapes must have the same number of vertices or
  polygons.

 The first vertices of all the shapes must be aligned.

 Turn off Optimization if the shapes have different combinations
  of linear and curved segments.

Always try to create the object with the fewest possible steps
and vertices. The exception occurs when you need a large number
of faces for morphing or for some PXP routine that requires many
vertices, such as RIPPLE.PXP.
 

Loft Deformations

Deforming while lofting shapes allows you to create extremely
complex 3D objects. 3DS enables you to deform a shape by beveling
the edges, teetering it back and forth as it travels along the
lofting path, and a number of other deformations, which can even
be performed in combination with each other. One deformation is
the twist, which twists an object around while following a path.
You can use the twist deform to create a realistic-looking
braided rope. The Scale deform can create objects that scale up
or down in a smooth, curved sweep, such as a vase that gently
curves scales larger and smaller along its length.

The Fit lofts enable you to take the front, side, and top views
of an object and force or fit a 3Dmesh into these dimensions.
When creating Fit lofts, visualize the path object as the front
view. Think of the Fit X window as the top view and the Fit Y
window as the side view. If the objects don't appear in the
correct rotation when you get them from the 2D Shaper, return to
the 2D Shaper, rotate the polygon, and return to the 3D Lofter
again with the correct orientation. Choose Deform, Fit, and Gen
Path to create a path with steps that coincide with the vertices.
The generated path also helps you visualize how the Fit shapes
work together. The Deform and Preview command series enables you
to verify before attempting to create the Fit loft.

Shortcuts for 3D Lofter

Use the automatic mapping option in the Object Lofting Controls
dialog box to apply automatic mapping to lofted objects, which
saves time later in aligning mapping coordinates with unusually
shaped objects in the 3D editor.

When lofting an object that you don't plan to render from the
back, turn off the Cap End option. This setting leaves the
backside of the object open and saves many faces, especially with
lofted text. If you loft different shapes along the same path,
use the shortcut keys, PgUp and PgDn, to select the current level
on the path. PgUp moves up the path and PgDn moves down the path.

Materials Editor

The Materials Editor is the key to photorealistic renderings. No
matter how beautiful the model is or how sharp the shadows are,
if the surface materials don't look good, the scene isn't
convincing. The following discussion provides suggestions for
deriving the best results in the Materials Editor. 

Take time to experiment, and notice the subtle differences
between ambient and diffuse colors, and between the Transparency
and Transparency Fall-off values. When you see interesting
surfaces in daily life, mentally break down the components. Think
of the roughness of texture, and the kind of bump map that may
simulate the texture. Consider the reflectivity of the material
and whether it actually reflects other objects. Most important,
look at the colors. What color is the material in shadow or in
direct light? What color is the specular highlight? Daily
observation and skill with the Materials Editor are the keys to
photorealism. For more information on surface material
attributes, see the 3DS Reference Manual.

In the following paragraphs, you explore the use of material
libraries and also techniques for using external processes,
creating realistic reflections, and effects for using animated
maps.

Managing Material Libraries

Because one scene may have hundreds of different surface
materials, you need a way to keep these materials organized.
Material Libraries enable you to organize materials. Consider
creating separate libraries for each project; you may reuse
textures rarely. For architectural renderings, however, you may
find libraries established by material classifications--marbles,
wood grains, exterior finishes, and so on-worthwhile. Just
remember to use the library capabilities: creating new libraries;
merging existing libraries; and loading, saving, and deleting
libraries.

A default library, 3DS.MLI, is loaded each time you start 3DS.
However, you can start a new library by just choosing New from
the Library menu. If you need to copy just one or a few materials
from one library to another, follow these steps:

1. Pick the Material menu and choose Get Material from the source
library. The material loads into one of the five material boxes.

2. With the Library menu and the Load Library command, load the
new library.

3. After the new library loads, pick the Material menu; choose
Put Material to save the material to the new library.

Using this technique, you can transfer from one to five materials
at a time. To transfer more materials, use the Library menu to
access the Merge Library command.                  

3D Studio loads the library 3DS.MLI, when you start 3DS. To cut
the time 3DS takes to load and save the default library (3DS.MLI)
under a different name remove all the materials from the 3DS.MLI
library.

Using SXP Files Effectively

Solid-pattem external process files, or SXP files, bring a new
level of realism to 3DS. SXP files are procedurally generated
textures. Think of these files as synthetic textures, in that 3DS
generates the pattern based on a mathematical formula in the SXP
file. Although the advantages may not be apparent, SXP files
enable you to move as close as you want to an object without
losing detail. If you zoom too close to a surface with a
bit-mapped texture of wood grain, for example, you run the risk
of pixelation, where you see the individual pixels of the texture
map. 3DS Version 2.0 uses a technique to blur the pixels, but the
effect still is noticeable at close range.

Synthetic or SXP surfaces, however, are generated when needed and
at the needed resolution. If a table has an SXP simulating wood
grain, for example, you can move the camera from across the room
down to the table, skimming just centimeters above the surface.
Rather than seeing large globular pixels, you see crisp,
well-defined bars of wood grain, as if you actually were two
centimeters above a wood table.

Because SXP surfaces don't require mapping coordinates, these
surfaces are good candidates for difficult-to-map objects such as
helixes. The Wood SXP surface enables you to define the color,
spacing, and waviness of the wood grain. The Vary SXP enables you
to fade one color to another over time. Using Vary as an animated
opacity map can make an object appear or disappear over a number
of frames. When the Dents SXP is used as a bump map, realistic
fractal dents--realistic looking chips, indentations, and
imperfections--can be made to appear on any object.

Animated Maps

Animated maps have been available with 3DS since Version 1.0, but
few people use them to full capacity. Using rotoscoped video as a
self-illuminated, animated texture map produces a convincing
simulation of a live television screen. To simulate reflected
light on objects around the TV, you can use the same animated map
as a reflection map set to 50 percent or less with a
high-reflection blur setting.

T I P                                              

Animated texture maps also are effective as reflections for shiny
moving objects where you don't need fine detail. Another use of
animated texture and opacity maps is in simulating effects such
as fire, rain, snow, and explosions. You can create
realistic-looking people by rotoscoping faces as they talk and
look around. Then apply this animation as an animated texture map
to a generic 3D Face. The details of the rotoscoped face cover
small imperfections in the 3D Face.

Reflections--Flat and Curved Surfaces

When creating objects with flat surface reflections, assign the
mapping coordinates to the face level. If you assign mapping
coordinates at the object level, you cannot guarantee which face
has the reflection map. You can apply the same flat reflection to
multiple faces on the same object as long as the faces are
co-planar.

3DS Version 2.0 also provides a setting to adjust the amount of
reflection blur. This setting is located in the Materials Editor,
just above the shininess slider--the button is marked with the
letters BL. Most reflective objects look good with at least a
little blur. Turbo mode requires some reflection blur because
this mode only anti-aliases edges, and texture blur works only
with texture maps. Setting reflection blur to 3 or 4 usually is
sufficient to solve this problem. All mesh files created in 3DS
Version 1.0 and loaded in Version 2.0 have reflection maps
defaulted to a reflection blur setting of 0; you need to adjust
the reflection blur settings slightly.

Shortcuts for the Materials Editor

Make good use of existing textures as base settings for new
surfaces. When you need to make an adjustment to an existing
surface and keep a copy of the original, click the sample box and
drag the sample to the next available box. The Materials Editor
copies all materials settings, after which you can make new
adjustments or experiment.

If your display driver has the capability, set the resolution for
the Materials Editor to 640x480, which gives you a better preview
of the surface materials and doesn't take much more time to
render the samples. If you are fine-tuning the colors of a
surface material that uses a bump map, you can save time by
temporarily turning off the bump map button until you get the
colors adjusted. After you have the sample rendering the way you
want, you can turn the Bump button back on to preview the final
texture.

Another time-saver is pressing Esc during a sample rendering. As
the sample sphere or cube renders, you see the surface material
in monochrome. If you are readjusting the bump map, you may be
able to press Esc after setting the bump map correctly. Another 
case where this step saves time is in creating many surface 
materials with the same settings, but different maps. When creating 
flags, for example, all the settings are the same for each flag material.
Rather than doing a preview rendering of each flag, just choose
the new texture map and put the material into the library. The
preview render isn't required before putting material in the
library.

3D Editor

Everything--the models and surface textures--comes together in
the 3D Editor. Objects from the 2D Shaper and 3D Lofter are
combined with lights and cameras to provide illumination and
perspective. You can run procedural modeling external processes
(PXPs) to modify existing 3Dmesh geometry or create new geometry
from formulas.

In the following paragraphs, you explore techniques for getting
realistic and fast results from the 3D Editor. You also find
suggestions for effective use of Boolean operations, mapping
coordinates, external processes (PXPs), and turbo mode.

Creating Complex Objects from Geometric Primitives

Most geometric primitives that you can create in the 3D Editor
are efficient in terms of faces. At times, you even can create
complex models with these geometric primitives. To create a
rocket, for example, you can use a hemisphere and simply stretch
the bottom vertices down to make the body of the rocket. To
create the fins, you can make a box and then use Taper and Skew
to refine the shape.

Another trick to speed rendering is to delete all faces that
aren't seen in the rendering. If you have a table with a 3D table
top and are sure that the camera will never show the underside,
you can delete all the faces that enclose the bottom. This
deletion speeds rendering times and saves memory.

Object Manipulation Tips

When building complex 3D objects in the 3D Editor, always try to
keep the object oriented to the main viewports. If the finished
object needs to rotate at an unusual angle, create all the
geometry, apply surface materials, and run sample renderings to
verify correctness before
tilting the object off the standard axis. To work on an
object-oriented sample at an angle, use the User viewport along
with the Display, User View, Align command series to align the
user view to any face.

3DS determines the outside of objects by using face normals. By
rendering only the outside of solid objects, 3DS saves time in
rendering. If an object needs to be open to reveal its insides,
such as an open box, you can assign the object a two-sided
material. Two-sided materials render both sides of the face,
regardless of which side the normal is facing.

Sometimes you run into problems with face normals getting flipped
away from the camera, which can be caused by importing complex
geometry from DXF files. You usually can correct this problem
with the Surface, Normals, Unify command series. If a particular
face remains problematic, select Display, Geometry, and then Full
to show all faces. You then should see and fix the offending face
by choosing Surface, Normals, and then Face Flip. If this method
fails, use two-sided materials to render the face, regardless of
the normals.

Successful Boolean Operations

Given the inherent complexity of Boolean operations, exercise
care when creating them. Generally, the more faces, the better.
Both objects being modified must have many faces. If you
originally lofted the objects from the 3D Lofter, make sure that
you turn on Tween and turn off Optimization. Also, try to have an
equal number of faces on the objects you need to change. If the
Boolean operation doesn't work with Weld turned on, turn off
Weld. If the operation still fails, move the objects slightly or
re-create them with different faces.

Also remember that Boolean operations greatly increase the number
of faces on an object. If you perform multiple Boolean operations
on the same object, the object may become so complex that Boolean
operations cease to work on it. Always minimize the number of
Boolean operations; creative use of the 3D Lofter can help with
unusual shapes.

Mapping Techniques

3DS offers three basic kinds of mapping--Planar, Spherical, and
Cylindrical--but you can apply any kind to any object. A good
example is a sky dome, an open hemisphere turned upside-down atop
the 3D scene. By mapping a self-illuminating sky image to the
inside of this upsidedown hemisphere, you can get a realistic sky
that pans as the camera moves.
 

Because a hemisphere is basically spherical, you may think that
the spherical mapping type may work best; but the planar map
works equally well if mapped flat on top of the dome. Flattening
the dome slightly by choosing Modify, Object, and then 2D Scale
helps eliminate any streaking near the edges of the dome, where
the map may run perpendicular to the faces.

Spherical mapping tends to pinch the texture map at the poles, so
you must place the icon carefully to minimize this problem.
Spherical mapping can be used on the dome if the mapping icon is
rotated 90 degrees in any direction so that the pole of the
sphere faces the side of the dome rather than the top. You also
need to move the mapping icon slightly below the bottom of the
dome so that the pinched poles don't map on the edges of the
dome.

One last method is to use a reflection map with the sky image as
the reflection map. This technique eliminates the need for
mapping coordinates, but doesn't give as much control over the
placement of the clouds on the dome. Reflection maps also take
longer to render than texture maps.

Using PXP (Procedural Modeling External Processes)

Procedural modeling external processes (PXP) is an IPAS routine
that modifies the 3D geometry of meshes and returns the modified
meshes to the 3D Editor. Version 2.0 includes sample PXP
routines: RIPPLES.PXP radiates sine wave ripples through a 3D
object; WAVES.PXP creates linear waves through an object; and
BOX.PXP generates a bounding box object for a selected object.

Some PXP routines, such as Ripples and Waves, can copy, modify,
and return existing geometry to the 3D Editor. You can use this
modified geometry for morphing and animating the effect.
Structural test PXP, for example, can be written to test
geometry. The results then can be animated for easy evaluation.

How To Tune Anti-Aliasing-Turbo Mode/Filter Maps

To speed rendering, tune the anti-alias settings based on the
model you are rendering. First, look at the anti-aliasing level
buttons in the Render dialog box. These buttons control the
oversampling amount used for anti-aliasing. The defaults are 
stored in the 3DS.SET file: Low, 2(2x2); Medium, 3(3x3); and 
High, 4(4x4). For high anti-aliasing, 3DS renders 16 pixels 
for every one displayed on the screen.

Next, consider the anti-aliasing parameters: Anti-alias
threshold, Texture blur, Filter Maps, and Turbo. The threshold
determines how far apart the colors of two adjoining pixels need
to be before anti-aliasing occurs. For best results, set the
threshold at 0; this setting provides a minimum performance
increase with turbo mode turned on. Texture blur and Filter Maps
determine the way texture maps are anti-aliased. Turning on
Filter Maps smoothly scales down pixels in textures that are
smaller than the screen pixels. Texture maps with pixels larger
than the screen pixels get smoothed out automatically in 3DS.
Texture blur affects the blurring of textures when Filter Maps is
turned on.

Turbo mode only anti-aliases the edges of objects. When used with
filter maps, which take care of the texture maps, good quality
antialiasing can be achieved at a high speed. Reflection maps can
be softened with the Reflection Blur setting in the Materials
Editor, and the shadows can be softened by increasing the Sample
Range. With this much control over anti-aliasing, you rarely need
to anti-alias every pixel in an image.

Shortcuts for 3D Editor

You can use the User view to quickly render snapshots for
verifying scene details, such as texture placement. You also can
use the Zoom and Pan icons, as well as the Display, User View,
Align command series, to position the User view accurately and
quickly over any face.

Always be aware of the construction plane position. If you know
that you will create many objects at a certain Y axis height,
choose Display, Const, and Place to set the construction plane at
the correct height in the front or side viewports. This way,
objects created in the top viewport are set at the correct height
automatically.

Keyboard shortcuts take less time than choosing commands from the
menu; for example, to select objects, hold down Ctrl when the
cursor is a pickbox. The X key turns the center axis icon on and
off, and the Shift key clones vertices, faces, elements, or
objects.

Before using the Create, Object, Attach command series, switch to
box mode by pressing the B key. Box mode helps you see objects as
they are attached to each other because the two bounding boxes
merge together into one bounding box as two objects are merged
into one.

Being able to select vertices with windows and then manipulate
the vertices gives you immense control over 3D objects. You can
move, rotate, scale. skew, mirror, bend, taper, and delete
sections of objects.

Keyframer

The Keyframer, the choreographer of 3DS, controls the movement
and animation of 3D scenes. Because of Keyframer's complexity,
use the Hold button often. Hierarchical linking, looping,
repeating, and instances all are wonderfully advanced tools, but
this complexity can surprise you.

The following sections cover shortcuts for object and camera
movement, and the new Video Post feature.

Object Movement

Often you need to move an object within a set number of frames,
say 40 to 50. If you simply go to frame 50 and move the object,
3DS calculates the movement from the object's last movement. The
last movement may even be at frame 0. Here, you must use the
Tracks Info dialog box and copy the keys from frame 0 to frame
40, which results in the movement being limited between frames 40
and 50.

A better way to accomplish the same task is to just click,
without actually moving, the object at the beginning frame (40),
which creates a keyframe for the object's current position. Then,
when you go to frame 50 and make the object move, 3DS only
calculates the movement between the last keyframe (now frame 40)
and the new one, frame 50.

If you need to select a large number of objects or select an
object that isn't currently visible, use the Hit key by pressing
h. The Hit key displays a list of objects and their linkage in
the Keyframer.

Dummy objects are good all-around workhorses in the Keyframer.
These objects serve many purposes such as camera control, object
movements, and exploding objects. Because dummy objects don't
render, feel free to make them large so that they are easy to
pick and manipulate.

Morphing enables you to create an animation in which one shape
changes into another shape. The secret of successful morphing is
creating objects that are nearly identical in vertices or faces
and yet different in overall appearance. To meet these criteria,
clone existing objects in the 3D Editor (using the Shift key with
any Object, Modify command), or use the 3D Lofter.
 

When using the 3D Lofter to create morphable objects, be sure
that both objects have the same number of steps in the path. The
objects also must have the same number of vertices in the
original polygon to be lofted.

Copying Tracks

With 3DS Version 2.0, you can copy track information from one
object to another. Begin by selecting Object, Tracks, and then
Copy. Click the source and then the destination. The Copy Tracks
dialog box appears with options for tracks to copy. The important
choices are Relative and Absolute. Relative copies the movement,
position, and scale from the source object to the destination
object's relative movement, position, and scale. Choosing
Absolute makes the destination object's movement, position, and
scale the same as the source.

You can use the Relative option to save time in setting up
keyframes. If you animate a human model to walk, for example, you
can copy the model to another position and copy the tracks with
the Relative option turned on. This option copies the same
keyframe animation to the new model.

Setting Up and Editing Camera Paths

To keep things simple in complicated animations where multiple
objects move, get the scripting of the objects correct before
animating the camera. This method enables you to see each
object's movement without the distraction of camera panning. The
same suggestion applies to animated lights.

The Paths, Show-Hide command series is useful in editing
keyframes. When an object's path is displayed, choose Track or
Key Info and click any white key square to edit the key without
having to move to the frame.

Detailed Camera Control

The Key Info dialog settings give you access to the Tension,
Continuity, and Bias (TCB) values from keyframes. When adjusting
these settings, remember that objects may be inheriting movements
from other objects. A key also can be affected by preceding or
following keys. If the preceding key has a high Ease From value,
for example, the object approaches the next frame at a high
speed. Turning on the Path, ShowHide option for animated objects
reveals how the current TCB values affect the movement.

Use the field of view creatively for special effects. Use a
wide-angle lens to exaggerate perspective and give objects a
large, looming look. Use a telescopic lens from a distance to
downplay the perspective, making objects look small.

Use caution when setting up a walk-through of a proposed
building. If the camera approaches the building as if in an
airplane, the relative travel speed may be surprisingly fast. If
you want to fly to a building from a distance in a few seconds,
for example, and then tour cubicles inside, allow for more frames
to slow the camera as it moves into and through the office. If
you use the same number of frames as the exterior fly-by, you may
cruise the cubicles at 100 mph.

Using the Video Post

Consider the Video Post feature as comparable to a complete video
post-production system. You can combine multiple images,
animations, and image processing external processes (IXP). You
can combine the images with alpha channel compositing for
high-quality compositing. Transition effects also are built in.

Although Animator Pro can create smooth transitions, you can
program Video Post into a queue with the same transition and
compositing information and then save the queue. This feature
enables you to run 3DS overnight unattended to build an entire
animated sequence. If a section of the animation needs to be
redone, simply call up the saved queue, and 3DS recomposites the
entire sequence with the new animation.

Before working with Video Post, make sure that the files you want
to composite are in a map path directory. Create a directory to
store the queues; the 3DS installation program doesn't create
this directory.

The Frames Grid enables you to specify ranges of frames with
range bars. These ranges are where the associated image from the
queue is rendered to the animation sequence. The range bars are
somewhat similar to time lines in project schedulers. When moving
range bars, pay attention to the status line at the top of the
screen; it provides useful information.

After you finish creating the Video Post queue, make sure that
you exit with the OK button. If you press Esc or click Cancel,
the queue is erased. To save the Video Post queue with the 3DS
animated scene, use the File menu and select Save Project. To
save the queue separately-without the 3DS mesh file--use the Save
button in the Video Post dialog box. Make sure that you select a
directory for the VP (Video Post) files with the file selector
dialog box; the default directory is \3DS.

T I P

When creating animation for VGA displays with the STARS.IXP
process, choose a Brightest Star Value of less than 180. Brighter
values cause the stars to be big and chunky.  

Photorealism

Some 3D animators and artists attain photorealism and others do
not. Success in the quest depends on knowledge, good tools, and
the ability to look at real objects objectively, mentally
breaking the object's illuminations, coloring, surface qualities,
and more.

Nonsaturated Colors in Real Life

One of the easiest ways to make renderings more realistic is to
use few, if any, full saturated colors. In real life, few objects
have a saturated color. Most real-life colors have a large amount
of gray; even the brightest colors show tints of gray.

Lighting likewise shouldn't be harsh. Turn down the saturation
before clicking OK to create a light in 3DS.

Use lighting to create advanced effects like radiosity. Radiosity
is a technique for simulating the way light reflects off some
objects and illuminates others. The formula for calculating this
light reflection is derived from a formula to calculate heat
dispersion. To create the reflected light effect, identify the
areas of the scene that are most illuminated; then place
low-intensity omni lights at these locations. Be careful not to
make these lights too bright, or the effect is lost.

Composing in 3D

You can enhance realism with good composition; this rule applies
to both still images and animations. A simple rule is to divide
the screen into quarters and place the most important object in
one of these quarters.

With animation, use smooth camera movements. If you are flying in
a chrome logo to spin to a stop, use the Key Info dialog box and
adjust the Ease To values to make the logo ease to a stop.
Carefully watch movies and motion pictures to see what camera
effects are used. Pay
special attention to how transitional scenes are made. If you
need to move from an exterior shot of a building to an interior,
for example, don't just fly through a window or wall--be more
creative. Pan the camera over the building and land it on a tree
in the front landscape. Then, as a transition, make a smooth fade
to the same tree viewed from inside the building. From there,
continue the camera's journey through the interior.

You can simulate depth of field effects by rendering the
background at a lower resolution than the final image or
animation. After the background is rendered to a file, use that
file as a texture map on a flat wall behind the foreground. Then
render the scene again at normal resolution. Because the
background, which now is a texture map, has a lower resolution
than the screen, the background is blurred. You can create the
same effect by adjusting the Texture Blur setting.

Reflections

Don't forget to use reflections. 3DS Version 2.0 has automatic
reflection maps; use many of them. Chrome and other polished
metals aren't the only materials that need reflection maps.
Realism can be simulated by adding reflections to other shiny
surfaces such as wood, plastic, leather, and clear glass.

Environment maps work well for reflections, but these maps often
need to be customized. A good example is found in the movie
Terminator 2. Industrial Light and Magic sometimes used seven or
more images to create just the right environment maps for the
chrome man from the future. You can create custom reflection maps
by rendering views from the object that needs the map and
creating a standard reflection map or animated reflection map.

Water

WAVES.PXP and RIPPLE.PXP can create surfaces that mimic water.
Even with a large body of water, you may be able to keep the
geometry simple by creating as few as five morphing copies of the
original object. Another technique is to use the NOISE.SXP shader
as a bump map. With the right colors and viewed from a distance,
this technique gives a convincing illusion of a large body of
water.

Half the illusion of realistic water is to match the true color
of water as closely as possible. The ocean at sunset, for
example, usually has an ambient color of black, a specular color
of white, and a diffuse color of

an unsaturated deep blue. Be sure to position the camera to pick
up the specular highlights from the water object. Remember to
keep the saturation low for all kinds of water.

Shadows

Soft shadows add to the realistic effect. One way to create them
is to reduce the hot spot of the spotlight to a narrow beam,
around one percent. This setting causes the shadows to have a
soft, natural look. If the shadow is chunky or coarse, increase
the map size. If the shadow contains streaks or moire patterns,
increase the bias value. If the increased bias value detaches the
shadow, increase the map size to accommodate the bias value. To
make the shadow sharper at the edges, reduce the sample range. To
make the shadow edge softer, increase the sample range.

Applications and Markets for 3D Studio Output

With the advent of high performance, low-cost 3D modeling and
rendering software, new markets are opening daily. Some of these
markets are detailed in the following list:
 Archaeology
 Cartography
 Corporate video
 Entertainment--motion picture, television, multimedia
 Financial
 Forensics
 Paleontology
 Scientific visualization via IPAS routines

CompuServe

The value of the Autodesk Software Forum (ASOFT) on CompuServe
cannot be emphasized enough. Here you can download the latest
utilities from Autodesk. Textures, 3Dmesh files, rendered
animations, and stills are uploaded daily from some of the best 
3DS users all over the world. You also can obtain expert advice on 
any 3DS-related problem. Even Gary Yost of The Yost Group (creators
of 3D Studio and Animator Pro) frequently answers problem
questions. You also find a wealth of professional contacts who
offer services such as batch rendering, frame editing to
videotape, outputting to slides, and more.

Seeing up to 50 3DS related messages posted in the Forum every
day isn't unusual. Topics range from 3DS tips and tricks to
hardware compatibility and performance discussions. Tools to make
CompuServe more user-friendly (such as the CompuServe Information
Manager) or to make CompuServe less expensive (like TAPCIS) also
are available.

Summary

Just as AutoCAD first brought power to the masses by providing a
high technology application to low-end computer users, 3D Studio
has done the same for animation. Photorealistic rendering and
animation are now within the realm of hundreds of thousands of
people who use 386and 486-based PCs.

3D Studio Version 2 reaches new levels of creativity with 3D
graphics, with extensive features that satisfy even the most
advanced user. This chapter serves as an appetizer for your 3D
rendering and animation needs.

The following chapter begins a special section of this book,
covering the ways that you can customize the AutoCAD
configuration with network capability and special AutoCAD
settings.


6. B O O L E A N  O P E R A T I O N S  H I N T S

How-To for 3D Studio r2

Take the 'Boo!' Out of Boolean Ops

by Frank Zuccaro, AeroData

Autodesk 3D Studio release 2's
Editor module provides
three types of Boolean operations:
union, subtraction, and intersection.
A Boolean operation allows the user
to modify an object's geometry using a
second object. For instance, a cylindri-
cally-shaped object can be used to
"drill" a hole in a cube-shaped object.
The shape of the cylinder is removed
from the cube shape using the Subtract
feature. Two objects may be joined us-
ing the Union function, which removes
all faces and vertices within the over-
lapping portions of combined objects,
and leaves the rest as a combined ob-
ject. Conversely, the Intersection opera-
tion retains only the overlapping geom-
etry shared by two objects.

General Rules for Boolean Operations

Boolean operations can sometimes
fail. The user can maximize his or her
success rate by adhering to the follow-
ing guidelines.

1. Objects created with a "surface of
revolution" SurfRev path in the 3D
Lofter are prone to failure (see Item A).

2. The general rule is that the face on
one object should cut through no more
than one face on the second object (see
Items B and C).

3. Objects with many small faces are
good candidates for Boolean operations
(see Item C).

4. Objects with long triangular faces,
parallel faces, or thin faces are likely to
fail (see Item B).

When Good Booleans Go Bad

When a failure occurs, click on the
Continue button and the original two
objects will be restored.

1. Reposition the object(s) so they
overlap differently. (Hint: Adjust Dis-
play/Geometry/All Lines so you can
view the geometry completely.)

2. Increase the number of faces.

3. Alter your geometry if it's com-
posed of long faces, parallel faces, or
thin faces.

4. Select all vertices of each object
separately. Weld the group, and then
unify the object. Do this to both objects.
Try the Boolean operation again.

5. Select one or both objects. Use the
Save Selected function from the pull-
down menu to save the objects as a
.DXF file. Use a unique name to avoid
confusion. Delete the original object(s).

Use the Merge command from the
pull-down menu and reload the
object(s). A dialog box will appear,
where the default settings often work
fine, but you may choose to turn
Smoothing off.

6. Try flipping the faces of one of the
objects (see Item D).

Item A--SurfRev

Boolean operations do not work
properly on some objects created us-
ing a surface-of-revolution path in the
3D Lofter. Specifically, when you loft
a SurfRev object using Shapes/Align/
Left or Right, the central core of the
object consists of many coexistent ver-
tices and edges. Avoid this area when
overlapping two objects for a Boolean
operation. For more information, see
Path/Surf/Rev in Chapter 6, 'The 3D
Lofter," in your reference manual.

Item B--Facial Concerns

If a Boolean operation fails, either
move one of the objects or increase the
number of faces in one or both objects.
Boolean operations work best when
performed on objects with many small
faces. This also increases the speed of
the operation and improves the ap-
pearance of the rendered intersection
edge when smoothing is used. Objects
with long triangular faces and parallel
faces between the two objects are more
likely to cause failures during a Bool-
ean process.

Item C--Complexity

Boolean operations increase the
number of faces in an object. Repeat-
ing Boolean operations on the same
object compounds this effect exponen-
tially. To see this, turn on Display/
Geometry/AII Lines before perform-
ing a Boolean operation (see page 7-
199, reference manual).

Item D--Flip Out...or In

The direction of face normals on the
objects affects the result of the Boolean
operation because 3D Studio uses
normals to determine which portion of
geometry is overlapping. For example,
if you first flip the normals of only one
object, then perform the Subtraction
operation, the result is the same as the
Intersection operation. (See "Surface/
Normals," page 7-126 of the reference
manual.)

If all else fails...
Explain your problem in the "asoft"
forum on CompuServe, in the "3D
Render" forum on PCGnet, or send it
to your 3D Studio dealer (or 3D
Artist) to forward to Autodesk Tech-
nical Support. (Autodesk's support is
onlv publicly accessible via the CIS
"asoft" forum. Questions are often
answered there by members of the
Yost Group, which authors 3D Studio
and Animator Pro.)

Additional Notes

You can increase the number of faces
in an object manually or by using the
Create/Element-Face-Object/Tesselate
commands. Or you can create the ob
ject over again with more complexity,
perhaps lofting it with more vertices
using the Tween, Optimization, and
Steps settings.

It's not unusual to hear about fail-
ures in Autodesk's support system
where users can receive help only
through dealers (or in the asoft forum
on CompuServe. This article, adapted
from an AeroData Computer Services 
handout. shows how well :Autodesks
policy can work when it does work
You can see why this Michigan 
reseller, which also anchors a user
group, ranks high in 3D Studio

A Light Tip

We told AeroData's Frank Zuccaro
that we had a reader request for help
in better lighting and asked if he had
something related. His response was
no, but he had two recommendations
he likes to give 1) Look for photogra-
phy texts which show classical exer- .
cises in lighting. This information
which also appears in beginner paint
ing books, applies directly to 3D art. =
Such books can be found in any 
library or bookstore. 2) create a stage
file with lighting the way you like it
for most scenes, including omni and 
spot lights, and ambient light setting.
(As an aid familiar to photography
students, you could make an egg-like
object--flat-white with some surface 
roughness and stand it on end in the
middle of this scene. When test ren-
dering, it will help you to observe re-
sults of adjustments to your lighting.)
Save this file and use it to start new
scenes.

No Bit Player         

The word "Boolean" derives from
the English mathematician 
Boole. His work on "set" theory IS
used directly in combining sets of 3D
vertices. More importantly, this is the
guy who invented "on and off." A
"bit" is either on or off (one or zero)
and every last thing you do on a 
computer, at the most basic level, is done
in bits--eight to the byte. (One byte
can represent a decimal number range
from one to 256. Notice you don't hear
too much about 257-color images.)
Boole created mathematical ways to
describe logic. This developed into
Boolean algebra, which became the
foundation for digital computer pro
gramming. He presented his first
paper on this new math at around age
32, despite little formal education be-
yond elementary school. He was well
recognized in his own time as a
genius, though there wasn't immedia-
te use for some of his discoveries. The
first digital computer wouldn't appear
for another 99 years...in 1946.



7. 3 D  S T U D I O  R 2  R E V I E W

Autodesk Improves a Good Thing

3D STUDIO RELEASE 2

by Tim Forcade

Users waiting for the new release of
3D Studio will soon learn that good
things come to those who wait. Auto-
desk's Multimedia Division has included
automatic cubic environment maps,
faster rendering, improved shadows,
Boolean operations, and a CD-ROM con-
taining more than 500Mb of 3D objects
and materials.

USER INTERFACE/GLOBAL FEATURES

3D Studio's well-designed and inter-
active user interface remains consistent,
stable, and predictable. One of the most
significant improvements with Release 2
is the addition of an expanded set of key-
board alternatives to its mouse-based
command set. Wlth Release 2 comes the
welcomed capability to call up a dialog
box that lists all scene objects for selec-
tion from either the 3D editor or Key-
framer. This is a timesaver when working
with complex scene elements. In addi-
tion, the See Background and Adjust
Background commands work together to
provide a contrast- and scale-optimized
proxy view of the specified TGA, TIF, GIF,
FLC, FLI, or CEL file for tracing or align-
ing scene objects.

The new system options dialog box al-
lows the user to override parameters such
as pixel depth (16 or 24), the optional use
of the alpha channel, or the weld thresh-
old. These values may also be set in the
3DS.SET file, but the advantage of the
options dialog box is that the user doesn't
have to shell out of the session to make
these changes.

To speed processes like tracing back-
ground images, automatic scrolling of
the viewport can now be turned off,
which eliminates unnecessary redrawing
of displayed geometry. Also, Release 2 lets
you save and restore the zoom and pan
views, viewport settings, and the view
angle for the user viewports.
To the Current Status dialog box of
Release 1.0, which displays the status of
geometry, lights, and cameras in use in
the current session, Release 2 adds a list-
ing of memory use including available
memory, memory used, size of disk swap
file, and the number of times the swap file
was accessed during the current session.

Access to this information is extremely
useful for determining how your system
resources are being used, particularlv
with a RAM-hungry, high-performance
program like 3D Studio. For instance,
although the program can work with only
4Mb of RAM, most users will soon dis-
cover that rendering performance is
being seriously compromised, because
when the program runs out of RAM, it is
forced to swap to a virtual memory file on
the hard drive. The bottom line is that if
the status dialog indicates that disk swap-
ping is taking place, add more RAM.

Other nice touches to the Release 2
interface include the addition of HLS
(hue, luminance, saturation) sliders to
boxes with color controls, such as spot-
light and omni light color; the display of
a safe frame area in camera views; and
the ability to select specific objects when
merging scene files.

MOVING FROM THE 2D SHAPER TO THE 3D LOFTER

3D Studio's 2D Shaper module allows
the user to create and edit two-dimen-
sional shapes and Bezier spline-based
polygons. The 2D Shaper has been ex-
panded to accept up to 9,999 vertices, up
from only 999 in the previous version.
This provides a more useful range for
importing or creating intricate shapes.
Adobe Illustrator files may now be di-
rectly loaded as well as PostScript Type 1
fonts (PFB). The Shaper comes with 20
Bezier fonts.

The new Freehand Draw command
lets you create polygonal lines by simply
drawing freehand. A smoothness setting
varies the number of vertices, which are
Inserted over the length of the current
freehand polygon.

The 3D Display command is now
available for displaying 3D mesh objects
from the 3D editor This is essential for
quickly creating and scaling new geome-
try that aligns with existing geometry. An-
other timesaving function is the Modify
Vertex Weld and Weld Selected functions,
which merges adjacent vertices to pro-
duce closed polygons suitable for lofting
The 3D Lofter is a cross-sectional
modeler that lets the user extrude in up
to three dimensions simultaneously.
This can be done by using either single
or multiple shapes along an arbitrary
path in the z axis. The Release 2 Lofter
now adds the ability to apply mapping
coordinates to lofted objects along both
their length and perimeter Mapping
coordinates are used to specify the po-
sition and scale of texture, bump, or
opacity maps. This feature is particular-
ly useful for quickly providing mapping
coordinates that align textures with
curvilinear geometry (such as the pat-
tern of a snake skin).
Furthermore, the 3D Lofter now im-
ports both DXF and FLM (filmroll) files
for use as extrusion paths and also pro-
vides optimization to help moderate the
complexity of lofted 3D mesh objects.

EXPLORING THE 3D EDITOR

The 3D Editor is 3D Studio's startup
module. It is here that the various mem-
bers of a scene--models, cameras, and
lights--are created, positioned, and pre-
pared for either rendering as still images
or exporting to the Keyframer as subjects
for animation.

The editor now provides Boolean op-
erations: union, subtraction, and inter-
section, which are remarkably fast and
robust. Occasionally, a Boolean opera-
tion will fail but can usually be corrected
either by increasing the number of faces
or moving one or both objects. Another
technique is to first select all vertices for
each object separately, weld the group,
and then unify the object and try the
Boolean again.

The Surface Mapping command set
can now display and acquire another
object's mapping icon or coordinates as
well as define the mapping region by
simply dragging a box around the desired
area. Surface Smoothing now adds auto-
smoothing based on the edge angles of
selected geometry. This is a useful global
method of managing which faces will
appear faceted or smoothed.

Any object in the 3D Editor may now
be exploded into multiple elements based
on the angles of its edges. This results in a
series of separate objects that maybe eas-
ily texture mapped or animated. Also, a
selection set of objects may be copied ei-
ther as a single object or as multiple objects.
Shadows have been vastly improved
in Release 2. The granular look along the
shadow edges has been replaced by a
natural-looking transition that can be
varied from soft to sharp. Control of
shadow variables is also much improved
over Release 1.0. Local shadow control is
provided for each spotlight with control
of map size, bias, and sample range, or
the user may use the global settings from
the 3DS.SET file. Map size, which may be
varied to accommodate differences in
rendering output resolution and light
distance, controls the coarseness of each
shadow. The bias value controls shadow
offset, and the sample range control s the
softness of the shadow's edge.

Finally, DXF translation has been im-
proved in Release 2. Along with the abili-
ties to derive objects from layer, color, or
entity come automatic vertex-welding/
unifying and autosmoothing by a user-
specified angle. One bit of DXF obscurity:
If you plan to create geometry by origi-
nating it in 3D Studio for subsequent ex-
port into AutoCAD (for dimensioning, for
instance), be sure to avoid using either
hyphens or apostrophes in any 3D Studio
object or element name as this causes the
translation into AutoCAD to fail.

SETTING UP THE KEYFRAMER

The Keyframer is 3D Studio's keyframe
animation module. The Release 2 Key-
framer now loads incrementally num-
bered bitmap files, which maybe used for
producing animated image maps for ob-
jects or backgrounds. You may also view
high-resolution FLICs from within 3D
Studio at any resolution supported by
your VESA-compatible display adaptor
A new path branch to the Keyframer
allows you to graphically adjust motion
paths for objects, cameras, or lights.
Many of these functions are duplicated in
the Key Info dialog box but nonetheless
offer a welcome alternative for adjusting
path attributes. The New Path Follow
command provides a means of aligning
an object to a path and provides the op-
tion of banking up to a specific angle.
There is a new Video Post dialog box
that duplicates many of the processes
used in a video postproduction suite. Al-
most a complete program in its own
right, the Video Post dialog box dramati-
cally expands 3D Studio's usefulness for
single rendered images as well as animat-
ed sequences. Included here are func-
tions such as image compositing, cuts,
fades, and the application of traveling
mattes. In addition, you can access the
alpha channel of any image and call ex-
ternal programs for special effects such
as rotoscoping and starfield generation.
Other useful additions to the Keyframer
include a set speed command for real-
time accuracy of previews, size buttons that
reduce the output size of FLC files, a user-
specified number of frames to use for both
heads and tails, and assignment of an
image file list (IFL) or incrementally nurn-
bered files for disk to videotape recorder.

USING THE MATERIALS EDITOR/RENDERER

The materials editor is used for editing
the supplied materials library and for
creating unique materials. 3D Studio
Release 2 adds two methods of reflection
mapping: cubic reflection maps and
automatic reflection maps. Cubic reflec-
tion maps or environment maps are pro-
duced by first creating six bitmap views
taken from the center of a selected object
in the scene. These views are then used
as a reflection map. The process also pro-
duces an ASCII CUB file that may be edit-
ed to assign different images to the se-
lected object's reflection map. Automatic
reflection mapping consists of a combi-
nation of spherical and cubic maps that
are generated automatically. Although
automatic maps are slower, they are
more accurate because a unique map is
produced for each frame.

This release also adds several proce-
dural textures (SXP or Solid-pattern eX-
ternal Process) including marble, wood,
and dents. The external process or IPAS
system is a programming interface that
allows users and developers to further
customize 3D Studio using the C pro-
gramming language. SXPs are 3D textures
that produce seamless pattern variations
even in models that have been cut or sec-
tioned using Boolean functions. Each
procedural map is provided with variable
parameters such as color, size, or offset,
which are accessible through the materi-
als editor interface. In addition to the pat-
terns, 3D Studio comes with IXPs (Image-
processing eXternal Processes), PXPs
(Procedural-modeling eXternal Process-
es) and AXPs (Animated stand-in eXter-
nal Processes) to complete the Release 2
IPAS complement.

Other improvements to the materials
editor include the ability to use either a
sphere or cube for sample materials
renders, a transparency falloff option
that provides control over an object's
surface, and reflection blur, which varies
the sharpness of an object's reflections.
These additions to the materials editor,
most of which are mutually inclusive,
expand the variation of surfaces that can
be created to a mind-boggling scale.
Although the addition of RenderMan
would certainly be welcomed, it is not as
sorely missed as in the previous version.
3D Studio Release 2 is faster, much
faster. Direct comparisons of render
times between Release 1.0 and 2 in some
cases revealed an increase of more than
seven times in render speed. For in-
stance, rendering the RACECAR.3DS
demo file on a MEGA 486-33 with 32Mb
of RAM produced times of 52:04 and
7:19, for releases 1.0 and 2, respectively.
This frame was rendered to a NULL
buffer at 756 x 486 with a Phong shading
limit, antialiasing set to high, shadows
and textures on, and all other variables in
default mode. With antialiasing set to a
more realistic medium range, Release 1.0
and 2 yielded render times of 29:25 and
5:57. The program provides a number of
additional variables to further trim ren-
der times including the antialias thresh-
old and texture blur parameters.
Other practical features include com-
mand line rendering (which does not re-
quire the hardware key), field rendering,
and a rescale/tile option for background
images included in all render dialogs.

PLAYING WITH THE CD-ROM

As if all the new features of Release
2 weren't enough, how about 600Mb of
bitmap images suitable for all sorts of
maps, models, FLIs, FLCs, and assorted
example images? There are more than
60 marbles, more than 50 woods, rocks,
clouds, tile and wallpaper patterns,
trees, and some 200Mb in TGA, GIF,
and CEL formats. Most of the maps are
24- and 32-bit images with resolutions
from less than 200 lines to more than
1,200 lines.

There are more than 220Mb of FLC
and FLI animations, 60Mb of 3DS mod-
els, 60Mb of sample rendered images.
Sifting through all these files is fascinat-
ing, informative, and incredibly time-
consuming. With its models ranging
from the Grand Canyon, the city of
Houston, and St. Paul's Cathedral to a
sextant, a stapler, and a cow, the "World
Builder Toolkit" CD-ROM forms a useful
learning and production tool.

The creators of 3D Studio have taken
what was an exceptional 3D program
and dramatically improved and expand-
ed it. With its improved rendering speed
and quality, extended list of functions,
the CD-ROM disk, and a set of com-
pletely updated documentation includ-
ing new tutorials, this program main-
tains both its sophistication and its
accessibility. With its second release, 3D
Studio continues to be a robust and
thoroughly useful program to anyone
needing to visualize the real world
through virtual space.


8. A N I M A T I N G   W I T H   3 D   S T U D I O

Animating with 3D Studio

When you use Autodesk 3D Studio, you can create and animate objects in a vari-
ety of ways. Lights and cameras can be moved and altered for unusual visual
effects. For example, a red spotlight that moves across shiny silver text is 
simple to create with 3D Studio and the result is a look you frequently see 
on television.

As you have seen, 3D Studio has numerous animation options. You may wonder,
at first, how you can use these fancy tools in a presentation (particularly 
if the subject matter is dry or uninteresting). In this chapter, you'll 
find ideas on how to create exciting animation for your presentations by 
combining object movement with camera and lighting changes.

Before reading this chapter, you should read Chapter 7, in which you find the
basics of 3D Studio covered in detail. In particular, Chapter 7 explains how 
to import an AutoCAD model into 3D Studio (you see this principle in use 
within the exercises included later in this chapter).

Before proceeding with the exercises in this chapter, you should have 3D 
Studio properly installed and ready to use on your machine. If you do not 
have Autodesk 3D Studio, read this chapter to flnd out what can be done 
with your software.

3D Studio Concepts

Before looking at the animation options of 3D Studio, a few concepts need to be
understood. These concepts deal specifically with animation procedures within
3D Studio. (Concepts of animation differ significantly from those of still 
imaging.)

Keyframing

Animation with 3D Studio is set up in the Keyframer module. The keyframer takes
its name from traditional animation technology, in which each frame is drawn
and painted by hand. On a large project, such as a Walt Disney feature film, a
senior animator draws the extremes of two parts of a scene, and a junior 
animator uses these drawings as guides to draw the in-between frames. Thus, 
the senior animator's drawings are called keyframes, and the junior animator's
drawings are called in-betweens or tweens. When you use the 3D Studio
keyframer, you set each object's position in the keyframes and the keyframer 
figures the tweens.

The keyframer has many additional features as well. You can use it to smooth 
out the movement between keyframes, to preview your animation before rendering
and for many other tasks.

Animation Output

To view animation created with 3D Studio, you have two options: either to save
the animation as an Animator FLI file to view on your computer screen or to
record each frame individually to videotape and view on a videotape deck. Your
choice of animation output is determined by the hardware you have.

FLI Files

FLI files--or Flics--are disk files that have the extension FLI. FLI files 
consist of a series of frames that, when played, create the illusion of 
animation. The phrase "to play a flic" means to view the animation on your 
computer screen. This is referred to as real-time playing.

Any 3D Studio hardware configuration can create FLI files. These animations 
can be played by Autodesk Animator, Autodesk Animator Pro, or FLI players. 
You can also create FLI files directly from 3D Studio by highlighting the 
Flic button when you set up your rendering configuration or you can record 
the animation to video-FLl files can have any resolution, but they always 
contain a maximum of 256 colors. You can create FLI files with a resolution 
that is higher than your computer is capable of displaying; VGA computers can 
play an animation with a resolution of 320x200 or lower. There are practical 
uses for creating FLI files that have too high a resolution to be viewed on 
your computer. This technique is explored later in this chapter.

In order to look like animation, FLI files must be played back at speeds that 
fool the eye. FLI players allow you to adjust the playback speed but the 
computer just refreshes the screen for each frame. If the changes from one 
frame to the next are too great, the FLI will not play fast enough. When you 
create an FLI file that will eventually be played on a computer screen, do 
not create radical changes over too few frames.

A playback speed of fifteen frames per second is the minimum required for 
convincing animation. FLI players are capable of this speed, and higher 
speeds, if the changes between frames are smooth and not too extreme. 
Higher playback speeds, when possible, result in a smoother look.

One of the advantages of FLI files is that they can be enhanced with text or 
additional graphics in Animator or Animator Pro. The results can be very 
useful for presentation purposes. You can also record a FLI file to videotape 
in real time. For more information on this topic, see Chapter 11.

Single Frame Output

The best way to create smooth animation, if you have appropriate equipment, is
to record each frame to videotape individually. This method is referred to as
single-frame recording. Because videotape always plays at 30 frames per second,
regardless of the change from one frame to the next, the look of the final 
animation is more predictable than the results you get with FLI files. (For 
more information on single-frame recording, see Chapter 11.)

Another advantage of single-frame output is that you can create frames that have
more than 256 colors each. Each frame can be created and saved in Targa 
format, which allows thousands of colors in each frame. This format is named 
after a set of graphics cards, the Targa series, manufactured by Truevision. 
Targa format, which has the extension TGA, has become a standard in the PC 
graphics industry--many non-Truevision graphics cards display Targa files.

You can also render and save frames in two other formats, Color Tiff (CTIFF) or
Monochrome Tiff (MTIFF). Color Tiff files are useful for color printing and 
monochrome Tiff files can be imported into many desktop-publishing systems 
for inclusion in newsletters or reports.

You do not need a Targa or Targa-compatible graphics card to create TGA files;
3D Studio generates and saves these files to disk even if you are equipped only
with VGA. In order to record Targa files to videotape, however, you must be able
to display the images, so a Targa-compatible graphics card is required. The same
is true for CTIFF and MTIFF images.

Targa, CTIFF, and MTIFF files cannot be played in real time on your computer
screen. For images that have more than 256 colors, computer technology has not
advanced to the point in which the screen can be refreshed fast enough to look
like animation. The only way you can view animation made with these file formats
is to record each frame individually to videotape and play back the tape. (If 
you don't have a graphics card that displays your images, or you don't 
have single frame equipment, service bureaus are available to record Targa, 
CTIFF, or MTIFF files to videotape for you.) Another option is to render a 
Targa file for each frame, and then use the AutoFlix utility to compile the 
frames to FLI format.

If you have VGA-capable, single-frame equipment, and you are dissatisfied with
the look of your real-time FLI flles, you can record each frame of the FLI flle to
videotape individually. You are still limited to 256 colors but the movement 
will be smoother. (For more information on recording animation to videotape, 
see Chapter 11.)

If you are considering upgrading from real-time VGA to single-frame capability,
see Appendix D for information on the hardware required.

Choosing Animation Output

If you only have VGA equipment and you want to be able to see your animation.
you must first render to a 320x200 FLI file and play it in real time on your 
computer screen. If you have only VGA equipment, but you want the color 
richness and clarity of Targa images, you can render your animation frames to 
TGA flles and bring them to a service bureau for output to tape.

If you have Super VGA hardware, you can create FLI files at higher resolutions
than 320x200. SuperVGA equipment varies in resolution capability, but you can
create 256-color FLI flles at resolutions of at least 640x480. These FLI 
files can be played in real time with Animator Pro or a third-party 
high-resolution FLI player.

If you have Targa capability, you can test your animation first by rendering 
to a FLI flle and playing it on the screen. When you are satisfied that the 
animation is correctly set up, you can render to TGA flles and output to 
videotape. Rendering to FLI flles takes much less time than rendering to 
TGA flles. Table 9.1 displays output options based on your hardware.

Animation Exercises

The exercises in this chapter will all be rendered to 320x200 FLI output. 
This will minimize rendering time and show you results more quickly. 
Also, these FLI files can be used for the exercises in Chapter 10.

Preparing for Animation

Before moving on with the exercises in this chapter, you should load and 
configure 3D Studio for your computer according to the instructions that 
come with the software. You should also read Chapter 7 of this book and 
perform tutorials 1, 2, and 9 of the Autodesk 3D Studio Tutorials book 
that comes with the package. You can also do the rest of the tutorials 
if you like.

If you use AutoCAD to create your models, you should know how to import
AutoCAD models into 3D Studio as described in Chapter 7. lf you are not an
AutoCAD user or you have not created the AutoCAD drawings from the previous
chapters in this book, you can use the 3D Studio files on the included disk  
or use models of your own.

Animating Objects

The most obvious use of 3D Studio for animation is the movement of objects over
a series of frames. This section discusses the concepts involved in moving 
objects and changing their shapes.

Object Movement

An easy way to animate a model is to have the various parts fly apart and 
then fly back in and compose themselves into the original shape. For the 
following exercise, the caster model is used to illustrate this method 
of animation.

To start the exercise, load 3D Studio. If you have created the caster model 
described in Chapter 3, you made a filmroll of this drawing. If you have not 
already imported the caster filmroll file into 3D Studio, do so now. Use 
Modify/Object/Rename to name the objects as follows:

lf you have not created the caster model, you can use the file CASTER.3DS from
the enclosed disk or a model of your own. The caster appears in the four
viewports, as shown in figure 9.1.

Before you continue, switch to box mode by choosing Display/Geometry/Box.
Now it's time to move the objects, starting with the caster support (you can 
do the other objects later). Note that the support is represented by the large 
box around the entire model.

Place a camera in the scene, as in figure 9.2. Change the user viewport into 
the camera viewport. Your viewports should look like those shown figure 9.2.

ln addition, place an omni light at the front of the model. Go to the keyframer 
by choosing Keyframer from the File menu or by pressing F4.

Each part of the model is shown in white. The keyframer has automatically 
assigned 30 frames to this animation as the default. (This default will be 
used to create the animation over 30 frames.)

The current object positions for frame 0 are shown on the screen. Frame 0 shows
the position of each object as it was in the 3D editor. Advance to frame 1 by 
clicking on the single right arrow at the bottom right of the screen. Note that 
the outline of the objects changes to black.

Frame 0 is considered a keyframe, it positions the objects for the start of 
the animation. Frames 1 through 30 currently contain no keyframes. Keyframes 
display the objects in white; non-keyfrframes are displayed in black.

Because you work with 30 frames in this exercise, the caster parts will fly 
apart until frame 15, and then fly back in to their original positions on 
frame 30.

The first order of business is to set frame 15 as a keyframe. To get to frame 
15, click on the current frame button, which shows a 1. Click on OK to accept 
the default of 15. Your screen changes to show the object positions at frame 15. 
The objects in each viewport will appear exactly the same as frame 1, but the 
current frame box will show the number 15.

Choose Object/Move, and click on the support bracket in the front viewport.
Press Tab until the arrow points left and right only. Move the support bracket
over to the right, just past the rest of the caster, as in figure 9.3. Click 
again to set the support bracket.

To see the animation as it stands, choose Preview/Make and click on Preview.
After a few minutes, the preview will play. Note that the support bracket moves
over to the right, but it doesn't come back.

Even though a preview shows no colors or materials, it does use shades of gray 
to roughly show the effect of lights on the model in the animation. If you 
don't place any lights in the scene, the preview shows all objects as a flat 
shade of dark gray, making it difficult to distinguish the objects.

To make the bracket move back into position on frame 30, click on the down 
arrow at the lower right of the screen to display frame 30. Note that the 
support is shown in the same position as frame 15.

To take a straight-ahead approach to this animation, you might try to move the
bracket back into its original position. This method is unworkable, however, and
you'll probably miss the mark (the animation won't look right). A more exact 
approach is to copy the keyframe information about the support from frame 0 to
frame 30. This is accomplished by using the Track Info panel.

Using Track Info

The Track Info panel gives you a graphical representation of your keyframe 
information. To use it, click on Track Info and then click on the support 
in any viewport. The Track Info panel then appears.

This panel might seem unfamiliar to you, because Track Info is in chart form
rather than in 3D form. The numbers across the top of the chart refer to frame
numbers and each keyframe has two or more black dots underneath it. Currently,
frames 0 and 15 are the only columns that have black dots.

Note that the object name Support is shown at the upper right of the panel. The
chart shows keyframe information for this object only. If you move the slider 
bar under the object name, you will see the names of other objects appear. The 
track info for these objects will not show a dot under frame 15; only the 
support has a dot at frame 15, because it is the only object that has been 
changed from its original orientation on frame 0 (one exception is the 
object World, which shows a compilation of all the dots for each object).

Move the slider bar back so the object name Support shows at the upper right of
the panel. The dots under frame 0 ail look alike, but each one represents a 
specific position, rotation, or size for each object. The dots cannot tell 
you how much an object has moved or rotated; they only tell you that the 
frame is a keyframe and that a particular aspect of the object has changed 
on that frame. Under frame 15, note that there is a dot across from Position 
but not across from Rotate or Scale. These aspects of the object Support were 
not changed on this keyframe.

The position, rotation, and scale of an object can all be changed 
independently. These aspects of object movement are called Tracks. The last 
track, Morph, is also independent of the other three tracks. Morphing is 
covered in detail later in this chapter.

The label All Tracks contains a dot whenever any of the tracks for that frame 
have a dot. This makes it easy for you to glance across a large number of 
frames and spot the keyframes. Note the black line under the number 30, 
indicating that frame 30 is the last frame. The columns beyond 30 are used 
only when you set your total frames higher than 30.

You can copy the keyframe information from frame 0 to frame 30 by clicking on
the Copy button at the bottom of the panel. Click on the dot next to Position 
under frame 0, and move it as far to the right as it will go--under frame 30. 
Click to set the dot on the Position track for frame 30. A dot also appears 
under All Tracks to show that frame 30 is now a keyframe.

Because you copied the dot from the Position track of frame 0, the position of 
the object at frame 30 will be the same as the position of frame 0. If you now 
look at the dot under frame 30, it is impossible to tell where it was copied 
from, so you must exercise care in copying and moving keyframe dots.

You are now finished with the Track Info panel. Click on OK to set the new
keyframe information. The support is back in its original position on frame 30
and is outlined in white to show that this is a keyframe for the support. Do not 
be overly concerned if you are still not an expert on Track Info. As you do the 
exercises in this chapter, you will become more familiar with this 
powerful tool.

Choose Preview/Make to get a look at your new animation. If you approve, save
the model and the new keyframe information by choosing Save from the File
menu. Click on the + button to save the file as CASTER01.3DS. Your keyframe
information is saved along with the mesh file.

Smoothing the Movement

In the preview you just created, you may have noticed that the support starts
moving abruptly, slides gently to the fight, and then stops suddenly when it
moves back into the original position. When the support moves and reaches the
rightmost position, 3D Studio smoothes out the movement (it slows down, stops
and then starts up slowly and gains speed until it reaches the ofiginal 
position). The start and the end of the movement has not been smoothed.
On keyframes in the middle of an animation, 3D Studio automatically smoothes
the animation for you. It does not do this for the first and last keyframes 
of an animation, however. This explains why movement into and out of keyframe 
15 is smooth but is abrupt at frames 0 and 30.

This problem can be solved by adjusting the key information. First, go to frame 30
by clicking on the down arrow at the bottom fight of the screen. To access key
info about this keyframe, select the Key Info button, then click on the support.
The Key Info panel appears.

The object name Support appears at the top right of the panel. Several options
are displayed on the panel as well, including the Ease To/From sliders. The 
Ease To slider affects the speed of the object as it moves into the keyframe 
position. Move this slider to 5. When the object moves into the position on 
keyframe 30, it will slow down before it stops.

Now that the end of the animation looks smooth, you can also adjust the 
beginning motion. At the left of the Key Info panel, the Key# is shown. 
The support has three keys, numbered 1, 2, and 3, at frames 0, 15, and 30. 
Because you are working with frame 30, the current key number is 3.

To move to the first key, click twice on the minus sign (-) on the Key# slider. 
This moves you to key 1, which is frame 0. Because you want the movement coming
from frame 0 to be smoother, slide the Ease From value to 5. When the support
moves away from its original position, it will start slowly and then speed up to
normal velocity. The support's movement is smooth as it moves into and out of
each position.

You are now finished changing the Key Info panel. Click on OK to exit the panel,
and create a new preview to test your new animation. Save this animation file as
CASTERM.3DS to use in the exercise in the Linking section later in this chapter.

The same process used on the support bracket can be applied to the other objects
in the scene as well. Move to frame 15 and move all the objects away from their
ofiginal positions. Next, copy keyframe 0 to frame 30 in the Track Info panel 
for each object. All the objects fly apart until frame 15, and then return to 
position in frame 30. Use the Key Info panel to smooth the movement coming out 
of keyframe 0 and going into keyframe 30. These processes speed up the animation 
sequence and improve appearances.

Changing an Object's Shape

The keyframer has several built-in tools to change the shape of an object. Note
that although 2D Scale affects the object's size in only one or two directions, 3D
Scale changes the object's size in general. The way in which the 2D Scale com-
mand affects the object depends on the viewport you are in when you use this
command; the command is performed relative to the view of the object in the
viewport you choose. The same is true for the Taper and Rotate commands.

Because these three commands can be performed in any viewport, however, you
can 2D scale, rotate, or taper an object anywhere in three-dimensional space by
choosing the appropriate viewport for the operation. To practice, try squashing
the spring in the suspension system model. Make sure you move the struts in
synch with the spring.

Animating the Camera

Changing the camera's FOV and position, rather than moving the objects, creates
a simple yet striking imagery. For example, suppose you set the active camera's
lens length at 135 mm on the first frame and 22 mm on the 30th frame and you
pull the camera back from the objects over these frames.

The resulting animation has a dramatic effect that Hitchcock would have envied
and doesn't require the movement of any objects. This camera trick is easy to do
and is illustrated in the following exercise.

To start the exercise, load 3D Studio. This exercise uses the chapel model. If 
you have created this model in previous chapters, you should have created a 
filmroll file. If you have not already done so, import the chapel model into 
3D Studio and rename the objects as follows:

If you have not created the chapel model, you can use the file CHAPEL.3DS from
the AutoCAD 3D Design and Presentation Disk.

Hide the object Base to make the model easier to work with in this exercise. 
Select Display/Geometry/Box to show the model as a set of boxes. This helps to
speed up the redraw process as you do the exercise. You can choose Display/
Geometry/Full Detail at any time to redraw the model in full detail. You can 
also move between box mode and full detail mode as often as necessary. You can 
also press Alt+B to switch between these modes.

First, a 135 mm camera is placed in the chapel model. This camera will gradually
change to a 22 mm lens length over 30 frames. To do this, place a 135 mm camera
facing the front of the chapel with the Cameras/Create command, and name
the camera Zoomview. Don't be concerned if the placement is not precise.

To view the camera view from the 135 mm lens, choose Viewports from the Views
menu to access the viewports panel. Change the view in the lower right viewport
to the camera view. Because you have only one camera in your scene, 3D Studio
will not ask you to pick a camera for the viewport. Use Cameras/Move to position
the camera so you have the view shown in figure 9.6. You also can place an omni
light as shown in the figure.

In order to change the lens length to 22 mm on the 30th frame, you must move to
the keyframer by choosing keyframer from the Program menu or by pressing F4.
The scene you set up in the 3D editor is frame 0. You can now set the camera
lens length for frame 30.

If your lower right viewport is labeled No View Defined rather than Zoomview,
change this viewport to show the camera view. Go to frame 30 by clicking on the
down arrow at the extreme lower right of the screen. Ascertain that you are on
frame 30 by checking the current frame number at the lower right of the screen.

On frame 30, choose Cameras/Adjust and click on the camera. (The lens length
for this exercise is 22 mm, but this lens is not available as a stock lens.) 
You can create a custom 22 mm lens by clicking on the Lens button and entering 
22. Then click on the Calculate button to display the corresponding FOV. (You 
must always click on Calculate when you enter the lens length manually. 
Otherwise, 3D Studio negates your entry by using the FOV setting to calculate 
and reset the lens length when you exit the panel.)

The current view in the camera viewport is not very interesting and you must
move the camera to see the entire model. To move the camera and target, choose
Cameras/Move. Use figure 9.7 as a guide to get the camera and target in the 
right position.

You are now ready to make a preview of the animation. Click on Preview/Make,
then click in the camera viewport. On the Make Preview panel, click on OK to
start the preview. When the preview is created, it automatically plays back 
in a loop. The camera appears to pull back from the altar, and the FOV changes 
as the camera moves.

You may be surprised that the animation doesn't look so strange, in spite of the
odd lens settings you used. That's the whole idea; the audience should not 
notice that the lens length is changing, but should be drawn in by the movement.

Feature films frequently manipulate lens lengths to create illusions. If you 
are a movie buff, you recognize this camera move from Bonfire of the 
Vanities (in the scene in which Tom Hanks is accosted by two punks). The 
next time you're at the movies, see if you can spot where the camera lens 
length is manipulated.

Animating Lights

In an animated presentation, you often need an opening sequence. If you have
been pressed for time in preparing your presentations, perhaps you have simply
thrown together a crude title sequence. A simple yet striking opener can be made
with shiny, metallic 3D text and moving spotlights. Try this exercise with your
company name.

In the 2D shaper, locate the font that most resembles your company's logo. A
chart showing each font appears in the back of the 3D Studio Installation and
Performance Guide.

Enter your company logo in the 2D shaper, loft it in the lofter, and move to the
3D editor. Assign the material BLUE METALIC to the obiect.

Set up a camera that gives you a slightly angled view of the logo. Place one 
omni light at the front center of the scene. Set the RGB settings at 50,50,50. 
Place a spotlight at each end, but have the spotlights point off into space, as 
in figure 9.8.

Set one spotlight's color at RGB values 50,10,10 to make it a dark red; set the
other spotlight at 50,50,10 to make it dark yellow. Move to the keyframer and 
set the number of frames at 180. On frame 90, move the left spotlight target 
across to the other side of the logo. Click on the right spotlight target (do 
not move it), and then click again. This keeps it from moving during the first 
90 frames.

On frame 180, move the right spotlight target to the left side of the logo. In 
this animation, the colored spotlights move across the shiny letters and make 
subtle effects. The combination of shiny objects and colored lights has a rich 
look that is easy to achieve.

This simple animation can be an attractive opener for your presentations. As a
further exercise, you can change the spotlight colors at the keyframes for a
gradual change in color over the frames.

Animating Materials

FLI animation files can be used as texture, bump, opacity, or reflection maps. 
FLI flles (or flics) consist of a series of frames that comprise an animated 
sequence. They can be any resolution, but are limited to 256 colors. They can 
be created with 3D Studio, Animator, Animator Pro, or Chaos.

To apply a FLI file as a map, first create the FLI flle. This file must reside 
in the \3DS\MAPS directory or the \3DS\IMAGES directory. In the materials 
editor, choose the FLI flle as the map file, as you would for a GIF, CEL, or TGA 
file. Apply the material in the 3D editor as you would any other material, using 
the map icon for texture, bump, and opacity maps.

When you render your animation, 3D Studio uses the FLI flle as the map. In the
keyframer, frame 0 is rendered with the first frame of the flic, frame 1 uses 
the second frame of the flic, and so forth. Even if you begin rendering at a 
frame after frame 0, the keyframer figures out which frame of the flic to use, 
based on this method. When the flic has fewer frames than the 3D Studio 
animation, the keyframer continues to render and use the flic over again, 
starting with the first frame of the flic.

As an opener for your presentation, try the following procedure. Render all your
animation to FLI flles, in addition to your usual rendering output. Then map the
animation onto a series of spheres that fly across the screen. This animation 
will make a splendid "overture" to your presentation.

Morphing

In 3D Studio, you can have one object gradually change into another one over a
series of frames. This function is called morphing, from the word metamorphosis.

In the keyframer, one of the Object options is Morph. When you choose this 
option and click on an object, you are given a list of objects in the scene 
that have the same number of vertices as the object you chose. When you choose 
another object, the keyframer gradually turns the flrst object into the second 
object over a series of frames.

In order to morph one object into another, both objects must have exactly the
same number of vertices. Rather than trying to build two morphable objects, it 
is much easier to start with one object in the 3D editor, use Create/Object/Copy 
to make a copy of it, and modify the second object. In this way, both objects 
always have the same number of vertices, and you can morph from the first object 
to the second.

You can also create two morphable objects by lofting two shapes with the same
number of vertices, provided that they both use the same lofting parameters. A
circle, for example, has four vertices in the 2D shaper, just as a rectangle 
does. A cylinder made from a lofted circle can be morphed into a rectangular 
bar that has also been lofted.

The keyframer performs a morph by moving the vertices in the first object to the
position of corresponding vertices in the second object. This concept is helpful to
keep in mind when you modify objects to be morphed.

You may wonder how morphing can be used for presentations. For example, you
can cause changes to an object in your animation that cannot be addressed with
the Scale or Squash options in the keyframer. You can also make an object "grow"
from a flat surface, as illustrated in the following exercise.

To perform this morphing animation, you need the chapel model you created in
earlier chapters. If you did not create the chapel model, you can use the flle
CHAPEL.3DS on the enclosed disk or create a model of your own.

Change the lower fight viewport to the user view. For the purpose of this 
exercise hide the Roof object with the Display/Hide/By Name option. This enables 
you to see the interior of the chapel during the animation.

Place an omni light near the center of the chapel for basic illumination. Leave 
the light at the default color setting. Your viewports should look like figure 
9.9.

Choose Display/Geometry/Box to display the model in box mode. To make this
morph easier, attach all the objects together to form one large object. First,
choose Create/Object/Attach and click on two objects. Repeat this step until
there is only one object left. When you have attached all the objects, there 
will be just one large box remaining on the screen. You can rename the remaining 
object FlatChapel.

In order to morph the object, a copy must be created (one copy is morphed to the
other). Create a copy of this object in the top viewport. Choose Create/Object/
Copy and press Tab until the arrows point left and right only. Click on the 
object and create a new object to the right of the original one. Name this 
object FullChapel. Choose Display/Hide/By Name and hide the object FullChapel. 
This object remains the same, and the other object is squashed.

To squash the second chapel model flat against the front viewport. Choose 
Select/All to select all vertices. Next, choose Modify/Vertex/Align, click 
on the Selected button, and then click on the chapel model in the front 
viewport. Choose Display/Geometry/Full Detail. Your model should look like 
figure 9.10.

The model looks the same in the front viewport, but appears as a flat line in 
the top, left, and user viewports. The chapel has been squashed flat as a 
pancake against the front viewport.

Go to the keyframer and make sure all objects except FlatChapel are hidden. Go
to frame 30. Choose Object/Morph, and click on the chapel. A list appears that
contains two objects, FlatChapel and FullChapel. Choose FullChapel. Make a
preview of the animation in the user viewport. When it is done, watch the 
preview. The chapel will "grow" from 2D to 3D.

To get a better idea of exactly what is happening, you can also make a preview 
of the left viewport. Before doing this, move the omni light on frame 0 to the 
left side of the model for better illumination. To see a perspective view, 
assign a camera to the model and preview the camera viewport.

This morphing technique can be used to make a blueprint drawing turn into a 3D
object or cause a picture hanging on a wall to come to life.

Animated Backgrounds

Backgrounds can add a lot to an animation. Scanned images of landscapes make
attractive backgrounds for architectural exteriors. Your company logo, for 
example, can be pasted up as a background to an animated presentation. 3D 
Studio also enables you to use a FLI animation file (flic) as a background, 
adding yet another dimension to your animation.

When you use a background, your rendering time does not significantly increase.
Because 3D Studio does not have to compute the image, it simply pastes it up
before rendering the model for that frame.

One limitation to the use of backgrounds is that shadows cannot be cast upon
them. If you need shadows on your background, you must create a "plate" object
at the back of your model and texture-map the background image onto it. You
can then cast shadows on the plate, but the rendering time will be much greater
than if you had used the background option.

If you use an animated background, the resolution of your FLI file must be the
same as the rendering resolution. You can create FLI files of any resolution 
with 3D Studio or Animator Pro.

If you map the background animation onto a plate, the FLI file can be any 
resolution. Even if you are using VGA, you can create a high-resolution FLI 
file and use it as an animated texture map in a 320x200 animation.

Linking Objects

Occasionally, you may want to have objects move together within the animation.
You can always attach the objects in the 3D editor, but this means you have to
assign the same material to both objects.

Alternatively, you can use the keyframer to link objects so they move together 
but still remain separate objects. This process, called hierarchical linking, 
is covered in detail in 3D Studio Tutorial 13.

You can also use linking with lights and cameras. This technique, called slaving,
enables the light or camera to become a slave to the object to which it is 
connected, moving when the object moves. For an interesting effect, try linking 
a spotlight target to a moving object. The target moves as the object moves, 
so the object will be lit throughout the movement. If the spotlight casts shadows 
on another object, the shadow will, of course, move when the light moves. This 
type of animation can be used to create interesting effects.

The following exercise illustrates one use of a linked spotlight target casting 
a shadow on another object. In this exercise. you link a spotlight target to an 
object and see the shadows it casts on a background object. The caster model is 
used for this exercise.

1. In the 3D editor, load the caster file that you created in the Object 
Animation section above, CASTERM.3DS.

2. Create a plate to serve as a background to the animation. Choose Create/Cube 
and place a cube behind the caster in the top viewport. Name
the object Bgrd Plate. Choose Modify/Object/2D Scale to make the
cube into a plate shape. Your viewports should look like figure 9.11.

3. Place an omni light at the top front of the model for illumination and
place a camera at the right of the caster.

The model is illuminated by the omni light, but a spotlight is required
to cast shadows.

4. Place a spotlight in front of the model, slightly to the right, pointing
right at the caster. Turn the Shadow button on. (See figure 9.11 for
correct spotlight placement.)

5. Assign materials to the objects, if you like. The plastic materials are a
good option for spotlit objects as they have an attractive, shiny look.

6. Go to the keyframer.

NOTE

In linking, one object is referred to as the parent object and the other is the
child object. To determine which is which, remember these phrases:

- Children follow their parents wherever they go.

- Parents don't always stay home with their children.

The parent object is the leader while the child is the follower. in this case,
the light follows the support bracket, so the light is the child.

7. Choose Hierarchy/Link. When you are asked to choose a child object,
click on the spotlight target; when asked for the parent object, click on
the object Support.

8 Render the animation with Phong shading and shadows turned on. You
see a shadow on the background plate behind the support bracket as it
moves.

Note that shadows do not appear in a preview; you must render with Phong 
shading in order to see shadows.

Camera targets can also be slaved to objects. Try connecting the camera target 
to the support bracket for an interesting effect.

Animating Natural Phenomena

The natural environment consists mostly of objects with irregular surfaces and
colors. To create convincing models of mountains, water, and human beings thus
requires a little more ingenuity than making models of man-made objects.

For example, the effect of rippling water can be created with an animated bump
map. When the bump map is created, however, you must be careful to make the
ripples natural-looking. The process of making a map look organic often takes
longer than the process of making man-made maps. When you create animated
maps for natural phenomena, be sure to use several (10 or more) frames to make
random the motion of the map.

Animating Terrain

Terrain models offer some interesting opportunities for animation. You can 
create a classic fly-around or flatten the terrain and morph it to its final 
shape in the keyframer. A terrain mesh file can be imported into 3D Studio via 
a DXF file. For information on what packages to use to generate these DXF 
files, see Chapter 8.

To morph a terrain mesh. use the technique described previously in this chapter.
Make two copies of the mesh and flatten one of them with the Modify/Vertex/Align 
command. Morph from the flattened mesh to the full-size mesh in the keyframer.

Texture and bump maps can be used to give the terrain a rough surface. Chapter
8 describes how to create texture and bump maps that can be tiled over terrain
for a realistic look.

Animating Water

If you have an outdoor scene that includes water, consider animating the water
for an intriguing effect. Animated water can be created by using terrain-type
meshes and morphing techniques. First, obtain two or more dissimilar terrain
meshes. In the 3D editor, use the Modify/Object/2D Scale option to squash the
meshes so they are nearly flat.

In the keyframer, morph from one mesh to the next. The resulting effect looks 
like rippling water. If the meshes are very close to flat, the rippling will 
be subtle; meshes with more height create the effect of water on a very 
windy day.

For further realism, an animated texture map can be created in Animator or 
Animator Pro. A good animated water map can be created by making a flic with 
10 or more frames and using the Spray tool to spray various blue and white 
tones onto each frame. If you spray each frame individually, each one will 
look completely different. When the flic is mapped onto the water mesh, it 
adds to the illusion of rippling water.

An animated water bump map can also be applied to the moving water mesh. In
fact, the animated texture map can also be used as a bump map to get a match of
colors and bumps. When you create water material, set the shininess to 75% or
higher (water is very shiny).

The DXF mesh files used for water animation must be created with software other
than 3D Studio. Chapter 8 describes the tools required to make terrain-type mesh
files that can be used for water simulation.

Animating People

Three-dimensional human beings can sometimes be utilized in a presentation.
You can, for example, include a few persons around an architectural model to
show its scale, or you can place your mechanical design on a 3D hand to illus-
trate that it fits into a space that small.

It can be a time-consuming process to model realistic three-dimensional human
beings or body parts. If you need this type of model, you can purchase prefabri-
cated ones. Two companies that currently offer 3D Studio-compatible files of hu-
man beings are listed in Appendix B.

Models of people already have hierarchical linking built into them, which means
that when you move or rotate the hand, the arm moves as well.

If the 3D people in your rendering are located at some distance from the camera,
you don't need as much detail as these packages offer. Simpler models, built 
from scratch in 3D Studio, will look fine if they are very small compared to the  
size of the rendered image.

Animation Tips

Animation can often take a long time to create, especially if you have to render
the same animation several times before it is right. You can take certain steps,
however, to minimize the time between conception and final presentation.

The final rendering is usually the most time-consuming part of animation 
creation. There are a few ways to make sure you only have to render once 
and that you get it right the first time.

Frame Count

In the rush to finish your project, you might make the mistake of putting too 
few frames in the animation. This is especially important when creating FLI 
files to be played in real time on the computer screen.

A preview gives you an idea of what the final movement will look like. If you are
creating an FLI file, the preview also tells you if you are using enough frames for a
smooth animation. If your preview is jumpy and cannot refresh the screen fast
enough to look smooth, your FLI file will have the same problem.

Videotape plays at 30 frames per second. The default speed for FLI flles is about
15 frames per second. Before setting the total number of frames for your 
animation, run through the animation in your head, and time it as you envision 
it. In the keyframer, make sure you have set the total number of frames to match 
your visualization.

Previewing

Previewing is an extremely useful tool for testing your animation before you 
render it. Although a preview does not show mapping or shadows, it does show 
the motion of the objects. Always preview your animation before you render 
it, so you can see any problems that appear (and avoid wasting valuable 
computer time rendering a faulty animation).

It is recommended that you use frame numbers in your preview. When you
choose Preview/Make, click on the Yes button next to Numbers on the Make 
Preview panel. When you watch the preview, the frame numbers make it easier 
to pinpoint problem areas of the animation.

Trial Runs

When you complete a preview, you can check your mapping and shadows by 
performing a trial rendering at a low resolution. To render a trial 
animation, follow these steps:

1. Choose Renderer/Setup/Configure and click on 320x200. The correct
aspect ratio for this resolution is automatically calculated.

2. Change the width to 160 and the height to 100. Leave the aspect ratio
as is. The ratio for 160xlO0 is the same as the one for 320x200.

3. Choose Flic as your file output.

4. If your animation is long, don't render every frame in your trial run
(unless you have some particularly tricky animation). When you select
Renderer/Render, choose to render every 2nd or 5th frame, depending
on the length of your animation.

5. A 160xlO0 flic can be viewed with 3D Studio or Animator Pro. 3D Studio 
plays the flic at the default speed of 15 frames per second and won't
let you stop the flic. If you want to examine your trial flic, use Animator
Pro to step through the frames at your own pace.

Final Rendering

When you perform your final rendering, change the display driver to Null. This
greatly improves the speed at which 3D Studio renders each frame.

When 3D Studio renders an animation to Targa files, it saves each frame to a
separate TGA file. The file names consist of the first four letters that you 
specify, followed by four numbers. For example, if you render an animation 
and give the disk flle the name Taurus, the keyframer names the files 
as follows:

TAUR0000.TGA
TAUR0001.TGA
TAUR0002.TGA
TAUR0003.TGA

When you start the rendering at a frame other than 0, the keyframer still starts
the numbering sequence at 0000. This means that if you split your rendering
sessions over several days, you use the file name TAUR0000.TGA several times.
To keep from overwriting your TGA files, you must assign a file name that has
letters that are different from the file name used in the previous rendering ses-
sion.

Batch File Rendering

Rendering of 3D Studio models can also be performed from the DOS prompt. For
a detailed explanation of the commands and parameters involved, see the
README.DOC file that is included on your 3D Studio disks.

Summary

Animation in 3D Studio is accomplished with the keyframer, which can animate
objects, cameras, and lights.

The Track Info panel and Key Info panel can be used to make animation easier.
Use the Track Info panel to move or copy information to other frames; the 
movement is smoothed with the Key Info panel.

Animation of the camera is easy to create. When you alter the camera's lens
length, you can create clever cinematic effects. You can also use animated 
Spotlights for stunning images.

Morphing, in which all the vertices of one object move to gradually change into
another obiect, is useful for visual tricks.

Natural phenomena, such as water, terrain, and sky can be utilized with an
AutoCAD model to improve a presentation. Morphing helps to animate natural
phenomena realistically.



9.  W A L K T H R O U G H   A R T I C L E

Creating an animated walkthrough

HISTORY, MODERN TECHNOLOGY, AND GLASSWARE

By Gerald O'Conner
and Elizabeth Yawger

When commissioned to design and
animate the new Oglebay Glass Center,
which was to be built in Wheeling, West
Virginia, the team of Pittsburgh, Pa.-
based Bennardo & Churik Design and
Summit Graphics Inc. saw the potential 
to combine their talents. Bennardo
& Churik Design is an interior-design
firm specializing in retail design and
planning. Summit Graphics provides
three-dimensional computer rendering 
and animation services to architects 
and designers. The two firms have
worked together on numerous retail-
design projects, providing a complete
design and presentation package. The
new Glass Center presented a challenge 
uniquely suited to the combined
talents of these two firms.

The design was developed from his-
toric photos of a carriage house that
was formerly located on the Oglebay
property. The placement of the new
Glass Center, directly adjacent to the
historic visitor's center, created some
concerns among the park commission-
ers about the architectural compatibility 
of the two structures. These concerns, 
coupled with concerns
regarding the sensitive nature of the
landscape and the need for committee
approval of the project, necessitated
the use of computerized rendering and
animation.

The project included interior and
exterior animations of the new and 
existing facilities. The exterior animation
was created to illustrate the visual 
impact of the project on the existing site
and the interior animation 
demonstrated the display possibilities of the
first-floor gallery. Because of the size
and complexity of these drawings, the
interior and exterior 3D models were
created separately.

The exteriors of both buildings were
created in AutoCAD Release 11 from the
two-dimensional floor plans and 
elevations, using extruded lines, closed 
polylines, and 3D faces. The site was also
created in AutoCAD using 3D polylines,
ruled surfaces, and edge surfaces.
Three-dimensional boundaries of the
site were created at key points, such as
the grading against the building and
points along roadways and pathways,
using AutoCAD's 3D polylines. These
boundaries were then used to create
the ruled surfaces and 3D meshes of the
landscape.

The interior presented a different
challenge. In addition to creating the
building's interior and display fixtures,
it was also necessary to fill the store
with glassware to give a more accurate
representation of the design. The 
volume of detail required to create this 
impression and the resulting 3D Studio
file size presented a challenge not only
for the person creating the model but
also for the animation hardware and
software. The finished interior model
used more than 250,000 faces and re-
quired more than 70 megabytes of RAM
to render without a swap file. Pieces of
the interior model were created in
AutoCAD and 3D Studio and were
merged together in the final 3D Studio
file. The accuracy and speed of AutoCAD 
made it the clear choice for creating 
the bulk of the structure and fixtures, 
while some of the unique
modeling capabilities of 3D Studio
made it ideal for creating oddly shaped
objects such as the wing backed arm
chairs.

The final interior plan was imported
into 3D Studio without the glassware
using the DXF converter built into 3D
Studio Release 2. The glassware was not
added to the AutoCAD drawing 
because, once in 3D Studio, each piece
would require separate mapping 
coordinates. Instead, a separate 3D file was
created for each of the different glass
shapes and packages. Each object was
then given its own 3D Studio mapping
coordinates before being merged into
the final interior drawing. They were
then copied throughout the interior
and assigned various colors and 
textures without having to reapply 
mapping coordinates. The glass pieces were
constructed with a minimum number
of faces, and smoothed with 3D Studio's
smoothing groups., which allowed a
high level of detail, even though each
piece of glass required no more than
200 faces.

Colors and textures were applied to
objects in a variety of ways. 3D Studio
can create object textures using 
ambient, diffuse, and specular color
reflections as well as any combination 
of texture, bump, opaque, and reflection
mapping. We have found one of the
best methods for setting up materials in
3D Studio is to organize the AutoCAD
drawing so that each layer represents a
different material. When the AutoCAD
drawing is imported into 3D Studio
using the DXF converter and choosing
the layer option, an object is created for
each layer of the drawing and retains
the layer name as the object name. 
Because this method results in objects
with logical names, the assignment of
mapping coordinates and textures to
even the most complex models can be
done quickly and efficiently.

Creating an animation in 3D Studio
is a process of establishing a series of
keyframes through which the camera
or objects will travel at specific points in
time. As these keyframes are created,
3D Studio automatically creates a
smooth spline path through each 
keyframe, calculating the position of the
"tweens" (frames between the keys).
Changes made to the keyframes are 
instantly reflected in the tweens. Each
object in a scene can have its own 
keyframes and paths. A keyframe for one
object does not necessarily have to be a
keyframe for any other object. This
ability to specify independent points in
time for the movement of a camera, 
target point or object makes creating a
complex animation relatively simple
and intuitive.

The finished animation begins with
a wireframe view of the site from 
overhead, drops down to eye level, then
cross fades to a fully rendered image
while continuing to move. This effect
was created by rendering two seconds
worth (60 frames) of animation in both
wire frame and fully shaded modes
where the transition occurs. These
frames were then overlaid on one 
another using 3D Studio's video-post
fade-in option.

Once it is fully shaded, the 
animation completes a full circle of the site,
arriving at the front door of the new
Glass Center. As the camera approaches, 
the doors open automatically, 
revealing the interior gallery. The 
animation continues through the interior and
ends with a view of the historic Oglebay
buggy. This smooth transition between
interior and exterior was created by
merging the exterior animation path
into the interior drawing. The interior
animation was then added to this path,
creating a single animation path of the
interior and exterior. To create the 
transition frames where interior and 
exterior are seen simultaneously, a third
model was constructed by merging the
interior model into the exterior model.
Because both drawings were created
from the same original AutoCAD 2D
floor plan, they had identical origin
points and merged exactly in place.
A total of 1,935 frames were required
to produce an animation slightly longer
than one minute at 30 frames per 
second. These frames were rendered to
disk on three separate 486 machines
using 3D Studio's batch-rendering
mode. On 486/50 machines, the frame
time ranged from a low of 10 minutes
for some exterior frames to a high of 30
minutes for some interior frames. Once
the frames were rendered, they were
recorded frame by frame onto video
tape using a Sony EVO9650 recorder
and Diaquest controller software.
The use of three-dimensional 
visualization and animation throughout the
design phase was instrumental in creating 
a design that enhances, rather
than competes with, the existing site.
The animation created a clear, accurate
representation of all the elements 
involved in the design, and helped to 
instill a level of confidence and 
understanding that moved the project
through the approval process. 


10.  B E V E L E D  T E X T

Using 3D Studio with Corel Draw!

Every now and then a couple of
software packages come along
that complement each other very nicely.
CorelDraw! and 3D Studio are such a case
and I'll try to show the strengths of their
relationship. This article is written using
as its references CorelDraw version 3.0
rev. B, and Autodesk 3D Studio release 2.0
running on an Intel 80486DX-33.

The Yost Group's 2D Shaper module in
3DSr2 is a fine implementation of a 2D
draw program. But you may already have
experience designing with a full featured
vector-based illustration program like
CorelDraw, or just enjoy having instant
access to the 256 outline fonts that come
bundled with that program. Perhaps
you'd like to use some of CorelDraw's
special effects for a 3D logo design. Virtu-
ally any combination of text, lines, bezier
curves, polygons, etc., are exportable to
the 2D Shaper via the .AI format, pro-
vided you follow these steps:
 
1) Make absolutely sure that any closed
polygons have NO fill.
 
2) Use the Ungroup and Convert-to-
curves command on all entities.
 
3) Use the Break Apart command on all
entities, especially those containing
"nested" polygons.
 
4) Choose Illustrator 88 format, and Ex-
port-as-curves in the Export dialog box.

5) Load the file into 3DS's 2D Shaper.
Use the Select/Vertex/All command followed 
by the Modify/Vertex/Weld/Selected 
command to weld coincident vertices 
of closed polygons that "open" during
the export/import process. It's also a good
idea at this point to check on the validity
of the Shapes, just to be sure. Use the
Shape/Assign/AII command followed by
the Shape/Check command to verify that
you are starting with valid shapes. Now
you are ready to use the file the same way
you would use any 2D Shaper-defined
Shape. Try some headline-style text that
has been edited with CorelDraw's Edit
Envelope feature. After lofting and 
rendering in 3D Studio, you can end up with
quite an effect in very little time.

A great effect with 3D text is beveling.
To do this, first we will create two different 
sets of Shapes that represent the text--
one for the back and middle of the text,
and one for the front, or beveled edge.
These shapes will then be placed on three
vertices of a straight path for lofting into
3D objects. To reliably bevel complicated
typefaces (or any complex shape) in 3D
Studio, try the following steps. (If you are
dealing with a file imported from
CorelDraw, be sure to complete the above
steps first!)
 
1) In the Shaper, use the Create/Outline
command to make equal offsets of each
letter, including the inner "nested" poly-
gon, if it has one. The size of the offset
depends on the amount of bevel that you
are looking for.
 
2) After using Shape/Assign/AII to
designate all polygons as shapes, use the
Shape/Check command to check the
Shapes for "crossovers"--places where
two or more vertices/segments self-inter-
sected during the outline process. The
Shaper will highlight them for you, then
you can zoom in and adjust their positions
by hand so that they match the original
polygon's shape. You may also need to
adjust spline tension on some of the verices
at this time to retain the integrity of the 
typeface.

3) Use the Display/First/On command to
monitor the position of each Shape's
"first" vertex. It is very important that all
first verices are perfectly in line for the
lofting process to work. For instance, for
the letter B, if the first vertex of the
outer polygon occurs at the upper left-hand
corner, the smae must be true for it's offset.
The same is true for nested polygons and their
respective offsets. Use the Display/First/Choose
command to ensure that all first vertices occur
inline.

4) In the 3D Lofter, use Path/Insert to insert
a "middle" vertex on the path and position it so 
that it is near where the front of the typeface
will be (be sure to straighten the path before
lofting). The distance between these vertices will 
equal the "thickness" of the bevel (this is 
not critical and is easily edited later in the 3D Editor).
 
5) In the 2D Shaper, use the Shape/Assign 
command to assign all the outer-most
outlines as Shapes, as well as the 
innermost outlines on those letters that have
nested polygons. In the 3D Lofter, use the
Shape/Get/Shaper command to place this
set of Shapes on the "back" and "middle"
vertices of the path. Now repeat this 
process, this time using the Shapes that form
the inner outlines of the letters--the ones
that will make up the beveled face and
position them on the "front" vertex of the
path. Now you are ready for lofting
(tweening is optional and you may have
to turn off optimization).

If the rendered result looks impossibly
"inside-out," re-check the Shapes' first
vertex placement, which is sure to be the
culprit!

Also noteworthy about CorelDraw is its
ability to "rasterize" a file. This means that
any combination of text, vector graphics,
and imported bitmaps can be exported as
high-resolution 24-bit TGAs or other
bitmap format. This opens up countless
ways to produce texture, opacity, and
bump maps for 3D Studio, often more
quickly and easily than a paint program
would. Product/packaging prototypes are
easy when the art, text, etc. is assembled in
CorelDraw, and exported as a texture map
for quick rendering in 3D Studio.

Need to create a 3D version of a
scanned document? Use CorelDraw's
CorelTrace to accurately "vectorize"
bitmaps to import and loft in 3D Studio.

Finally, CorelDraw is a fine tool to size
composite, and further embellish your
3DS renderings for printing or going out
to slides especially if further headlines,
text, etc. are needed.

For input or output, CorelDraw is a
good companion program to Autodesk's
3D Studio.


11.  Q U I C K  T U T O R I A L

Using 3D Studio
Version 2.0

Autodesk offers another excellent PC-based graphics package (in
addition to AutoCAD) called 3D Studio. 3D Studio (3DS) is a 3D
modeling, rendering, and animation package that creates images and
animations that appear photorealistic. Photorealistic is a term often
used to describe computer-generated images that look so realistic that
you can mistake the images for photographs. You can create these
complex scenes with the advanced modeling tools of 3DS and then give
the scenes dimension with animation that you can save on videotape or
play back in real-time on VGA PCs. You can add textures, surfaces, and
reflections to give 3D objects depth and a photographic quality. You
also can import models from AutoCAD by means of the DXF image
format.

Three basic functions are covered by 3D Studio: modeling, rendering,
and animation. Modeling is the process of building or creating 3D
scenes. Modeling also involves assigning surfaces to wireframe objects;
this process gives wireframe objects a realistic appearance.

Rendering uses the scenes that were modeled and generates still 
pictures. Although this definition may sound simple, the computer must
perform tremendous mathematical calculations to simulate real-world
conditions in creating a still image.

Animation involves choreographing the movement of objects. This
single task is so complex that many 3D rendering systems cover only
modeling and rendering. For animation, the program creates many still
images and then plays them back at a high speed, much as motion pic-
tures are created and shown. 3DS offers an intuitive modeler with a
high-speed, realistic renderer. When combined with smooth animation,
the results are beautiful.

3D Studio Version 2.0 brings such advanced capabilities as visible cam-
era paths, procedurally generated surface textures, Boolean modeling,
and automatic environment maps for simulating accurate reflections
and a new programming interface called IPAS. IPAS is an acronym 
derived from the four kinds of external processes available: 
Image-processing, Procedural modeling, Animated stand-in, and Solid pattern.
IPAS gives you the capability of creating stand-in C language routines.
These routines can interface directly with 3DS to create special effects
such as particle systems, algorithmic-based object transformations,
and photographic-style special effects.

The rendering engine of 3DS now comes with a site license so that you
can run batch mode renderings on as many PCs as you have available
(the modeler, however, still is licensed to only one machine at a time
and is enforced by a hardware lock).

The advantage of 3DS over AutoCAD's AME (Advanced Modeling 
Extension) is the fast, free-form modeling control you have without regard to
dimensions and precision settings. Compared to AutoShade, 3DS also
gives much more rendering control, with a Materials Editor, on-screen
camera, and light positioning. The Materials Editor enables you to
quickly create realistic textures and surfaces; you then can apply these
features to your geometry.

Because the animations are created in 3D, the output is much more
realistic than hand-drawn animations created in Animator or Animator
Pro, which are flat 2D graphics. After the animation is finished 
rendering, you can load the graphics in Animator or Animator Pro for 
retouching and special hand-painted effects.

Although 3DS is a capable 3D rendering, modeling, and animation 
package, it may not handle geometry or models as large as AutoCAD and
AutoShade can handle. Another limitation is 3DS's lack of support for
any high-speed graphics processors such as the Intel i860. 3DS is more
oriented toward graphic art than engineering purposes, so the output
options are directed to media such as video editing systems, rather
than to plotters.

Installation and Hardware
Requirements

The user-friendly 3DS installation program prompts you for paths and
default settings. If you lack sufficient hard disk space, the installation
program notifies you and ends the installation procedure.

3DS requires a 386 CPU or better and a math coprocessor (Intel or
Weitek). Although performance on a 386-based PC may be acceptable,
the added cost of purchasing a 387 math coprocessor may put you in
the price range of a 486DX-based PC. A 486DX not only has the Intel
math coprocessor built in but also performs better than a 386 PC.

To install 3D Studio, place the #1 disk in drive A and enter a:install at
the DOS prompt. The installation program prompts you for any neces-
sary information. When presented with a list of items, use the up- and
down-arrow keys to make selections. For the first installation, accept
the default settings for disk paths.


Hard Disk Space

You need at least 12M for the program and sample data files, with a
minimum of 10M for working storage. (A typical VGA animation with
100 frames can easily take 2M. lf rendered to videotape and played
back at 30 frames per second, the animation lasts only a little more
than 3 seconds!) Hard disk space also is used for swap files when 3DS
runs out of memory so that you use the hard disk to emulate memory
storage.

Memory

You can never have enough memory; with 3DS, memory is a significant
performance enhancer. When 3DS runs out of memory, 3DS starts
swapping to disk. Because disk swapping can slow rendering to a crawl,
install enough RAM (random-access memory) to prevent this situation.
The 3DS Installation and PeffonTlance Guide states that the minimum
RAM requirement is 4M, but you may find 8M a more realistic figure. If
you want to create high-resolution images for video or large models
with tens of thousands of faces, consider getting 16, 32, or even 64
megabytes of RAM. Before purchasing any PC to be used with 3DS,
make sure that the motherboard can hold at least up to 32 megabytes
and preferably 64 megabytes of RAM.

Math Coprocessor

During installation, you are asked what kind of math coprocessor you
installed. lf you are running on a 486DX-based PC, an Intel math
coprocessor is built into the CPU. For more information on choices for
CPUs and coprocessors, see ChaPter 2.

For more speed, 3DS supports the Weitek math coprocessor, which has
built-in 3D math functions. With the Weitek, you can expect a speed
increase of about 200 percent with 386-based PCs but only about 30
percent on 486DX-based PCs. Before you purchase a Weitek chip, make
sure that your 486/386 motherboard supports the coprocessor; some
lower-priced motherboards lack a Weitek math coprocessor slot. Some
386-based motherboards use a combination math coprocessor socket
for either the Weitek or the Intel chip. Therefore, you can install only
one chip at a time. lf so, remember that AutoCAD doesn't support the
Weitek math coprocessor. lf you need to run both AutoCAD and 3DS,
beware of this potential conflict.

CAUTION: Math coprocessors can be difficult to install. lf the
Intel coprocessor is incorrectly installed or if the pins aren't fully
seated, 3DS cannot run. lf the Weitek coprocessor is incorrectly
seated, 3DS attempts to run and then reports spurious errors in
the 3DS.SET file. lf 3DS starts to run sporadically and you haven't
changed the 3DS.SET file, check the Weitek chip--chances are
high that the chip isn't seated correctly.

Video Adapter

All VGA (8-bit color) video adapters work well with 3DS. You can even
render and view truecolor (16/24/32-bit color) images on a VGA by
palette-optimizing and dithering. Optimizing looks at all colors needed
for an image and determines which of the 256 colors best represents
the image. You then can dither (closely mix) these colors to give the
illusion of even more colors. lf you print most of your work with true-
color output, consider a true-color frame buffer such as the Targa+
adapter from Truevision. The Targa+ adapter not only provides up to
32 bits of color information, but also outputs NTSC-compatible signals
for easy transfer to video editing systems (NTSC stands for National
Television Standards Committee--or Never The Same Color).

When choosing a video adapter for 3DS, make sure that the adapter has
good driver support. Drivers come either in the form of an ADI driver
bundled with the video adapter or from third-party driver developers
such as Panacea Inc. or Vibrant Graphics Inc. A high-resolution display
driver is efficient because you don't often need to zoom in to see
details.

Another consideration with the drivers is the possibility of graphics
coprocessors such as the Texas Instruments 34020 and 8514/A
compatible video adapters. The speed of video adapters that use
hardware-based line drawing routines can be outstanding. Some adapt-
ers are fast enough to give real-time previews of animation in wireframe
mode, which eliminates the need to render preview animations to
check an object's movement.

Video Monitor

Your best choice is a high-resolution monitor, at least 1024x768 non-
interlaced. A cheaper monitor that uses interlacing to access higher
resolutions may flicker and cause eyestrain. The higher the refresh
rate, the better; a vertical refresh rate of 72 kHz or higher is ideal.
To take full advantage of higher resolutions such as 1024x768 and
1280x1024, consider a 19- or 20-inch screen. lf you are using a true-color
video adapter such as the Targa+, a dual-monitor configuration may be
preferable. With this configuration, you can view the rendering and the
drawing viewports at the same time.

Output Devices

3DS supports raster hard-copy devices that use ADI drivers. The pro-
gram also supports various file formats, including PostScript, Targa,
TIFF (Tagged Image File Format), and GIF (Graphics Interchange For-
mat). Multiple images can be rendered to create an animation. You can
save the resulting animation as a series in any of these file types or
save the animation in Autodesk Animator or Animator Pro format. lf the
frame buffer is connected to video editing equipment, you can render
each frame directly to tape, without storing the frames sequentially on
a hard disk.

Optimizing 3D Studio

After installation is complete, the \3DS directory contains extra files
that you can deletc particularly MID files. The installation procedure
leaves these files on disk, but the files provide no useful function after
3DS is installed.

All 3DS default settings are stored in a configuration filc 3DS.SET. This
file contains paths, driver information, external program paths, and
modeling defaults. For now, leave this file as installed. Complete details
on all available settings in the SET file are documented in the Installa-
tion and PeffonTlance Guide.

The main 3DS executable program file comes in two versions: 3DS.EXE
and 3DSHELL.EXE. 3DS.EXE uses all available memory and gives the
best performance. 3DSHELL.EXE reserves memory for use by DOS. This
version also enables you to run other (external) programs such as
Animator without having to leave 3DS. The advantage of this capability
is quick access to other applications. The downside is a lower limit on
available memory. The choice depends on the amount of memory in
the system and the kind of rendering you plan to do. lf you don't often
need to exit to Animator to make texture maps, or you have 8M or less
RAM, use the 3DS.EXE version.

3DS uses the PharLap 386 DOS/Extender, which requires no special
memory drivers in the CONFIG.SYS file. lf you must run other PC appli-
cations that require an expanded, extended, or XMS memory, PharLap
is compatible with DOS 5.0's HIMEM.SYS and Quarterdeck's
QEMM386.SYS files.

Learning 3D Studio

After you install 3DS, you are ready to familiarize yourself with the
application's features while learning the steps to create an animation.
Begin by creating a chrome flying logo, which spins in a continuous
loop.

Although you will create 3D objects and environments, most computers
cannot display stereo 3D or accept 3D input. You are limited to viewing
the 3D world on a 2D screen and to manipulating that world with a 2D
input device (the mouse or digitizer). Because of these limitations, you
must do 3D creations in a controlled, organized manner. To make this
task easier, 3DS is divided into five modules. Three modules are dedi-
cated to the creation of 3D objects, one module is for creating textures
and surfaces for the objects, and the final module is for animating the
objects.

Depending on the version, begin the tutorial by typing 3ds or 3dshell at
the prompt in the directory where 3DS is installed. After 3DS loads, the
program takes you to the 3D Editor module. You may recognize the
standard Autodesk user interface with command column on the right, a
status line and menu bar at the top of the screen, viewports in the
middle, and a prompt line at the bottom (see fig. 20.1). You also see an
icon panel for quick access to standard operations. The icon panel is
located in the lower right corner of the screen, below the command
column.

You start creating many 3D objects with the 2D Shaper. Here you can
draw a 2D outline, which you later can loft (or extrude) in three dimen-
sions by taking the following steps:

1. Move the cursor to the top of the 3D Editor screen. As the cursor
moves to the status line, a menu appears, which you can use to
access information about the current 3D model, files, views, and
program--which takes you between the different modules of 3DS.

2. Click Program. When the menu drops, you see that all five 3DS
modules are listed. You use this menu to change modules.

Each module is assigned a function key. The 2D Shaper is assigned
to F1.

3. To select the 2D Shaper, click 2D Shaper or press F1.

4. Select Create from the command column. An indented submenu
opens.

5. Choose Text from the submenu. Text offers three more choices:
Font, Enter, and Place.

6. Select Font. This option opens a dialog box that lists the fonts in
the \3DS2\FONTS\ directory.

7. Select NEWEXBLD.FNT and click OK (see fig. 20.2).

  The extension FNT is used for 3DS format fonts. With 3DS
  Version 2.0, you also can load Type 1 PostScript fonts (PS).

T I P 

As you review the tutorials, make mental notes of and practice using
the shortcut keys. These keys save time when you use 3DS.


After you choose a font, type the logo text. To enter the text, follow
these steps:

1. Select Create from the command column, and then select Text.

2. Select Enter. This option opens a dialog box where you type the
text.

3. Type QUE and press Enter or click the OK button to close the
dialog box.
The following steps transfer the text to 2D Shaper.

4. From the Text option in the command column, select Place.
Move the cursor to the viewport. Press and hold the Ctrl key
down and create a box similar to figure 20.3. Click once to set the
er left corner and click again to set the lower right corner. The
word QUE should appear in the NewExBld font.

After the first click, release the Ctrl key. Holding down Ctrl for the
first click keeps the correct aspect ratio (height and width of the
original font) of the text during placement.

Your work in the 2D Shaper is done and you are ready to move on to
the 3D Lofter.

3D Lofter

The 3D Lofter takes any 2D shape that is in the 2D Shaper and extrudes
(lofts) the 2D shape into three dimensions. A shape that you loft can be
deformed by scaling smaller or larger, twisting, beveling the edges, and
many other deformations. Follow these steps to extrude the word QUE
into three dimensions:

1. To enter the 3D Lofter, select the Program menu and then se-
lect 3D Lofter. If you prefer to use the keyboard, press F2 (see
fig. 20.4).

2. Select Shape from the command column; from the submenus,
choose Get and then Shaper.

This series of commands pulls in the text from the 2D Shaper.
Because you didn't specify polygons you wanted to loft in the 2D
Shaper, 3DS displays a dialog box asking if you want to use all 2D
Shaper polygons. Select Yes.

3. When QUE appears, select Shapes and then Center to center the
text on the lofting path. The lofting path is the direction the 2D
shape follows to be extruded into a 3D object. This path can be a
straight line as in the example, or you can create a more complex
lofting path, such as a circle for revolving a cross-section of a
glass. Text even can be used as a lofting path; lofting a circle along
the path of text creates a "neon" sign.

To better see the text, use the Zoom Extents button (the 3D
square box) in the icon panel below the side menu. When you
click this button, the current viewport is magnified to display all
visible objects.

4. Right-click the Zoom Extents button. Notice that this action magni-
fies all viewports simultaneously (see fig. 20.5).

5. Because you are making the text 3D without special deformations,
click Objects and then Preview in the command column. 3DS dis-
plays the Preview Controls dialog box (see fig. 20.6).

6. To accept the defaults, choose the Preview button in the dialog
box. In the other viewports, you see a preview of how the logo will
be shaped when extruded. The Preview option saves time in veri-
fying the shape of a loft before actually creating the 3D object.

7. Because the text loft looks fine, choose Objects and then Make
from the command column. 3DS displays the Object Lofting Con-
trols dialog box (see fig. 20.7).

8. Type LOGO in the Object Name text box and then click the Create
hl]ttnn

3DS displays a status box (Object Creation Progress), showing
the faces being created for the 3D version of the logo text (see
fig. 20.8).

9. After the object is created, switch to the 3D Editor by pressing F3.

3D Editor

The 3D Editor is where most work in 3D Studio is done. You can as-
semble 3D objects that you lofted. You can create new geometric 3D
objects such as spheres, tori, cubes, cylinders, and so on. The 3D Edi-
tor also is the module that assigns surface textures to objects. You also
can create lights and cameras in the 3D Editor.

After you enter the 3D Editor, you see four viewports, with Top, Front,
Left, and User views selected (see fig. 20.9). The User view is an 
isometric view of the current 3D scene. The first step to take with any new 
3D scene is to create lights to illuminate objects and a camera to view 
objects in perspective.

Follow these steps to create the lights:

1. To zoom out all four viewports at the same time (so you can see
the model better), right-click the Zoom Extents icon and then
right-click the Zoom Out icon (four arrows pointing away from
the center) twice. Right-clicking the zooming icons causes all the
viewports to be zoomed. Left-clicking only zooms the current or
active viewport.

2. Before you can render the model, you need to add a light source.
If no lights are added, the model or scene is in the dark and a
black screen appears if you try to render it. Three different kinds
of lights are available: ambient, omni, and spot. For now you will
deal only with omni lights. From the command column, choose
Lights, Omni, and then Create.

3. Move the cursor to the lower right corner of the Top viewport and
click the Light Definition dialog box to create omni lights (see fig.
20.10). With this box, you can create lights to illuminate with any
color, identify a light's name, adjust the color and brightness with
the RGB (Red, Green, and Blue) or HLS (Hue, Luminance, and
Saturation) slider bars, and use the On and Off buttons to turn a
light on or off. For now, accept the default settings by clicking
Create.

4. To illuminate the model from more than one direction, you create
a second light source. This second light will be much dimmer than
the first so that it doesn't overpower the first light. Move the cur-
sor to the upper left corner of the Top viewport and click to cre-
ate a second light. Drag the Luminance (L) slider to read about 70,
and click Create. This setting causes the second light to dim

To speed test renderings, temporarily turn off extra lights that may
not be needed.

Next, create a camera to view the text in perspective. Follow these
steps:

1. Select Cameras and then Create from the command column.

To create a camera, you must click to set the location of the cam-
era and then click again to set the location of the target (the loca-
tion where the camera is pointed).

2. Move the cursor to the middle of the lower left quadrant of the
Top viewport and click once.

3. Move the cursor to the middle of the 3D text (see fig. 20.11). No-
tice that a line extends from the camera to the cursor. Click once
when the cursor is in the middle of the 3D text.

The Camera Definition dialog box appears with many settings that
you can adjust.

4. For this example, accept the default settings by clicking Create
(see fig. 20.12).

5. Click the User viewport to make it the current viewport.

6. Change the User viewport to the Camera viewport by pressing c
on the keyboard. You should see the 3D text in perspective (see
fig. 20.13).

Before rendering the CameraO1 viewport, you need to choose a 
material so the text will look like shiny chrome. Follow these steps:

1. Select Surface, Material, and then Choose to display the Material
Selector dialog box. Materials are grouped into files known as
Material Libraries. Libraries can contain many different materials.
Usually, you create a library for each project on which you work.
When you use the Material Choose command, you see a dialog
box that contains a list of the materials in the current library (see
fig. 20.14).

2. Use the slider bar to move down to the chrome materials and
double-click CHROME VALLEY.

3. To apply the CHROME VALLEY material to the text, choose Sur-
face, Material, Assign, and then Object.

4. Click the 3D text. 3DS asks Assign "Chrome Valley" to Logo?
Click OK to accept this action.

5. Choose Renderer and then Render; click the CameraO1 viewport
to display the Render Still Image dialog box. Alt-R chooses the
same command for the currently active viewport. To accept the
default settings, click the Render button (see fig. 20.15).

In a few seconds, the Rendering in Progress status box appears, dis-
playing information about the status of the rendering (see fig. 20.16).
To switch the view to the image being rendered, press the space bar;
to cancel the rendering, press Esc.

If you are rendering to a VGA display, the program uses the
palette that was last generated. The colors may be incorrect
until the rendering is finished and 3DS has computed a new
optimized palette. Wait for the VGA image to be redrawn in
the new palette before pressing Esc.

When the rendering is finished, the screen switches to the image dis-
play and shows the rendered image (see fig. 20.17). At this point, you
can press Esc to return to the 3D Editor.


12.  A I R P O R T  P R E S E N T A T I O N

Without question, the Greater Rockford 
Airport Authority's Regional 
Transportation Study contained elements that
were ideal for computer modeling and
rendering: a greatly expanding airport 
infrastructure, a new terminal, and a high-
speed rail connection to Chicago's
O'Hare Airport. The challenge was not
just getting this information into 3D-mesh 
form, but to create output in a
medium suitable for four-color printing.
The steps involved in taking a series of
AutoCAD drawings for the airport project
and bringing them into 3D Studio for
high-resolution rendering included 
organizing the associated AutoCAD and 3D
Studio files, preparing the AutoCAD
drawings for import into 3D Studio, 
modifying the final model in 3D Studio's 
editor, and rendering the image suitable for
printing.

FILE ORGANIZATION
One of the keys to successfully producing 
this project was the organization
of the various files shuffled between
AutoCAD and 3D Studio. Since this project 
was active, it was important to maintain 
the ability to make design changes
consistently and easily across all files. A
technique was developed based on three
primary files and any number of component 
or sub-files. The original plans for
the airport and its expansion were created 
by the airport-engineering firm of
Crawford, Murphy and Tilly Inc. of
Springfield, Ill. The original plan, as
shown in Figure 1, was drawn with AutoCAD 
Release 10 and was 1.2Mb in size.
rhe information in this engineering
drawing was distilled into a less complicated 
presentation drawing that served
as the working design drawing and a
source for presentation plots. As the project 
evolved, design and engineering
changes, made by the architecture firm of
Larson & Darby Inc. of Rockford, Ill.,
would appear in this file.
The entities that formed the basic
geometry of the model were then
Wblocked out of the presentation 
drawing to create a working 3D drawing. In
this drawing, 2D entities were converted
into 3D entities, along with other modifications, 
then written to a DXF file to create 
the final 3D Studio model. Some components 
of this project were determined
to be sufficiently complex to deserve
their own files. These components were
Wblocked out of the presentation drawing 
and became component files that
would eventually be merged back into
the final 3D Studio model. The entire
document flow is shown in Figure 2.
This file organization allowed a rapid
response to design changes without
wasting effort on repositioning changed
objects. For example, if a runway is
moved in the presentation drawing the
changed entities are Wblocked out and
inserted into the working 3D drawing,
where they are cleaned up for use with 3D
Studio. These new 3D entities are then
written to a DXF file and imported into
3D Studio using the Merge option. The
newly imported runway is then assigned
the material and mapping coordinates
acquired from the old runway, which, in
turn, is deleted after the merge is complete. 
It is useful to note that no repositioning 
of objects is necessary as long as
the blocks are created and inserted at
(0,0,0). Using this technique preserves
the relationship between the presenta-
tion drawing, the 3D drawing, and the 3D
rendering.

THE AUTOCAD DRAWINGS
The first step was to transform the
presentation drawing into the 3D working 
drawing. To begin the process, all text,
line-types, and other non-real-world 
features were stripped out. Each plan 
element that would become an individual
object in the 3D Studio model (such as
runways, rivers, roads, parking, aprons,
and building plans) was placed on its
own layer. Layer organization was fine
tuned according to the eventual material
and mapping requirements of each ob-
ject. The lines and arcs composing each
piece were then transformed into closed
Plines using either the Pedit/Join command 
or a custom AutoLISP routine that
connected selected entities into Plines.
These Plines were assigned slightly different 
elevations so that higher-elevated
Plines would obscure lower Plines. 
Continuing with this overlay approach, a
large 3DFace was placed at the bottom
(ELEV = 0) as the world followed by the
Pline representing the surrounding land
slightly higher, rivers and airport property 
higher yet, followed by roads and airfields, 
runway text and parking, and finally, 
buildings. This spatial-overlay
technique allows rendering objects on
top of each other without having to precisely 
model the edges between the objects. 
It saved modeling and rendering
time because it required far fewer faces.
A key in transforming an AutoCAD
drawing into a renderable mesh lies in
understanding how the original entities
will be interpreted by 3D Studio. In short,
all 3D entities (Polymeshes, 3DFaces, and
extruded Line and Arc segments) will be
converted into groups of triangular faces.
Adjacent faces that share two vertices will
be netted together to form elements. All
elements that were organized on the
same AutoCAD Layer will now belong to
the same 3D Studio object. In addition,
all closed Plines and Circles will be
capped (replaced with a mesh defined by
the entity's perimeter) while extruded 
entities will have their edges converted to
vertical face meshes and their tops and
bottoms capped. In our opinion, 3D Studio's 
ability to cap objects is an understated 
modeling feature and a big time-saving tool.
Before going any further, the project
goals needed to be defined, because the
accuracy and detail of the model now had
to be balanced with the final output. The
determination of an object's accuracy is
two fold: how close will the observer get
to any particular object, and what will be
the final output resolution. (Note: when
creating animations, the speed at which
an object will pass across the screen 
creates yet a third accuracy factor.) When
building a model, an object's required detail 
and accuracy is determined by its
final size on output. In video, the screen
resolution is typically 512-by-486 pixels,
where a large object might be only 400
pixels wide, compared to a photographic
print at a resolution of 3,072 by 2,048 
pixels where the same object would be 2,000
pixels wide, requiring much more detail.
It was decided that there would be a
number of aerial shots showing the 
relationships between existing and proposed
airport areas and at least one dramatic
close up of the proposed terminal with
the high-speed-rail train and overhead
aircraft visible. The published document
would require one or more full-cover renderings 
at 8 by 11 inches. After a discussion 
with the printing service, the resolution 
of these renderings was set at
2,125-by-2,750 pixels making it possible
to define a detail hierarchy. The terminal
and train connections were the most 
detailed and the remaining elements fell off
in detail based on their proximity to the
terminal. Due to the high resolution of
the image, most objects had to be fairly
detailed to be convincing.
An object's detail translates into using
more vertices and faces, which correlates
directly into the memory and time required 
to generate the renderings. The
complexity of all objects had to be held to
the absolute minimum to maintain manageable 
rendering times. In a model of
this size, a major concern was the vast
amount of Arc entities found within the
roads and parking areas. arc entities do
not directly exist within 3D Studio; they
become segmented when the DXF file is
imported. The detail or coarseness of the
arc segments is currently fixed at one 
segment for every nine to 10 degrees of arc.
For this project, that amount of detail was
excessive for the multitude of distant
road arcs, yet too coarse for arcs very near
the terminal, the restriction was overcome 
by using the Yost Group's
DXF3DS.EXE program--a conversion
utility available on CompuServe that 
improves DXF import for 3D Studio Release
1.0. It has the added feature of an "arc
smoothness" parameter, which allows
you to define the angle of arc per segment. 
Because we needed to constantly
evaluate the coarseness of arcs within the
AutoCAD model, the choice for this project 
was to use a custom AutoLISP routine 
that allowed the transformation of
arc entities into segments with differing
levels of coarseness within the same file.
All Arcs were copied to a separate arc
layer. The original arcs were converted to
Plines according to the following hierarchy: 
30-degree arc segments were used at
extreme distances and three-to-five-degree 
arc segments were used near the terminal, 
with other arcs falling in between.
One advantage of this method was that
similar arcs, such as a grass edge over or
parallel to a curb edge, were assured of
aligning in the 3D Studio file.
After the Plines were put on the correct 
elevation, some were given thicknesses. 
This step was applied sparingly
because the simple action of extruding a
rectangular Pline increased it from a 
two-face four-vertex object to a 12-face eight-
vertex object. All building outlines were
given somewhat random but representative 
thicknesses using the CHPROP command. 
The existing terminal building
was given extra detail because it was a
known object and attracted the viewer's
attention. The proposed terminal was
deemed a project unto itself and was
Wblocked out to be worked on as a 
separate drawing.

THE 3D STUDIO MODEL
To begin the 3D Studio mesh, all of the
landscape perimeters were exported out
of the 3D drawing with the DXFOUT
command using the selected entities
method and, to be safe, 16 decimal places
of accuracy. Once in 3D Studio, the DXF
file was imported by layer with the
smoothing angle set to zero. Work then
began on analyzing each object's normals, 
smoothing groups, and name. 3D
Studio unifies normals on an element
and then an object level. Closed polygons
are assured of being unified meshes with
normals determined by whether they are
an extruded 3D entity or a flat 2D entity.
Extruded polygons always come in with
normals facing outward, somewhat like a
wrapped present. Flat polygons are 
another matter, because they have no depth
and we have no z-axis centroid from
which to calculate normals. These flat
polygons import with seemingly arbitrary, 
yet unified, normals. The option 
exists to ignore this condition by rendering
objects as two-sided. Unfortunately, it
doubles the amount of faces calculated
during rendering and that overhead is
unacceptable on a large model. With all
objects hidden, each flat object was 
unhidden in turn. The Display/Backface
command made it easy to see which 
elements needed to be flipped either one by
one with Surface/Normals/Element Flip
or on a selection-set basis with
Surface/Normals/Face Flip. It was also a
good time to assign the Object's initial
material and, to show that it had been
modified, change its name by deleting
the "01" suffix.

Take a closer look at how 3D Studio 
names objects that are brought in by
layer from AutoCAD. 3D Studio takes the
first eight letters of the layer name alld
appends them with "01." If it results in
name duplication, such as having layers
named TERMINAL-WINDOWS and 
TERMINAL-DOORS, those layers are 
appended with "01" and "02" to become
TERMINAL01 and TERMINAL02. Be
careful when naming layers for import
into 3D Studio because it will truncate
any trailing numerals used in a layer
name (for example, layers named
ROUTE55 and ROUTE251 become 
Objects ROUTE01, and ROUTE02).
So far, the model had a very manageable 
12,080 faces. The existing terminal
was modeled from the original construction 
documents and inserted into the 3D
drawing for placement and site integration. 
Finished independently in 3D Studio, 
the existing terminal (550 faces) was
then merged into the master 3D Studio
model. After several revisions, the 
proposed terminal was created independently 
on its own local site in AutoCAD,
then finished as a separate 3D Studio
Mesh (3,800 faces). Since neither model
required the detail of cut-out openings,
both models were created primarily with
extruded and offset Plines. Care was
taken to organize the layers according to
building material (copings, glass, 
skylight, high wall, low wall, pavement, and
so on) to assist with material assignments
A small modeling problem presented
itself. The train route was designed to
emerge from the terminal through a terraced 
valley which cut into the already
modeled grass entity. A Pline of this
cutout was prepared on a separate Layer
and subtracted from the grass object
using the Create/Object/Boolean/Subtract 
operation in 3D Studio.
To emphasize the scale of the project,
it was desired that one plane be included
for each potential gate. Since this requirement 
amounted to 45 planes, it became
essential to create a very simple, yet 
believable aircraft model. The plane was 
defined with eight-sided cross sections for
the engines and fuselage and six-sided
airfoil sections for the wings and tail. 
Relying on 3D Studio's smoothing function
for streamlining, these defining Plines
were connected with Rulesurfs using a
minimum number of 3D faces. The result
was a 727 model with an economical 362
faces. Specific layers were created for the
wing, tail, fuselage, engine and intake 
entities so that materials and smoothing
groups could be applied globally to the
objects once they were in 3D Studio. The
original 2D airplane block in the 3D
drawing was then redefined with the new
3D airplane drawing, creating an air force
properly placed and ready for import into
3D Studio. The train engine and its cars
were modeled in a similar fashion.
These 45 planes added 16,290 faces to
the assembled model, which now 
contained approximately 35,000 faces, the
upper limit of a manageable model on
most systems. To give a sense of greater
detail and complexity, texture and bump
maps were placed on nearly every object.
This approach also served to break-up
the color banding that can occur with
some printing processes or if the image
was rendered to a file format with less
than 24-bit color.

MAPPING
Except for special maps created and
used for the train and close-up planes,
the purpose of mapping in the remainder
of the model was primarily for artistic 
effect. The correct scale of the map was 
determined by the visual effect desired and
not by any real-world correlation. This
procedure used somewhat of an "apply,
render, and see" method for the first
maps until requirements of a particular
scene were discovered (for example, the
grass maps were scaled with mapping
icons at 300 to 500 feet, when they would
normally have been four to eight feet in
an architectural model). This approach
even holds true for the proposed 
terminal building, because the camera is 
nearly 1/2 mile away in the closest scene. If
the terminal had been properly mapped
to represent stone, stucco, and so on, the
result would have been a smooth, even,
cartoon-like tone from this distance.
Mapping was further used to distinguish 
the airport's property from that of
its surroundings. An abstract rectangular
pattern was created to simulate a typical
grid of streets and land. This pattern was
applied as a texture map and a modified
version was applied as a bump map. To
achieve the proper look with a minimum
of repetition, both maps required 1,024-
by-768-pixel-resolution GIF images.
These large GIFs made the land-grid map
the most costly material, requiring 4.7Mb
RAM (1,024 by 768 pixels by three bytes
by two maps) at rendering time.
The terminal close-up was the only
scene where we felt that shadows were
necessary to provide extra depth and life
to the composition. Although multiple
spotlights were used, only one was shadow 
casting. This spotlight, placed directly
over the proposed terminal, was given a
1,000-line shadow map. Because many of
the objects in this model floated above
one another (such as the pavement
above the grass) it was important to turn
off their shadow-casting attributes to
avoid calling attention to this technique.
The final mapping challenge was the
creation of a realistic sky background.
The sky map used was a 756-by-486-pixel
TGA file provided on 3D Studio's CD-ROM. 
Ideally, a background map should
be the exact size of the final resolution (so
that no extra time or memory is required
at rendering time), or tile seamlessly so
that its size is not an issue. In the close-
up, the sky had to match the output resolution 
in one direction and the distance
from the model's horizon to the top of the
image in the other. With the final resolution 
set, the background TGA file was
rescaled to 2,125 by 850 pixels, saved, and
used with the Tile/Background option.

FINAL OUTPUT
With the model complete, the scenes
were rendered to a TGA file at a resolution
of 2,125 by 2,750 pixels. When rendering
images that are not intended for screen
display, rendering times can be reduced
by configuring for a NULL video driver.
Select the NULL driver from the 
Render/Setup/Configure dialog. To maintain
consistency and ultimate control of the
final covers, it was decided that titles and
other text would be composited with the
rendered images prior to being sent to
the printer. Text could have been added
to this image in many ways with various
image editing and desktop-publishing
programs, but because we did not have
one available, 3D Studio was used. One
advantage that 3D Studio provided was
that few image-editing programs have an
anti-aliasing feature as good as the 3D
Studio renderer. The text was created in
the Shaper using Type- 1 PostScript fonts,
Lofted with 10 steps per arc and optimization 
on, then positioned within the
3D Editor. Since the text needed to be flat,
its side and rear faces were deleted. 
Although it is possible to import shapes 
directly from the Shaper and bypass these
last two steps, the optimization feature is
not available for this task. Using 
optimization in the Lofter greatly reduced the
number of faces used by the text and was
well worth the effort.
The key to placing the text properly
was to use 3D Studio's View/See 
Background function to check the text 
location against the rendered image of the
airport. The three-tone proxy image 
created by this command is extremely useful
but can be very time consuming if the
background image selected is large. A
stand-in background was created by 
scaling the high-resolution image of the 
airport down to 200 by 256 pixels. This
image loaded in a matter of seconds as
opposed to the nearly 30 minutes 
required to load the full-color 
high-resolution image. Prior to rendering, 
View/See Background was turned off and the
stand-in background bitmap was replaced 
with the high-resolution image.
Figure 2 shows the image with composited text.

PRINTING
Since an image's final print size 
determines the required resolution, it was 
important to have the client decide early to
print to 8-by-11-inch covers. Also, the
amount of detail needed for final output
depends on the hard-copy process being
used, the image "crispness" desired, and
the printing technology being employed.
Whenever an image is to be printed using
a four-color offset process, it is vitally 
important to have an early discussion with
the printing service. Printers may have
slightly different screening methods and
preferred rules for image resolution. 
Discussions with the printer determined
that this project would require a 
resolution of 250 dpi to achieve the desired
image quality. At 250 dpi, the 8-by-11-
inch covers required TGA or TIFF files at
a resolution of 2,125 by 2,750 pixels.
Again, verify the preferred file format
with the printer.
Transferring the image files becomes
the final and possibly most difficult step.
With image files ranging from 15Mb to
17.5Mb for this project, choosing a 
storage medium that is acceptable to the
printing service becomes important. The
most common mass-storage devices are
removable cartridges such as Syquest or
Rernoulli drives. Other choices include
optical drives (fairly rare but growing in
popularity), tape (slow), or splitting the
file across multiple diskettes using 
software (the slowest). If creating a high-resolution 
image for printing is a rare occurrence 
for you, then the multiple-floppy
method of transfer would probably be
adequate. If you plan to produce many
images for print, you should invest in the
same storage technology used by the 
majority of printers in you area.

NEW HEIGHTS
As with most undertakings, producing
high-resolution renderings for print 
requires organization and planning. Many
people wrongly assume that transferring
their AutoCAD drawings to 3D Studio 
involves some kind of wizardry or extreme
effort. The reality is that good file 
organization simplifies the work and maintains
the relationship between the files so that
they stay up-to-date during the course of
an evolving project. The goal in combining 
AutoCAD with 3D Studio is to exploit
the strengths of each program while 
sidestepping the weaknesses. This article
should give you some ideas as to how you
can set up your own projects in a manner
that will minimize the time and effort 
required for rendering presentations with
3D Studio. While it might seem to be 
considerable work to produce these images,
the response that you get from high-quality 
renderings can make the effort
quite worthwhile.



END
