home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!usenet.coe.montana.edu!news.u.washington.edu!milton.u.washington.edu!hlab
- From: georgef@endeavor18.wv.tek.com (George Forman = GHF)
- Newsgroups: sci.virtual-worlds
- Subject: TECH: OIFF an object interchange file format
- Message-ID: <1992Aug21.232127.29246@u.washington.edu>
- Date: 21 Aug 92 22:18:02 GMT
- Sender: news@u.washington.edu (USENET News System)
- Organization: University of Washington
- Lines: 52
- Approved: cyberoid@milton.u.washington.edu
- Originator: hlab@milton.u.washington.edu
-
-
-
- I have a few suggestions for OIFF:
-
- 1. You might include a version # in the header. This will probably
- save grief later.
-
- 2. In the 2-D world, people have found it very useful to include a
- bounding box in the Postscript files. This saves scanning the whole
- thing to find out if it's in the field of vision or whatever. Also
- aids scaling-- suppose I want to put Jim's world in a shoebox on my
- shelf.
-
- 3. Perhaps some notion of subroutines could be incorporated to reduce
- data size. A 1 page description in a slightly higher level graphics
- language can produce a scene filled with 1000s of polygons. The OIFF
- files will blow up geometrically-- each of the 40 houses each have 15
- boxes which each have 6 sides. A very low-level OIFF will have to
- represent an awful lot of data, but a slightly higher-level OIFF can
- be much smaller. I suggest such simple languages as found in computer
- graphics course textbooks.
-
- 4. The file could specify the 6 light reflecting characteristics of a
- flat surface (RGB reflected & RGB diffuse) & let the *user's* display
- program decide whether to just quickly use a flat color model (just
- using the first 3 numbers), or to produce a beautifully lighted model
- using all 6. To save space, suggest that the OIFF define a color map
- which the polygons then use by reference. In the future, this color
- map could become sophisticated w/ texture & tiling information. You
- certainly wouldn't want each polygon to have to specify all this
- information explicitly each time-- takes a lot of polygons of the same
- color to make a blue tree. These are probably frills for the second
- version.
-
- 5. Suggest making the file more readable & printable. Right now it
- seems extremely vertical-- could waste a lot of printer paper.
- Wouldn't hurt to use a tab character between components of a vertex &
- newline after each vertex, or commas and tabs.
-
- 6. I'm not so certain that things should be broken up according to the
- 5 senses: SOUN,VISI,HAPT,OLFA,TAST. Yes, it's stable over centuries,
- but I'd think it'd be better to break things up according to the
- virtual devices on which they are to be displayed. The 3-D world
- device (probably HUD, but perhaps not.), a much higher quality 2-D
- device (for text, explanations, detailed menus, high-resolution
- pictures etc. which won't work well on HUD), general room lighting
- device (RGB diffuse light? Strobe?), pack vibrator, floor vibrator,
- room heater...?
-
- Good luck,
-
- GHF
-