"Light Makes Right"
January 23, 1995
Volume 8, Number 1
All contents are copyright (c) 1994,1995, all rights reserved by the individual authors
Archive locations: anonymous FTP at
wuarchive.wustl.edu:/graphics/graphics/RTNews, and many others.
You may also want to check out the Ray Tracing News issue guide and the ray tracing FAQ.
Well, I was going to talk about the great research presented at SIGGRAPH this year, but if you went then you already know, if you didn't then you should consider getting the Proceedings. For some reason I found more of the papers directly useful to me this year than in the past few years.
So, instead, I'll talk about something totally mundane and banal, and yet totally amazing: Saturday morning cartoons. Specifically, "Reboot" on ABC. It's an entirely 3D computer graphics rendered cartoon. The character animation is not fantastic, there are no shadows, and one can make a whole bunch of other quibbles, but who cares?! This show makes it clear how far computer graphics has come, that it's used for a weekly cartoon show. It's fun to watch and figure out how effects were done (and there are some nice animated textures and whatnot). Some of the models are wonderfully elaborate - I watch it and say to myself "hours and hours of modeling". The show even answered a question I've had: what does it look like if you move something through a static turbulence function? That is, you apply Perlin noise and get a rendering of a marble ball; now, what happens if the ball moves but the noise function does not move along with it? I figured a "swimming through marble" effect would happen. They use this effect ("it's a feature, not a bug") for an animated worm-like thing that crawls around on the main bad guy. Anyway, how can you dislike a show where the old, wise character is named Phong?
The other banal-yet-cool thing I've noticed is "Magic Carpet" for the Intel PC. There's a demo version at ftp.uml.edu and ftp.uwp.edu in msdos/games/demos/magic.zip. Pretty amazing stuff: you fly over textured 3D terrain and can bank and change altitude, sorta like DOOM without the "fixed head" perspective. Definitely worth checking out, even if you don't like games. The real game includes output for 3D glasses (well, I can see 3D with my regular glasses, but you know what I mean) and also random dot stereogram output in realtime. Amusing!
As usual, there's a big backlog of stuff I haven't put in this issue, but I thought I'd get this one out the door. Still, there should be more than enough to keep you busy. The announcement of a RenderMan implementation from a non-Pixar source (and shareware at that) is probably the most amazing thing in this issue, and there are lots of other interesting sites and resources. More when I find the time...
back to contents
Stefan Didak (email@example.com) Doelstraat 147 3011 AJ Rotterdam The Netherlands +31 10-412.57.19
One line description: 3D Studio Die Hard User (some call it expert :)
Title : managing director Company : Animagic & Animagic Development Actually : doing computer graphics, animations, and development. (i.e. having fun and getting paid along the way :) Using : PC Platform - 3D Studio, proprietary tools. SGI Platform - Alias PowerAnimator, proprietary tools.
Almost anything that is related to three-dimensional computer graphics, animations, raytracing, commercial productions, broadcast design, high-end digital video and commercial production, software development (graphics).
Autodesk's 3D Studio and development of IPAS routines for 3D Studio.
So what about me ?:
I'm both a programmer and computer animator that has been able to combine the best of both worlds. It has been five years since I formed my own business called Animagic specialized in 3D computer graphics for broadcast design. Alongside this, development of software also played a major role. Currently the development and production departments are separated due to the confusion created by our extensive list of services. The development department is currently developing IPAS routines for 3D Studio and in-house proprietary tools for both the PC and the Silicon Graphics platforms. It has been five wonderful years, and with the current state of progress the next five years will be even better in terms of new development.
I started with computer graphics back in late 1985, when the Commodore Amiga computers were first introduced. This is where I got my first taste for computer graphics and animations with a package called Videoscape from Aegis, designed by Allen Hastings (currently the developer of Lightwave). By today's standards this package would be a real old-timer where 3D geometry had to be entered by hand (coordinates and polygons) in ASCII. So back in those days creating a twenty second animation was about a month's work. Today of course, this can be done in just hours :) I've been able to follow the evolution in the computer graphics industry from that point on, and quickly started to do some experimental raytracing with the Amiga port of DBW Render. That's where the fascination of raytracing got the best of me :) I've used those good old Amigas for a few years and about four years ago switched to bigger and better platforms and software (and never regretted it).
back to contents
Hadley Woolf (firstname.lastname@example.org) wrote:
>I am looking for a good RAY TRACING program
>Preferably WINDOWS for ease of use, but otherwise DOS ok.
>I hear there are good shareware/freeware of this type of program available.
>Please reply if convenient by phone (voice) 011 648 6629 or otherwise
Please take this to alt.personals.fetish.ray-tracers. And for the record, this radiosity stuff is sick sick sick.
back to contents
VolPack - fast, high-quality volume rendering library from Lacroute/Levoy
Here is the FTP site info:
graphics.stanford.edu [184.108.40.206]: pub/volpack/volpack.1.0b2.tar.Z - current version of the *VolPack* volume rendering library. Includes source, example programs, user's manual and man pages. Data sets and a Tcl-based application are also available. Phil Lacroute (email@example.com)
Here is the WWW site info:
http://www-graphics.stanford.edu/software/volpack - VolPack
[The paper related to this new, much faster volume rendering technique is in SIGGRAPH '94, p. 451. -EAH]
- Phil Lacroute (lacroute@weevil.Stanford.EDU)
The book _Texturing and Modeling: A Procedural Approach_, edited by David Ebert, from Academic Press is now out (see RTNv7n3 for a review). The other co-authors are Ken Musgrave, Darwyn Peachey, Ken Perlin, and Steve Worley. They have graciously provided the book's disk's code at the FTP site archive.cs.umbc.edu:/texture. Even if you have the disk you might want to download this code, as all bugfixes and additions reside here.
L Systems Trees, images and program
For those of you out there that were interested in my tree pictures, they are available on WWW now. The URL is :
There are only the pictures there at the moment, but I intend to write some stuff about L Systems etc. so check it out in a while (a few weeks). A copy of the executable may be there soon as well. [I recently grabbed the executable, though I haven't had a chance to unzip it yet. His images are great. -EAH]
- Phil Drinkwater (P.J.Drinkwater1@student.lut.ac.uk)
The Graphics Gems I-IV code distribution has had a few minor bugfixes and additions. Personally, I've diddled a bit with, yes, my favorite, the point in polygon test and made a new routine which is a whopping 15 0.000000aster for triangles. Hey, it keeps me off the streets. The distribution is available by FTP from princeton.edu:/pub/Graphics (though this may move to Stanford soon, since Craig Kolb is now there).
Computer Graphics Labs WWW Site
I have finally joined the 1990s and set up a www home page. It has several images and postscript copies of some of my papers. It also has a listing of computer graphics labs-- please send me yours to include!
[Also, ftp://ray.graphics.cornell.edu/ has stuff, for those without Mosaic]
Several people have asked for copies of some of my hard-to-find papers, so to exploit parallelism I have put them in an ftp site.
site: ftp.cs.indiana.edu (220.127.116.11) directory pub/shirley
- Pete Shirley (firstname.lastname@example.org)
I have just finished installing a rough WWW document structure for the BSP Tree FAQ. If you are interested in keeping track of my progress, you can look here. The FAQ is just getting underway, and may take a while, so don't expect a whole lot right now...
Thanks to everybody who has encouraged me or provided assistance!
- Bretton Wade (email@example.com)
L-system Educational Package
Anne Burns (firstname.lastname@example.org) has written an L-System set of tutorials which run on an Intel PC for teaching her college class. She has made them available via FTP on cgcr.gsu.edu, with login and password "cgcr". The file is /pub/IBM_PC/flora/aburns.zip . I have looked at it briefly, and it looks pretty interesting: there are programs to grow various 2D L-system plants while also teaching about the algorithms used and pointing to references for more information. Also, even old PC's look to be able to run this software.
It looks like the time has come to throw in the towel.
I have a "version 4.1" floating around, waiting for me to get to some documentation-hacking and general cleaning up. But between dissertation research/writing, moving to the opposite coast, and attempting to have something resembling a Life, it looks unlikely that I'll be able to wrap things up before the end of the year.
As such, it's probably best if I simply stopped work on the freely-available version of rayshade for the time being, and let whoever wants to release a rayshade-based "netshade" (or whatever) go to it. If I am not mistaken, a group of interested people have already set up a mailing list (David? Nick?) to discuss such matters -- I'll let the "TraceCore" folks speak up for themselves on this.
I'll keep the mailing list going until the "netshade" future looks a bit more firm.
Thanks to everybody who has contributed to rayshade over the years, and thanks for your patience during the last couple of years in particular.
- Craig Kolb (email@example.com, though he's now at Stanford)
Since Craig bowed out there's been confusion on what the latest version is or should be. Here's one of the latest:
This is an unofficial release of Rayshade, called 18.104.22.168 [was 4.1] because of the new features included. I expect there will be problems and such, and welcome patches and suggestions.
I have placed the file on ftp://tbird.cc.iastate.edu/pub/rayshade/rayshade-22.214.171.124.tar.gz so grab it and make comments. There is a lot of cleaning up to do yet, which I hope to do over the next few weeks between working and preparing for classes this semester.
Comments, bug fixes, suggestions, flames, and hard drives welcome. :)
- Michael Graff (firstname.lastname@example.org)
Rayshade Flame-Texture Enhancement Available
This is an enhancement to rayshade version 4.0 so it can generate animated fire to further enhance the realism of scenes.
You can get the enhancement via anonymous FTP from iamsun.unibe.ch (126.96.36.199): /Graphics/flame.tar.gz and /Graphics/flame.examples.tar
This work is based on the Article of Ken Perlin (Ken Perlin, An Image Synthesizer, p. 285-296, SIGGRAPH '85 Number 3, July 1985) and the book of A. and M. Watt (A. Watt and M. Watt, Advanced Animation and Rendering Techniques, Addison-Wesley, 1992) where all the principles used here are described. To explain my work in few words, it can be said, that the flame is generated by simulating and animating turbulence.
Feel free to contact me for questions and suggestions...
- Reto Mani (email@example.com)
Latest Amiga Rayshade
I just uploaded rayshade 188.8.131.52 (source + 68020/68881 binaries + several utilities) on wuarchive. It should be on an aminet site near you soon, in gfx/3d/rayshade-184.108.40.206.lha (e.g., wuarchive.wustl.edu:/pub/aminet/...).
Kriton Kyrimis (kriton!kyrimistheseas.ntua.gr@Princeton.EDU)
More Rayshade Web Sites
http://www-graphics.stanford.edu/~cek/rayshade/ - The new "official" site (bop up a level and see a picture of Craig)
http://web.mit.edu/afs/athena/activity/c/cgs/lib/html/rayshade/guide/guide.html - Rayshade user's guide and reference manual, by Craig Kolb
http://web.mit.edu/afs/athena/activity/c/cgs/lib/rayshade/surf/ - archive of various material definitions and other *.rh files
Bridget Testa (BrigittaT@aol.com) has an article in the November 1994 issue of Computer Graphics World about the Internet and computer graphics. There are some nice images made with POV, pointers to net resources (already dated), and quotes from comp.graphics luminaries like Steve Hollasch, John Grieggs, John Cristy/Christy, and some guy named Erich Haines.
Check out the August 1994 issue of Computer Graphics - the cover is an image done by one of my students. He is an undergraduate and this was his first & only graphics course. The students write their ray tracers from scratch. [The cover shows a comparison of a simple scene rendered with 3D Studio, POV-Ray, and Radiance; there is also a short article about the cover. -EAH]
- Scott Owen (matgso@ironduke.CS.Gsu.EDU), Georgia State University
Tcl-SIPP (tsipp) 3.1a available
Tcl-SIPP provides a 3D image specification and rendering toolkit for use with Tcl and Tk. It is based on SIPP, the SImple Polygon Processor, a library for creating 3-dimensional scenes and rendering them using a scan-line z-buffer algorithm. Tcl is an easy to use, powerful, interpretive programming language, that is designed to be imbeddable in applications. Tk is a X11 toolkit programmed in Tcl.
It is available from:
ftp.neosoft.com:/pub/tcl/distrib/tsipp3.1a.tar.gz or harbor.ecn.purdue.edu:/pub/tcl/extensions/tsipp3.1a.tar.gz
- Mark Diekhans (markd@sugar.NeoSoft.COM)
You can subscribe to the Wavelet Digest mailing list of the Industrial Mathematics Initiative, Department of Mathematics, University of South Carolina by writing firstname.lastname@example.org.
More information, archives and papers can be found by anonymous ftp to ftp.math.scarolina.edu, in the directories under /pub/wavelet and /pub/imi_93, and by connecting to our gopher server at gopher.math.scarolina.edu.
- Wim Sweldens, Bjorn Jawerth
As you should know, avalon.chinalake.navy.mil is a 3D model archive (which is like saying Mount Everest is a big hill). Its name will change soon to avalon.vislab.navy.mil, though the old name will still work for awhile.
The bad news is that Kubota's mirroring site ftp.kpc.com is no longer mirroring avalon. If you can offer your site as a mirror, let Francisco DeJesus, email@example.com, know.
The good news is that Avalon's resources continue to grow. There are working demos of Caligari's TrueSpace 3D Modeling program (look for file "tsdemo*.zip") and for Real3D (files "real3d*.zip", unzip them onto disks and load from there), a windows-based graphics/animation package. Real3d does inverse kinematics, spline-based modelling, collision detection, particles, etc. The site is also fast becoming a hot spot for 3D Studio users, with IPAS add-ons and other goodies becoming available.
Another Optical Ray Tracer
Jonathan Abourbih writes:
>I am looking for a program for my Macintosh (IIci) that will do light-ray
>tracing diagrams. i.e. I want to be able to enter focal lengths, diameters,
>lens thicknesses, distances between lenses etc. and it should draw the path
>of a light ray through the lens system.
In the last 6 months or so, a company called Stellar Software came out with a Mac version of their Beam Two (and Three, Four?) optical ray tracer programs. People I know at NASA-Ames who used the PC versions seemed happy with them. The Two, Three, etc refer to two-dim, three-dim etc optics placement and views.
Beam Two: $89 Beam Three: $289 Beam Four: $889
[as Bob Knowlden noted, what four-dim lighting could mean is anyone's guess... - EAH]
Stellar Software P.O. Box 10183 Berkeley, CA 94709 (510) 845-8405 voice (510) 845-2139 fax
- Amara Graps (firstname.lastname@example.org)
[see RTNv7n5 for other optical ray tracers. -EAH]
Ray Dream Designer Mailing List
List owner of raydream-l, an Internet mailing list for discussion & announcements of products from Ray Dream, Inc. For subscriptions, send a message to email@example.com with the following body text:
subscribe raydream-l FIRSTNAME LASTNAME
Your address will be picked up by the listserv software.
p.s. Ray Dream Designer 3.1 for PowerMac has shipped!
- John T. Chapman (firstname.lastname@example.org)
If you're interested in looking at DOOM-like game engines, check:
ftp.funet.fi: //pub/msdos/games/programming/3dlab103.zip or garbo.uwasa.fi://pc/turbopas/3dlab103.zip
- Kimmo Fredriksson (kfredrik@cc.Helsinki.FI)
There is a cooperative InterNet effort (headed by Chris Laurel) to create a sorta Doom-like 3d engine called "wt" ("what's that" -never capitalized).
The main ftp site for it is magoo.uwsuper.edu and the mailing list subscription is email@example.com. Mail to the group is firstname.lastname@example.org.
The current version available (I think) is wt 0.05...
Tiny Review of _Ray Tracing Worlds with POV-Ray_ [v2.2] (Waite Group book)
I have had it for a couple of weeks now. From what I have seen so far, it is a good book to use for a tutorial sort of thing for Moray, and it does go into some detail with regard to editing povray source, but I'm not so sure that I would recommend it strictly as a reference guide for povray. As a reference source for povray, it comes nowhere near as complete as the povray manual.
- Dan Kressin (email@example.com)
Latest version of XAERO is v1.5.1 [see RTNv7n5 for info on this package]
- Andreas Ziegler (firstname.lastname@example.org)
For issues of this electronic magazine [mentioned in RTNv7n1], check out:
Mosaic/WWW users: http://barchetta.stu.rpi.edu
FTP users: barchetta.stu.rpi.edu:/pub/gfx-news/current
If your name server has difficulty resolving our address, the IP # is 220.127.116.11 Also, if the FTP server is badly lagged or not responding, please let me know and I'll provide an alternate server to use.
[I personally couldn't connect when I tried, but who knows? - EAH]
- Eric Hsiao (email@example.com)
I've had a couple people express interest in a compilation of POV I made on a RS6000 (model 250, i think). If anybody wants it, ftp to: ftp.povray.org The file, povray.rs6, and an accompanying text file are in the "pub/povray/incoming/misc" directory. No guarantees, but I hope it helps!
-Michael Sanchez (firstname.lastname@example.org)
POV related WWW sites:
http://www.cm.cf.ac.uk:/Ray.Tracing/ [excellent for all ray tracing! - EAH]
[this last one did not work for me - EAH]
- Albert Goodman (email@example.com)
My BBS is carrying all back issues of RTN because nobody does it in Belgium (to my knowledge), so if a european guy (who can't use ftp :) ask you where he can get RTN issues, feel free to give him my number: Virtual Access BBS - Belgium +18.104.22.168.77
- David Delabassee (delabass@nic.INbe.net)
Major Announcement : New POV FTP Site
Despite the following, ftp.uwa.edu.au continues to be the official secondary POV-Ray FTP site until further notice.
A new FTP site has been set up for POV-Ray. It is ftp.povray.org ! Yes, our very own domain. And instead of a 50mB disk allocation, it currently has about 1.7 gigabytes of disk space available. Currently it has more or less what was (and still is) up at uniwa, but you can expect it to expand over time, particularly when the POV CD-ROM is published.
This machine is connected via a T1 to the Internet. Currently, it shares the same T1 as wcarchive.cdrom.com, so in reality you'll be sharing the link with up to 140 other people at peak load times. But the machine itself is more or less ours for the time being ; it is provided as a free community service by Walnut Creek CD-ROM and has no other FTP sites hosted right now. [This will probably change sooner or later.]
Please start using this site and do let me know of problems via email to firstname.lastname@example.org ; I've only just finished configuring the FTPD and it's possible there may be a few glitches here and there.
BTW, I'd like to thank Walnut Creek for offering us this free, no-strings- attached service. Get their catalogue by fingering email@example.com if you want more information. Also, many thanks go to Toivo Pedaste of the UWA for tolerating us for so long (:-).
[In reality, this is *the* site for POV Ray related stuff - the uwa site is no longer updated, near as I can tell. -EAH]
- Chris Cason (firstname.lastname@example.org)
Dear Radiance User,
This is another reminder that if you want to sign up for the unmoderated Radiance discussion list, send mail requesting subscription to:
Please, do not send administrative requests to the discussion list itself (email@example.com), as they will be immediately forwarded to the whole discussion group, which is a rather poor self-introduction.
Also, you can pick up the most recent as well as past digests by anonymous ftp from hobbes.lbl.gov (22.214.171.124) in the /pub/digest/ directory.
- Greg Ward (firstname.lastname@example.org)
back to contents
http://www.cm.cf.ac.uk:/Ray.Tracing/ - actually, an old site, but it bears repeating. Points to all sorts of ray tracing related things. A great place to start explorations. Contact: Ian 'The Grim' Grimstead (Ian.Grimstead@cm.cf.ac.uk)
http://mambo.ucsc.edu/psl/cg.html - A great site jump off spot for other computer graphics programs and sites.
http://www.graphics.cornell.edu/ - A good place as any to jump off and see what's going on in research at various computer graphics programs around the world.
http://www.mcs.com/~bcleach/www/illusions.html - "Illusions - a computer graphic/animation related magazine" covers 3D Studio, Lightwave, Alias & Wavefront as well as SGI in general. Contact: Brian Leach (email@example.com)
http://www.lightside.com/~dani/ - 3D Web, a site about the computer graphics industry. Includes resumes of computer graphics people interested in various sorts of work, pointers to various production houses, etc.
http://www.tgax.com - The Graphics Alternative BBS now has a Web homepage. Under construction, but has some nice bits to it. Contact: Adam Shiffman (firstname.lastname@example.org, email@example.com)
http://www.sgi.com/ - Silicon Graphics Inc. homepage, pretty, but annoying if
you have a slow link, so here are some direct addresses:
http://www.sgi.com/grafica/ - GraficaObscura, an eclectic collection of images, hints, and essays related to graphics. Contact: Paul Haeberli (firstname.lastname@example.org)
http://www.sgi.com/tech/openGL/openGL.html - OpenGL related
http://www.algorithm.com/~cwatkins/algor.html - Chris Watkins' company's homepage, also includes a number of pointers to other pages worth checking out. Give it a look.
http://metallica.prakinf.tu-ilmenau.de/GOOD.html - the "GOOD" raytracer and modeller and whatnot (see RTNv7n1 and RTNv7n3). Contact: Ekki Beier (Ekkehard.Beier@Prakinf.TU-Ilmenau.DE)
http://www.cs.tu-berlin.de/~lulli/xray/xray.html - the XRay ray tracer, a new one in pre-pre-release for MS DOS and Windows (though the code's portable to others). Contact: Christian Schnell (email@example.com)
http://www.eecs.uic.edu/~malic/ - various MPEG animations, including one on 3D morphing. Contact: Mihailo Alic (firstname.lastname@example.org)
http://archpropplan.auckland.ac.nz - For things related to the Radiance program: http://archpropplan.auckland.ac.nz/Graphics/radiance/radiance.html Contact: Paul Bourke (email@example.com)
http://siggraph.org/library/bibliography/bibliography.hmtl - WWW version of SIGGRAPH on-line bibliography. Contact: Stephen Spencer (firstname.lastname@example.org)
http://info.acm.org/~joyces/homepage.html - Shawn Joyce's homepage, a nice little jump-off point for various commercial renderers and other stuff.
http://www.petech.ac.za/~kon - yet another images site; interesting in that it's based in South Africa. Contact: Kon Wilms (email@example.com)
http://www.clbooks.com/ - Computer Literacy Books, who sell Ray Tracing Creations (second edition) & Ray Tracing Worlds, among other things. A nice little place to browse information on books out there.
back to contents
Here is a reprint of a recent posting...
Announcing MGF: Materials and Geometry Format
Here we have yet another proposal for a standard graphics scene representation, but with a few notable differences:
1) The material descriptions are physically-based and therefore usable for lighting simulation: radiosity and ray-tracing methods for global illumination. 2) The package includes an ANSI-C parser that makes writing a translator to any native format quick and painless. It also includes a nice library of objects and scenes culled from the Radiance distribution. 3) Use of the parser and library is free.
The language was designed by the author and some other lighting and computer graphics experts/enthusiasts. It includes what it needs to include plus a little bit. It is by no means the ultimate scene description, and future attempts to improve the format must be weighed against the benefits of keeping things simple. Also, enhancements to the standard will be made in such a way that they do not place new demands on the programmers who support it. This will be accomplished by updating the parser along with the standard, so each programmer need only support those entities s/he knows how to translate. The parser will translate the rest.
Some other basics about MGF:
o It is a compact, human-readable ASCII format o Its syntax permits it to be embedded in TCL (I think) o It is strictly boundary-representation (for now) o It is mostly polyhedral, but supports a few conics o Vertices may be named, may have normals, and may be shared o Colors and materials may also be named and put in libraries o It supports full-spectral colors and arbitrary basis functions o It supports one-sided and two-sided surfaces o It supports rigid-body transformations and instancing o Materials may reflect, transmit and emit light
To pick up the parser by anonymous ftp, look in the /www/mgf directory of hobbes.lbl.gov (126.96.36.199), or from WWW access the URL:
(By the way, we'd welcome links to other WWW sites, as we're isolated at the moment.)
Comments are welcome.
-Greg Ward Lighting Research Group Lawrence Berkeley Laboratory Berkeley, California GJWard@Lbl.Gov
back to contents
The toolkit consists of a full implementation of the RenderMan standard which supports ray tracing, radiosity, area light sources, texture and environment mapping, programmable shading in the RenderMan Shading Language, motion blur, automatic ray cast shadows, and other advanced features. The toolkit also contains quick RIB previewers (using GL or X11) to allow "pencil tests" of scenes and animations.
The BMRT is being distributed as shareware. Feel free to grab a copy for evaluation. If you like it and continue to use it, a "donation" is requested. In exchange, this makes you a registered user, which means that you will be informed of bugs and updates, get some support, etc. It's all outlined in the License file that comes with the distribution.
The toolkit is distributed as binaries for the following platforms: SGI running IRIX 4.05 or later, HP 9000 8xx/7xx running HP-UX, Sun SPARC, NeXTSTEP (HP, Motorola, and Intel). Other platforms may be supported if requested. Source code is not distributed, so please don't even bother asking for it. The toolkit comes with full documentation, several example scenes which illustrate the various nonstandard features, and several example shaders.
The BMRT is available via anonymous FTP: FTP to ftp.gwu.edu login as "anonymous", use your email address as the password cd /pub/graphics/BMRT get the appropriate file for your architecture.
For more details, feel free to contact me.
Larry adds later:
I've recently had a period of unexpected productivity, and have managed to fix all of the known limitations to the Shading Language compiler for my BMRT package. A number of other bug fixes and enhancements have been added recently. Updated binaries for SGI and HP are in the FTP site.
and this, excerpted from another note he posted:
The rendering methods of the two programs [Pixar's vs. BMRT] are very different (REYES vs. ray tracing/radiosity). Given this fact, I'd consider it very good any time BMRT was within an order of magnitude of prman. Some people have written to me with timings for particular scenes, and they seem to be saying that BMRT is about 1/4 the speed of prman. I think that's pretty good, but I'm not sure it proves much of anything. Nobody's given me timing data on "real" scenes, whatever that means.
I tend not to even try comparing the two programs in terms of speed, because I mostly think of them as having very different uses due to their different rendering techniques and capabilities. You can let prman run all day, and you still won't get a radiosity solution. On the other hand, you can't get true displacements with BMRT (or any ray tracer).
Errata on using Darwyn Peachey's textures from _Modeling and Texturing: A Procedural Approach_ (Ebert et al.) with the Blue Moon Rendering Tools.
Unfortunately, when the FTP site was set up [for the book], the examples from chapter 8 (Ken Musgrave's chapter) weren't included. We've since put together all the (working!) shaders described in this chapter and also sample RIB scene description files which demonstrate their use, and put them in the FTP area. The chapter 8 examples are specifically in the file musgrave_renderman.tar.Z. The other subdirectories contain examples from the other chapters. FTP from archive.cs.umbc.edu:/texture
A couple minor incompatibilities made an earlier version of rendrib choke on some of these shaders. It's since been fixed, so if you have trouble with any of the shaders that I say work, just get the newest copy of the BMRT.
All of the shaders from chapter 2 work fine, except for the following:
brickant.sl - has a problem due to use of the Du and Dv functions. In BMRT, these functions can only take s or t as parameters. Peachey uses them as follows (lines 34-35):
swidth = abs(Du(ss)*du) + abs(Dv(ss)*dv); twidth = abs(Du(tt)*du) + abs(Dv(tt)*dv);
The following change (which is perfectly equivalent) makes them work properly with both prman and BMRT:
swidth = abs(Du(s)*du) + abs(Dv(s)*dv) / BMWIDTH; twidth = abs(Du(t)*du) + abs(Dv(t)*dv) / BMHEIGHT;
bluemarb.sl - Peachey uses a function definition, which unfortunately is broken in BMRT. To make it BMRT-compatible, change this line in Peachey's code:
Ct = marble_color(marble);
to the following:
Ct = color spline(clamp(2*marble + .75, 0, 1), PALE_BLUE, PALE_BLUE, MEDIUM_BLUE, MEDIUM_BLUE, MEDIUM_BLUE, PALE_BLUE, PALE_BLUE, DARK_BLUE, DARK_BLUE, DARKER_BLUE, DARKER_BLUE, PALE_BLUE, DARKER_BLUE);
This is simply folding the function definition into the main routine. Also don't forget to delete the function definition itself.
perturb.sl - works, but note that rendrib takes TIFF files directly as texture maps, not the .tx files that prman takes.
back to contents
back to contents
So here's an overview of the directory structure and contents:
3ds, 40 megs total: 3D Studio files 3ds/3dsmesh, 22 megs of 3DS mesh files 3ds/ipas1 and ipas2, 1 meg, about 20-25 IPAS filters total 3ds/maps, 14 megs of texture GIFs (same as on the first Syndesis CD ROM, from what I can tell) 3ds/util, 1 meg of utilities for 3DS anim_utl, 3 megs total: various animation utilities, such as DTA, MPEG makers, etc anime, 103 megs total: not Japanimation, rather animations; about 120 FLC/FLI animations and other stuff audio, 4 megs total: various sound editing programs demos, 11 megs total: sundry graphics demos, like "Unreal", and Doom. djgpp, 4 megs, the GNU C++ compiler, v1.11 dos_view, 4 megs, various dos viewers and image processors fractal, 11 megs, various fractal programs, including Fractinit, Lparser, and Mandy f_prot, 1 meg, popular anti-virus program gfxlib, 4 megs, graphics libraries with source, including SVGAKIT and TGE (The Graphics Engine) gfxsrc, 6 megs, graphics source code, including Graphics Gems I-III, rend386, vr386, and other stuff. imag_utl, 4 megs, various DOS image utilities images, 204 megs, all sorts of stuff, mostly in GIF format: images/astrogif, 18 megs, 191 images, various NASA images images/fracgif, 7 megs, 48 images, 2D & 3D fractal images images/handgif, 5 megs, 49 images, hand painted images images/raygif, 61 megs, 384 images, ray traced images (8 bit mapped) images/rayjpg, 5 megs, 43 images, ray traced JPEG (lossy 24 bit) images/raytga, 69 megs, 88 images, ray traced TGA (24 bit) images/textures, 11 megs, 73 images, GIF, TGA and TIF textures, some tile, the TIF's are from Paul Bourke's collection images/tga, 23 megs, not Targa files, but rather works by Adam Schiffman (who made this CD ROM); some nice stuff misc, 1 meg of various, such as arj and zip unpackers, Visual Basic DLL Also, very useful, is a file description list which briefly describes many of the utilities and whatnot. os2_view, 1 meg of OS/2 image viewers paint, 7 megs of various paint and image processing programs ray, 15 megs of ray tracers, including OORT, Polyray, POV, Rayshade, Vivid, and the Truespace demos and WorldRender3D demo ray_utl, 7 megs of ray tracing utilities, mostly for POV, Polyray, Vivid and fairly selective (i.e. not just every utility ever made). Modelers include Moray, POVCAD, WinBlobs, PV3D. rtnews, 2 megs of utter drivel. For some reason, RTNv6n1 is missing. win_view, 8 megs of Windows viewers and whatnot.
As you can see, while the focus is ray tracing, with about 250+ of the 450 megs of stuff directly related to this topic, there are also some other things of interest, such as 3D Studio related resources, fractal generators, other images, paint programs, graphics libraries, the GNU C++ compiler, audio utilities, etc etc. A nice feature for users is that most of the files are uncompressed and can be executed directly from the CD.
The image libraries are fairly nice, and unlike the "Tracers and Tracings" CD ROM (reviewed in RTNv7n4), there appears to be more editorial work done on the collections. Many of the images are of some interest either visually or for educational purposes, e.g. focus.gif shows depth of field effects (although there's an interest [also educational] bug: the reflections are in focus!) and 6gear1.gif shows the SPD gears database rendered with 6 different ray tracers (the bug in Rayshade has been fixed!). Some of the images in the ray traced image directories are not ray traced but rather produced by 3D Studio, but I'm no purist. One annoying thing with the image libraries is that there are no catalog files, as on "Tracers and Tracings". The creator says in the readme file that he decided against this because the user could create catalogs, but this excuse is a bit weak. In terms of portability, time spent, and hard-disk space, having the catalogs on the CD ROM is a lot better than making and storing these on a particular machine. Given that there are 200+ megs of unused CD ROM disk space, storage would not have been a problem.
The addition of 3D Studio related resources is worthwhile, and of course the textures are usable elsewhere and the meshes are convertible to other formats using 3DS2POV and other utilities. The meshes and textures for the 3DS section do have catalog images on the disk, which is extremely helpful.
Editing has also been done on the various program collections; there are no cases of "8 versions of the same software" clutter. However, in checking for some of the better image utilities, I didn't find Image Alchemy (which "Tracers and Tracings" did have) image converter or the LView windows image viewer, so either a little too much editing occurred or these worthwhile programs were unknown to the creator.
All in all, this is a nice collection of software and images for hobbyists and others who are interested in modeling and image generation, and beats the daylights out of downloading all this stuff from the net. I use a fairly reasonable Internet connection, but it's a heck of a lot easier to grab stuff from CD ROMs than deal with unavailable or dropped connections and the amounts of thumb-twiddling time spent downloading (let alone finding) material on the net. Also, if I find some worthwhile software on a CD ROM I can always try to find the latest version on the net and use that. All in all, I consider CD ROM file collections to be a great boon to users of the Internet, saving much time and energy and bandwidth for us all. This CD ROM is a welcome addition.
Ordering info (straight from his advertisement):
$39.95 + $3.00 shipping and handling - California Residents will be charged applicable sales tax. $6.00 S/H for International orders.
The price includes the TGA CD-ROM with 450MB of quality graphics-related material AND a FREE 30 day subscription to TGA. For more details call TGA and download TGACDROM.ZIP - this archive gives detailed information on all the material contained on the disc.
The TGA CD-ROM may be ordered online with a VISA or Mastercard by calling the BBS and typing CREDIT from the main menu. The CD purchase also includes a 30 day complimentary subscription to TGA.
Node 1: 510-524-2780 - USR Dual 2400-14.4k, New Users
Node 2: 510-PRI-VATE - Hayes Optima 288, 2400-28.8k, Subscribers
Node 3: 510-PRI-VATE - USR Dual V.FC, 2400-28.8k, Subscribers
Or you may send a bank draft (No Personal Checks Please!!) or money order to the following address:
TGA 190 El Cerrito Plaza #107 El Cerrito, CA. 94530
Internet e-mail: firstname.lastname@example.org World Wide Web: http://www.tgax.com
back to contents
I was looking around on avalon.chinalake.navy.mil the other day and noticed an ad for SYNDESIS AVALON CDROM. I called up Syndesis [(414) 674-5200] and ordered it on monday (8/22) for $49.95 + shipping, and received it on Wednesday (8/24).
It arrived with a CDROM and a booklet. The booklet contains a Copyright page which basically says that they (Syndesis) doesn't claim ownership of the files on the CDROM, just the Copyright to the contents of the CDROM. Permission must be obtained from them to make it available on a BBS, multiuser network, or network.
I'm not a lawyer (obviously) but I had a bit of trouble understanding the distinctions they are trying to make here. Perhaps someone else would share some insight here.
In addition to the Copyright notice there is an introduction, and a section on installation. The introduction thanks Francisco DeJesus (which I heartly agree with). The Installation section talks about the fact that the disk is an ISO-9660 CDROM which can be read on any computer.
The rest of the document is just a listing of the files on the CDROM. The list serves two purposes, 1) it lets you see what's on the disk with motoring through the directories, and 2) it shows the mapping between the long file names used on the avalon site (I assume avalon is a UNIX machine) and shorter MSDOS names.
After looking through this I popped the CDROM into my MSDOS machine, brought up the MS-Windows file manager and started looking through it. The first place I looked was under the format_specs directory. The listing in the manual showed a dozen or so files, but the MS-Windows file manager showed an empty directory.
I started randomly looking around. Most directories matched the listing, but many important ones were empty including utils/converters.
I sent E-mail to Syndesis. They said I had a driver problem with my PC and that it would work on another PC. I tried my PC at work, it had the same problem. I sent that info to syndesis. They said oops, they have a mastering error and will have to re-master the disk. This will take about 3 weeks and they will send a free update to anyone who requests one.
Aside from that problem, the disk seems to be a direct copy of the avalon.chinalake.navy.mil site. The only problems I see are that:
1) I need several tools on my PC that I'm not sure where to find to make this disk useful. These tools are: a) An MSDOS equivalent of UNIX uncompress. b) An MSDOS equivalent of UNIX Tar c) An MSDOS equivalent of gunzip. d) PKUNZIP (which I already have) 2) I'll probably have to get a larger harddisk for all the cool stuff I'll be moving off the CDROM.
I would have preferred that syndesis had uncompressed all the files and then made their master disk. One of the advantages of CDROMS is that I don't have to have all that stuff on may HD. Unfortunately, I need to pull the models I'm interested off the CDROM to be able to use them. Too bad.
All in all I like the CDROM. I'm looking forward to the update, and will be looking for MSDOS equivalents of the UNIX tools I need to access a large number of the files on this disk.
[For ordering info, contact: John Foust / Syndesis Corporation (76004.1763@CompuServe.COM)]
[I had hoped to review Syndesis' second volume of models, but won't for now. Check out RTNv6n3 for a review of the first. John is putting together a third volume of models on a CD ROM right now; if you want to get a free one, you can get one by having him accept your model for inclusion on the CD. Write him for details. -EAH]
back to contents
What worked for me was looking into my computer screen at one and looking at a reflection in the screen to get my eyes to defocus; there are at least five other methods of doing it, and I leave it to you to find them in the FAQ (next paragraph) or in books. My other piece of advice: don't try to cross your eyes, as this is not usually the way to do it as it makes the images reverse themselves in depth.
So anyway, I thought I'd try to add this capability to my renderer. After puzzling through the SIRDS FAQ (which can be FTPed from katz.anu.edu.au [188.8.131.52]: /pub/stereograms, and the FAQ is maintained by Stuart Inglis (email@example.com)) and trying out a few wrong ways, I came up with a very short addition to my renderer which made some reasonable SIS (Single Image Stereograms). I was going to explain it here in depth, but it turns out there are now a few books that deal with the topic, so who am I to ruin capitalism?
Actually, I'll give you the five minute description. In creating these images you take a repeating texture pattern and vary it slightly. The pattern repeats horizontally every couple of inches or less, i.e. the repeat pattern must be less than the width between your eyes (so you can defocus and get a separate repetition of the pattern for each eye), and should probably be less than 1-1/2".
Say you just repeat the texture pattern again and again horizontally without variation. You defocus, and the left eye sees a pattern, and the right eye sees the same pattern, but is actually looking at the repetition of that pattern one to the right. Still with me? (if not, think of a picture, say Babe Ruth on a baseball card, and a copy of the Babe right next to it - defocussing makes the left eye look at the left image, the right eye at the right). You all (should) know about stereo pairs, where one image is shown to the left eye and the other to the right and your brain puts them together and gets a 3D effect. This idea of having separate similar images go to each eye is key to SIS.
Now, since the pattern is repeated a number of times, once defocussed, your left eye and right eye can move over the entire image and continue to be in sync. You will perceive the entire image to be at a distance farther from where it really is, as the angle formed by left eye/perceived thing/right eye is a lot smaller than if you just looked at the image normally. Your brain knows that when this angle is small, you're (supposed to be) looking at something far away.
Now, if you vary this left eye/perceived thing/right eye angle over an image, your brain will interpret these variations as changes in depth and you'll see a 3D effect. I've read a fair bit of gobbledy gook about how to vary the depth, about dealing with factors like the left eye seeing parts of 3D objects that the right eye doesn't, and vice versa, but forget all that. Just think that, given the basic repeating pattern being seen by each eye, if the right eye then has to look a little bit to the left of this base pattern at some feature (compared to what the left eye is seeing), then the brain will interpret this feature as being closer, since the angle has become larger.
You now know the essential secret, believe it or not. The basic algorithm starts with a texture pattern that the left eye is going to see and will be repeated. The "hidden object" which is to appear is used to vary this pattern. Say the hidden image is a square floating above a plane. When the background of this hidden image is visible, the repeat pattern is not shifted, when the square is seen, the repeat pattern is then shifted to the left. The closer the object is, the more left-shift there is. [Actually, in practice these algorithms look at the distance from pixel to pixel, and you can start from the right or from the middle of the image, but this horizontal shifting is the basic mechanism] If you look at the image (without defocussing) you'll notice that the pattern repetition is relatively rapid on the square that is closer - this makes your eyes form a larger angle with the perceived object and so you think it is closer.
So, we've started with a pattern on the left (call this pattern #1) and made a copy of this pattern just to the right (call this pattern #2), shifting the data in pattern #1 to the left as required by the hidden image to form #2. Now we move the process over to pattern #2 and use this as the starting point for the next repetition, where we form pattern #3 by shifting pattern #2 as needed. Pattern #3 is used to form pattern #4, etc, until the image is done. Algorithmically it's even simpler than this.
One great feature of these algorithms is that each scan line is entirely independent of the others. That is, you can generate a scan line without any knowledge or effect on the other scan lines, so the entire depth image does not have to be saved. Using the left-to-right algorithm above, you don't even need to keep the whole scan line of z-depths around, since you're shifting as you go along. This is what makes the algorithm easy for ray tracers: you're simply using the ray tracer to get depths as you go and shifting the pattern as you go. Note that you don't really need a ray tracer, though (hey, I had to mention ray tracing, right?) - a zbuffer renderer works just fine for the hidden images, and even just stored images used as depth maps work fine, too. Notice that a single depth map is used for generating the image, so there's nothing special about the image generation process.
The good news is that once you have a half-decent renderer it's pretty easy to add this algorithm to it. My first attempt was about 30 new lines of code to access and shift the texture pattern using the z-depth. The bad news is that once you do the easy algorithm, you notice problems. There's echoing, where pattern repetitions and elements get intertwined in a way that false 3D features appear. There's also a limit to the number of levels of depth in the basic algorithm, making the 3D object look like it was terraced for farming by miniature Southeast Asian farmers. You might try interpolating colors in the texture to get rid of this artifact (I did), and then it just looks blurry on the right side of the image. Me, I finally wrote some additional code that maintains the exact links between repetitions and uses these precise numbers to interpolate between samples for output, which reduces blurring (though I still get some echoing). Anyway, even with this additional code we're talking of a total of around 100 new lines (including whitespace and comments).
I hope this article has given you a flavor of how these images work and are generated. If you're seriously interested, look at the reviews that follow of two books which discuss in depth how to generate these images. Playing with these algorithms (and seeing the sorts of goofs you can make) has given me a greater appreciation of well-constructed commercial images made using this technique. It's also interesting to figure out what went wrong with the bad ones out there. All in all, a cute technique, but watch it: if you defocus too much your eyes will get stuck and you'll see everything in 3D.
back to contents
Richardson's book collects some of the better stereogram generating tools together and describes how to use them and how they work. He spends a little bit of time describing stereogram history and algorithms, then discusses many of the different ways the autostereogram effect can be used. He covers normal text stereograms, random text stereograms, random dot stereograms, image mapped stereograms, and icon based stereograms. Most of these (the exception being normal text stereograms, which are usually made by hand anyway) have one or more free or shareware programs presented along with them.
For random text stereogram generation the SIRTSER program, written by the author, is provided on the disk and does the job. The next program presented, RDSdraw, is a lot more fun. It's an interactive program which is tailored towards making depth images for generating random dot stereograms (SIRDS). It's a fun program in which you can put various objects (boxes, pyramids, ellipses, etc) and text on the screen and turn the image into a SIRDS. Another SIRDS program, MindImages, is also presented. I personally have a tough time getting MindImages images to work for me on my 15" screen because they have a wider repetition distance than RDSdraw, but they are nice when they do work. There is also a cute program called Shimmer which ameliorates echoing and helps some people "lock on" to the 3D image quicker. It's cute, and the effect reminds me of the book _Snow Crash_ (which is a great trashy cyberpunk [or cyberpunk parody, it's hard to tell the difference] novel).
My favorite program is RDSGEN (by Fred Feucht, from a program originally by Alexander Enzmann), for two reasons. One is that it generates image stereograms just like you see in malls, and the other is that source code is actually provided, a boon for explorers who would like to get their hands dirty with a little code. The code provided in the book has evidently been cleaned up a bit from earlier versions out there. If you have internet access, you can find some of these programs at katz.anu.edu.au [184.108.40.206]: /pub/stereograms, though I didn't notice RDSGEN there.
The book then goes on to discuss other ways of creating depth maps. The wonderful fractal program Fracint is described in this context, and the book's diskette provides some parameter and map files useful for making SIRDS (Fractint itself is not included due to space limitations, but is easy to find on the net at oak.oakland.edu:/pub/msdos/graphics/frain182.zip). Using the POV Ray ray tracer with fog to generate depth maps is also discussed, and some sample scenes are provided. To end, there's a short chapter on icon based stereograms and how to generate these with POV Ray.
Hankinson & Hermida's book focusses on random dot stereograms. The principles involved in generating and viewing stereograms are made easily understandable. Many variations and pitfalls involved in the process are covered. Along with the book is the Windows program POPOUT-LITE and some sample depth maps and random dot stereograms. Installation was a snap, and the program is easy to use for generating SIRDS. Creating depth maps is not so easy; the program relies on the Windows Paintbrush program to generate these. Using Paintbrush finesses having to present and distribute (let alone write) new software, and this program is understood by many users already. However, it is not tailored for creating depth maps and so makes for a lot of work for the user to do anything non-trivial. I find RDSdraw much better for the task, as it is made to create simple 3D object depth maps as part of the program.
Since I've begun a comparison of this book to Richardson's (which includes RDSdraw), let's continue. The software offering of Richardson's is richer. It includes programs for a wide variety of generation techniques and also includes a little source code. Hankinson includes essentially one program; the nice thing about it is that it's a Windows program which uses the standard BMP format. However, after a short while you reach its limits. You have to have an artist's abilities and patience to generate depth images using Paintbrush. For example, Hankinson would have you draw each depth slice of a tilted cylinder separately by hand, since Paintbrush doesn't know anything about 3D objects. This time would be better spent generating depth maps with appropriate tools. Hankinson does briefly discuss other methods of generating these, e.g. via POLYRAY. However, POPOUT-LITE has only 16 levels for depth information (as does RDSdraw), so images generated with it all have serious terracing artifacts (where it looks like the object is made out of layers of corrugated cardboard glued together).
Hankinson does discuss image stereograms (like those seen in malls) and puts a few in the set of plates and in the text. However, there is no software to generate these images, unlike Richardson's book. Throughout the book there are references to POPOUT PRO, which does generate these images and uses 256 depth levels, but it is made available only by paying another $17.50 (discounted from $35) for the program. There is also a confusion between GP POPOUT and POPOUT PRO which I have not figured out - they seem to be the same thing, or maybe different things; it's hard to tell from the text. Perhaps my review copy did not come with the discount coupon described in the book, which might have cleared up some of the mystery.
Whatever the case, it is misleading to put a sunflower pattern image stereogram on the cover with a hidden teapot and say "Create Your Own Incredible 3-D Stereograms with Software Included", when in fact you cannot generate image stereograms or the teapot depth map with the software provided. This is a poor way of treating readers and it mars an otherwise fine book.
Hankinson is good at fully presenting the algorithms involved. He is thorough about discussing the various problems which are encountered and ways to fix these. About two-thirds of the book purely is about the algorithms, the rest about describing how to run software.
In summary, both books have their strengths. Richardson's gives a quick once-over of the algorithms, then gets you using the programs in order to learn more. Hankinson's spends time giving a coherent, full presentation of the algorithms involved, then allows some play time to try out SIRDS. Both are well written and nicely presented. Neither book gives a formal presentation on stereogram theory (nor should they) - for this you'll need to hit the scientific literature (check the FAQ). Richardson's software offering is excellent, but the reader should be comfortable with DOS and some basic image manipulation software (most readers of this newsletter qualify). Hankinson's software is simple to use and has on-line help, but doesn't actually do all that much. Hankinson explains things thoroughly enough that you could implement the algorithms after reading his book. Richardson points at net resources and provides code; this is both a good and bad thing - it allows you to get going quickly, but without as deep an understanding of the techniques available and the problems you'll encounter. If you want to read about random dot stereograms in a methodical fashion and don't care so much about playing (or would rather surf the Internet for software), get Hankinson's book. If you would rather play as you go and learn about a wider variety of stereograms, get Richardson's.
back to contents
THE EXHIBIT FLOOR
Instead of reporting on the ultra high end stuff I am reporting on low end stuff that's pretty impressive. Well, maybe a little high end.
Rendermorphics from Reality Lab (408) 486-6956 Email firstname.lastname@example.org
At the Rendermorphics booth they were showing an API, which was explained to me as a software only render engine which is being licensed to 3D software makers. This product was rendering like a hardware renderer. In the booth they had the Matrox 3D engine (PC)and a Cirrus Logic render engine. In a very fair test, it matched the Matrox and beat the Cirrus, though both were indeed respectable. The boards also featured Open GL support, while Rendermorphics said that they plan to support this also.
[Other vendors at SIGGRAPH included Ithaca Software's new HOOPS on the PC, and Criterion's RenderWare. - EAH]
January update: Rendermorphics is now directly supported by 3D Labs' GLINT chip, a 3D rendering engine on a chip being widely adapted by the PC-Mac-Windows NT board manufacturing community. It will be even faster as a result when used with hardware acceleration.
Cirrus Logic- (510) 226-6218
The primary advantage to the Rendermorphics API is that they are making it cross platform compatible on the PC, Mac, Power PC, Windows NT and SGI (especially the INDY). Expect to start seeing this incorporated into Lower Cost 3D software in the next few months. The test that I saw was rendering at speeds comparable to my Indigo 2 Extreme. It featured Phong, Gouraud, textured, reflected and I believe faceted shading and permitted real time movement of a lit object.
They were demoing with a 200 polygon or so torus and a 16,000 poly skeleton.
Animation Master had some very cool demos on their reel this year. One was a human head so real looking people thought it was actually video. More sophisticated users thought it was Cyberscanned. But it was done at home by this guy (Rod Altshul I think was his name) in LA with photos he took of himself.
Another piece on their reel was done by a teacher at UCLA extension. Very morphy and a good, clear story. It should have been in the Electronic Theater.
The product had very cool inverse kinematics and the ability to take character actions and add them to a list of character actions in a Direction module. All versions were under $1K.
January update: A new version of AM is due out within the next 6-8 weeks, which features an integrated interface and lots of new features. Their number is (206) 750-0042 and Email is Animmaster@aol.com.
Real 3D takes the programmability award for a PC based 3D product. You can write macros for everything. It also has very impressive dynamics, inverse kinematics (though cruder than Animation Master) and a lot of high end features you normally find on high end programs. Character animation looks challenging, and I'd like to see how fast it renders complex objects. I think also around $1K.
Truespace from Caligari was similarly impressive, with animatable lattices and some impressive modeling capabilities. Very fast rendering.
Reported earlier in this news group (by someone else) Newtek announced PC, Power PC, SGI and Windows NT versions of their popular Lightwave program.
January update: Due out this month, I hear, for $995
Most Needed unusual product award goes to Digital Image Design for a product called "THE MONKEY".
This is a Claymation style armature set up to pose characters in 3D programs like Softimage, though I've been told several PC software companies were interested in supporting it.
It allows you to pose characters as a whole without having to do the select a joint, move a joint with the mouse method of character animation. It seems more useful and flexible than most of the motion capture I saw.
Cost: around $10K. (212) 222-5236.
Synthetic Images had the "Reality Blaster", a VR card(s) for PCs. $2500 to about $5K. (407) 282-7740.
I didn't see a demo, but the guy gave me the literature on the airplane back to LA. Strangely, the company appears to be located in Orlando.
3D Xpres is a poor mans digitizer ($199 at the show) where you put a handyman's right angle in front of the object to be digitized, take pictures from at least two angles and click on the features similar in the two photos (they have to be scanned first). The right angle is used to reference the x-z plane for the object.
(818) 707-6000. One of the better finds.
Supports DXF and I hear OBJ.
At Animo, they had a dialogue system that supposedly could take spoken dialogue and automatically pick mouth shapes from a list based on phonemes.
The product also featured resolution independence, impressive inbetweening, an exposure sheet and other character specific features.
Also a compositor, ink and paint and Bill and Sue Kroyer as demonstrators.
This is a 2D animation product. PC based.
(818) 762-6466 Email: email@example.com ___________________________
Linker System demoed their 2D package for the MAC and PPC. Very capable with too many features to list. I didn't see any Audio support.
Their demo artist seemed to be their main programmer, a big mistake. I saw an artist from Film Roman, the largest non-Disney animation producer in this country (The Simpsons, The Critic, Garfield, Cro, Mighty Max and more). She wanted to talk art and animation and he wanted to talk tech.
She didn't understand a word he said, and shouldn't have had to. Especially for a MAC program.
Tel # (714) 552-1904
January update: I have since heard this same story repeated by a very well known animation director in Hollywood, who actually bought the product and doesn't understand how to use it, and the company still doesn't know how to talk his language.
It sits idle and mute in his Quadra, mutely mocking him and his staff.
Another 2D animation Product was Pixiscan. This runs on PCs and SGI.
Telephone # (33-1) 49 85 17 18 (FRANCE). Looks impressive, but I didn't get a real good look.
I would like to point out that Softimage demoed the most esoteric dynamics products to date. Seems they're digging deeper and deeper into obscure physics and mathematics textbooks. I don't know what you'd use these things for, but it's pretty interesting to watch.
Softimage still doesn't support a text interface.
Wavefront has the best inverse kinematics to date for polygonal programs. Their "Smart Skin" solves a lot of problems encountered in Softimage.
back to contents
>How much faster is scanline rendering compared to raytracing ? I am
>considering whether or not a scanline rendering previewer (able to read POV
>scenes) might not be a good idea, if it were *really* fast.
My guess is, no, better yet, just a sec. OK, POV 2.0beta+ rendered tetra, 4096 polygons, in 56 seconds. Subtracting out the time to read the database, call it 52 seconds. My general software z-buffer renderer (not including database read) did it in 2 seconds [and I'm working to make it faster].
Chris also asked about whether doing a scanline previewer for POV-Ray would be worthwhile. I replied:
I'd love to see a reasonable scanline renderer. The main problem is turning everything into polygons. Also, that's where the cost comes in: a sphere is just a sphere to POV, but is 2000 polygons to a scanline renderer (or less, depending on what quality your talking here. You could get by with maybe 200 polygons per sphere). Also, as far as CSG - forget it trying to get the polygonal version of a CSG model, people form companies around doing this problem. Though there have been rumors on the net of people doing this in upcoming shareware...
Anyway, if you can polygonalize the beastie (or even if it was just a simple guess), there are at least two resources for doing scanline images. One is SIPP, which is pretty good (no front end scripting language, but then you don't need one for POV integration) - it includes texture mapping and shadowing (!). There's also VRend, which is more X windows based but the code itself (cobbled together from various sources) does indeed seem to render stuff.
SIPP's home (it's also elsewhere, tell me if you have problems): isy.liu.se [220.127.116.11]: pub/sipp/sipp-3.0.tar.Z - *SIPP* scan line z-buffer and Phong shading renderer. Jonas Yngvesson (firstname.lastname@example.org)
There are some demo programs included, give it a try!
Info on ZRend is in RTNv7n5.
Anyway, it'd be great to have a POV scanline previewer, even with limitations (since most people work using polygons and simple quadrics and quartics, it'll make 90% the people out there happy). You could probably cobble one together in a few days, adding a front end to SIPP that is simply the POV script language. Do it! Fame! Fortune! (well, fame, at least...).
The above was a pretty old exchange of messages, but hey, I can always hope that someone will actually make a reasonable z-buffer renderer integrated with a shareware modeller or as a standalone program. It shouldn't be *that* hard.
back to contents
-Monte Carlo integration to compute diffuse illumination: This permits to have refracted shadows, specular and diffuse reflections, blurry reflections, translucency, texture mapping and any geometry of light sources. -Adaptive subdivision: Objects with big radiosity gradients inside are dynamically subdivided during the radiosity process. -Ray tracing: A postprocess of ray tracing is used to calculate the image. The main characteristics of the algorithm used are: -Depth of field. -Stochastic anti-aliasing. -Optional 3D glasses stereo view. -Acceleration techniques: Three different user selectable techniques are incorporated: -Binary Object Partitioning -Adaptive voxel subdivision (with optimal coordinate system selection) -Z-Buffer for first ray acceleration -Geometry input: It reads 3D Studio v2 geometry files so the only geometric entity supported is the triangle. -Completely written in ANSI C: This way it may be compiled in any system. Actually, my work was done in a PC but I've tried and succeeded in compiling it with no modifications on a DEC OSF/1 and a HP-UX server.
If anyone wants a copy of my program (it's free), please feel free to contact me. [This program is also available now at avalon.chinalake.navy.mil via FTP in pub/utils/raygcc10.arj, so check there first. -EAH]
Eduardo Bustillo Iceta (email@example.com) Particular de Basterra 1 48990 Getxo (VIZCAYA) SPAIN
[There's been some interest in this package from the 3D Studio crowd, since it reads 3D Studio v2 files (unfortunately, the package doesn't seem to ignore packets of information which it doesn't understand - with luck the author will fix this in the future. Anyway, the demo images are interesting; a Sillion-like image, and computed rapidly at that. -EAH]
I asked about benchmarking, and Eduardo replied:
I have done some quick benchmarking tonight. The only 3DS to ray tracer file exchanger that I have is the one I wrote last year for POV. I used two example scenes:
-Scene 1: Three spheres (mirror, glass and diffuse) with one light. About 800 polygons. Resolution 320x240:
POV 2.2 RAYSMITH(default configuration) 736 s 337 s
-Scene 2: Based upon F. Sillion's example in Foley-Van Dam (plate III.27) with some added features. Raysmith showed interobject reflections (even through reflective or transmitting objects) while POV didn't (it's a ray-tracer). On the other hand Raysmith can't handle sharp shadows (needed in some cases). About 6200 polygons. Resolution 320x240:
POV 2.2 RAYSMITH(default configuration) 340 s 129 s
I don't have a file exchanger for RAD386 or Rayshade but in the version that I have access to, both of them use a similar accelerating structure: uniform voxel subdivision. Raysmith has user selectable accelerating parameters (it uses adaptive voxel subdivision) such as grid resolution, maximum grid depth or minimum number of objects per voxel. I think that with these options it should be theoretically faster (or at least as fast).
back to contents
Enclosed are the results of my survey on which (if any) of the following terms were preferred to describe making computed images that relate to how things look in the real world (I don't want to say "realistic :^}). I make no explicit suggestions of what this survey is useful for, but I hope everyone will remember how much easier it is to write a paper now that we all have the same term for "radiance".
Many people thought rendering was the same as "scan conversion" or rasterizing. Others thought that rendering and image synthesis are synonymous. It is clear that there is not agreement on what rendering means. It is also clear that some think that the "realistic" is implicit in "image synthesis". One thing is for sure-- there is no global agreement!
TERM VOTES rendering 3 realistic rendering 6 photorealistic rendering 3 image synthesis 4 realistic image synthesis 16 photorealistic image synthesis 9 (other term) 2 (no pref) 2 ----------- 45 * rendering 12 * image synthesis 29 * 7 realistic * 22 photorealistic * 12
There were three "other term" votes that also expressed a preference among the six offerings, so I counted them as one of the first six categories. The other terms listed were:
physically based image synthesis photosimulation virtual image idealistic image synthesis pseudorealistic image synthesis
back to contents
Those of you with an interest in radiative transfer theory (also known as "radiosity") may want to download the new release of /pub/doc/RadBib94.Z via anonymous ftp from (hobbes.lbl.gov). It's an online bibliography that lists over 500 references to papers, theses, books and technical reports concerning radiosity, radiative transfer theory and related topics. Many of the references concern computer graphics, which is where most of the action is these days in radiative transfer theory research. However, it also includes references to most of the important papers on radiative transfer theory from the illumination and thermal engineering literature. If you do not have ftp capabilities, send me an e-mail message with a request for a free photocopy or an ASCII text file on a 3.5" MS-DOS diskette (please specify which). Sorry, but at 121 KB (88 pages), the file is too big to send as an e-mail message.
P.S. - thanks to Greg Ward of Lawrence Berkeley Laboratory for providing space on his ftp server for this resource.
P.P.S. - if anyone finds any omissions in the bibliography, please let me know so that I can add them to the next release.
back to contents
An interesting paper on this topic is:
A David Jevans T Object Space Temporal Coherence for Ray Tracing J Proceedings of Graphics Interface '92 I Canadian Information Processing Society C Toronto, Ontario D May 1992 870154C4 176-183 K temporal coherence, efficiency, space subdivision
His method should be easy to do with Rayshade: essentially, you record each ray shot in each voxel the ray hits. When something moves in a voxel or to another voxel, only those rays in those voxels which have had changes (movement, entry/exit, etc of objects) need to be recomputed for the next image. Jevans claims an almost 4x speedup. Only works for a static camera, of course (i.e. the camera doesn't move).
Memory intensive, of course (lots of rays get stored), but Jevans has a method of keeping this reasonable: he needs only 64 bytes per voxel. He stores a resolution independent grid of 16x16 bits, essentially dividing the image into 16x16 blocks. Each eye ray (and its children) then has a particular bit which gets set for it; as you propagate the ray it marks this bit in each voxel hit by it and its children. For the next frame you move objects and note which voxels are touched. "or" together all the dirty voxel bitmasks and you get a list of which of the 16x16 blocks need to get recomputed. Simple stuff, the messy part for Rayshade is keeping the frame by frame coherence info around in a nice fashion.
back to contents
You might want to look at _Encyclopedia of Graphics File Formats_ by James D. Murray and William van Ryper, published by O'Reilly and Associates, July, 1994 (928 pages, softcover, with CD ROM, ISBN 1-56592-058-9, $59.95). Email: firstname.lastname@example.org
There's a CDROM readable by Mac, PC, and Unix systems with a bunch of utilities (most of which you can find, with some amount of effort, on the net if you know where to look). While I can quibble with the claim on the back cover that it is "the only file format book you'll ever need," it is quite complete.
It attempts to cover a lot more ground that Kay and Levine and does most of it fairly well (though there is a lot missing, too. For instance, for legal reasons many of the proprietary formats, such as Kodak Photo CD, are covered with not much more than a pointer to the "owner" of the format).
The book is organized into three parts, an Overview, describing the basics of graphics file formats, their care, and feeding, as well as other bits and pieces of useful tutorial information. The second part covers the formats themselves. The final part is a description of the CDROM and pointers to getting additional software.
There are a few glitches in the index. For instance, if you look up "PICT" format, you find only a single reference to a brief mention of the format in the introduction, but if you look up "Macintosh PICT," you find a reference to both the introduction and an article in the main body of the text. Unfortunately, the article doesn't really tell you much about PICT format either, but just points you at the Apple technical note. There is software on the CDROM, which I haven't tried, that apparently does read and handle PICT, though.
In my copy there is also an error in binding which may not be in every copy. The list of figures is inserted in the middle of the table of contents entries for section two.
In summary, _Encyclopedia of Graphics File Formats_ is an useful, if sometimes necessarily incomplete, resource.
[Disclaimer: I was sent a free review copy of the book by O'Reilly and Associates, so I may have a small bias. However, I have no connection with the publisher other than as a small contributor to their financial success due to twenty one of their other publications gracing my bookshelves that I've personally bought over the years. :-)]
Another short review:
From John Grieggs' comp.graphics FAQ:
Good introduction to graphics file format issues for both vector and raster formats, plus specific descriptions of nearly 100 file formats. CD-ROM includes sample images, original format spec documents where available, and C code snippets. Also a lot of free and shareware image conversion/manipulation software for Unix, DOS, Windows, and Mac. Much of this is available on the net (and indeed the book tells you where), but having it all pulled together is very useful. Tom Lane (email@example.com) says: "My only complaint is that there are too many typos in the printed text. Check the original spec document whenever you find something unclear or dubious."
Brian Erwin writes:
We have received several requests for a complete listing of all of the format specification documents on the CD-ROM accompanying our new book, _The Encyclopedia of Graphics File Formats_ by James D. Murray and William vanRyper. The CD-ROM contains specs from such vendors as Adobe, Aldus, Apple, IBM, Microsoft, and Silicon Graphics, along with test images and code examples for many of the formats. Also on the CD-ROM is a set of free or public domain software and shareware--for MS-DOS, Windows, OS/2, UNIX, and Macintosh platforms--that will let readers convert, view, and manipulate graphics files and images.
To access this listing: http://ora.com/gnn/bus/ora/catalog/
Graphical file format FTP site
As author of GBM, which I believe is the source code include on the CD with the book [I'm not sure what he means; the code on the disk is by the authors and does not appear to resemble this code. - EAH], I would like to point out that the latest version of GBM source is usually available on one of :-
[There are also many ray-tracing and other rendering related things for OS/2 here, too. -EAH]
I particularly mention this because the latest GBM has fixes and a faster GIF/TIFF LZW save algorithm.
- Andy Key (firstname.lastname@example.org)
back to contents
apple_mac, cray_ymp, dec_alpha, dec_mips, dec_vax, hp_700, ibm_rs6000, next_next, sgi_4d, sun_3, sun_sparc
I also understand they will be providing source in the near future.
Image types supported:
bmp, cur, eps, gif, hdf, ico, icon, iff, mpnt, pbm, pcx, pgm, pic, pict, pix, ppm, pnm, ps, ras, rgb, rla, rle, synu, tga, tiff, viff, x, xbm, xwd
back to contents
i . file format translators:
- compuserve's gif - zsoft's pcx - truevision's targa - sgi's rgb - aldus-microsoft's tiff - alias' pix - wavefront's rla - urt's rle - pbm/pgm/ppm - rgb - raw - rayshade's heightfield - jpeg
ii . image editing functions:
- resampling - blur - sharpening - cut - paste - median filter - histogram equalization - ... and so on
iii . image viewers:
- for Silicon Graphics and IBM RISC 6000 ( GL Library ) - for HP9000 series ( Starbase Library ) - for any kind of X11 platforms ( Xlib Library ) - for PC386/486 ( GRX Library ) - for Linux ( VGA and VGAGL libraries )
... and it's a LIBRARY!!!, not a program. You! will be the person who write the programs.
If your machine is a PC with MSDOS, you'll need the DJGCC compiler (or an equivalent).
The original site distributor of this package is telva.ccu.uniovi.es (18.104.22.168): /uniovi/mathdept/src/liblug-1.0.tar.gz.
Please, send me a E-mail so I can track all the users of LUG and send you news, bugs, etc ...
LUG is free! (what a wonderful word :-) but it's distributed under the GNU LIBRARY GENERAL PUBLIC LICENSE.
back to contents