Slashdot Mirror

Sketches list appears in June (if we are lucky). But there's the SIGGRAPH 2004 course on Shrek 2, a paper on GI for animated films, a piece accepted at the prestigious Electronic Theatre and of course a VES Festival presentation on the film as well, so it's definately the creme de la creme (sp?):

Visual Effects Society Festival schedule
Shrek 2 course at SIGGRAPH 2004
SIGGRAPH 2004 Electronic Theatre list
An Approximate Global Illumination System for Computer-Generated Films

Sketches list appears in June (if we are lucky). But there's the SIGGRAPH 2004 course on Shrek 2, a paper on GI for animated films, a piece accepted at the prestigious Electronic Theatre and of course a VES Festival presentation on the film as well, so it's definately the creme de la creme (sp?):

Visual Effects Society Festival schedule
Shrek 2 course at SIGGRAPH 2004
SIGGRAPH 2004 Electronic Theatre list
An Approximate Global Illumination System for Computer-Generated Films

Sketches list appears in June (if we are lucky). But there's the SIGGRAPH 2004 course on Shrek 2, a paper on GI for animated films, a piece accepted at the prestigious Electronic Theatre and of course a VES Festival presentation on the film as well, so it's definately the creme de la creme (sp?):

Visual Effects Society Festival schedule
Shrek 2 course at SIGGRAPH 2004
SIGGRAPH 2004 Electronic Theatre list
An Approximate Global Illumination System for Computer-Generated Films

j

As for organizations beating slashdot to the punch on this one, that's true... but it's good to see this getting even more exposure. :)

GPGPU (General-Purpose computation on GPUs) was a hot topic at various conferences in 2003; a number of papers were published on the subject. At SIGGRAPH 2004 there will be a full-day course on GPGPU given by eight of the experts in the field (including myself).

Mark Harris of NVIDIA maintains a website dedicated to GPGPU topics, including discussion forums and news postings. Well worth a browse if you're interested in GPGPU topics.

I look forward to seeing some of you at SIGGRAPH! :)

--Cliff

Tim Rowley's SIGGRAPH 2004 Index has links to available preprints or you can go straight to the SIGGRAPH 2004 program for the official program. There's a category called "Large Meshes and GPU Programming" this year, although there might be GPU related papers in some other categories too.

GPUs pass input and output from GPU memory at 4-12 bytes per flop. This is much faster than CPUs which are limited by bus speeds that are likely to deliver a number every sever several operations. So CPU benchmarks are bogus, using algorithms that use internal memory over and over again.

Its not always easy to reformulate algorithms to fit streaming memory and other limitations of GPUs. This issue has come up in earlier generations of custom computers. So, there are things like cyclic matrices tha map multi-dimensional matrix operations into 1-D streams, and so on.

The 2003 SIGGRAPH had a session on this topic showing you could implement a wide variety of algorithms outside of graphics.

they're only good for games

Guess again. Medical volume visualization.

Now, if you're point is that for MOST consumers, they're only good for games, you may have a point. But the other way to look at it is that, since consumers have demanded such amazing video technology, the price to deliver advanced medical visualizations to doctors has dropped dramatically.

What you used to need a $40,000 SGI O2 for, now you can do with a $1000 computer from Best Buy. That computer might actually save your life some day. Pretty amazing, if you think about it.

They don't need to manage it in real time, or at all. Photorealistic rendering for film is all about hacks and cheats. James Blinn's 1988 paper started it all, and every SIGGRAPH paper from a Pixar employee since then has followed in his footsteps, even stealing the title. Do you know that Pixar's Photorealistic Renderman, the defacto standard for rendering CGI for motion pictures, didn't even support ray-tracing until the latest version?

Interestingly, BMRT was free as in $$$ but not as in Free Software. This was one of the first software packages where I first recognized how big this distinction is. (A free as in Free Software program probably would have continued on as people may have coded around some of the disputed intellectual property - a free as in $$$ program was possible to kill with the carrot and stick of a lawsuit and buyout opportunity)

I dunno, I decided to get 'serious' a while ago, and just converted an old laptop [Kaypro II] into a luggable desktop. It has everything I need, including a 10K RPM SCSI RAID, etc.. Also this finally put to rest the squabbles we used to get into at SIGGRAPH over who's laptop could render the fastest. OK, the solid steel reinforced frame puts it in at over 50 pounds, but I'm happy when I get where I'm goin. I know it's not for everyone, but I am not the 'I want to watch my own inflight movie, mew, mew, mew' type anyhow. It's great for going to conventions and being able to get some serious work done, not to mntion a great tribute to a great machine, long live Darth Vader's Lunchbox!

actually MPEG-2 is 24 bit

I remember seeing an HDR display at siggraph, it was 30 times brighter than any commercially available display technology while producing a black that is 10 times darker. They used an array of bright LEDs behind the monitor.

..the ratio is 60,000:1 from the darkest to the lightest portion of the screen. Compare this to the 600:1 contrast ratio LCD monitors that are offered currently.

If you don't know anything about HDR, check out this information from Siggraph 2003.
Soon, you may not want to render directly into the sun, you may go blind.

Ah, a fellow "simulator sickness" sufferer. At least you don't get nauseous from FPS games like CounterStrike, like I do.

If the image fills most of your field of vision, the movement on screen will trick your brain into thinking it is actually happening. Without the corresponding movement in your body, things can get very weird.

This could be why the more expensive simulators, like flight training simulators or entertainment simulators like "Star Tours" are equipped with all those hydraulics to make the physical motion sync with the visual motion. Well, that, and to make it really "realistic". Read more about simulator sickness: Simulator Sickness

The keynote speaker at the 2003 SIGGRAPH conference in San Diego was the British astrophysicist Anthony Lasenby. He claimed that a new kind of unified Euclidean and hyperbolic geometry could explain acceleration and deceleration in the Big Bang. He was talking at SIGGRAPH because his new unification of geometry is supposed to be more elegant for computer graphics modeling than the current homogeneous coordinates now used. He wrote a book about the geometry. But I have been unable to find a paper relating to the cosmological application on the web.

This is not the first time geometry has been used to unify and simplify physics. Previous examples are Galilean coordinates, special relativity, and general relativity.

At SIGGRAPH 2003 this year, we saw a demonstration by Masahiro Fujita, the head of Sony Intelligent Dynamics Laboratory, of this robot during the Android Dreams: The Present and Future of Autonomous Robotics session.

They showed it dancing (to its own music), flashing its lights, walking around and finding its way using dynamic path planning and environment sensing. It responded to an extensive voice vocabulary and mubbled some stuff in Japanese.

A PC was wirelessly to it linked showing some graphical and textual output from the robot. Anyway, that PC was running Unix, not sure if it was Linux. I guess the robot also ran a Unix-based OS.

Very old news. "Teddy" was developed by Takeo Igarashi at the University of Tokyo, and presented at SIGGRAPH 1999. 8-13-99 Schedule

I remember Esther Dyson speaking at SIGGRAPH 2002 in San Antonio. She talked about how DNS was the first truly global, flat address space where everyone could have their own little plot of land, so to speak.

Apparently she'd never heard of (a) the postal system, or (b) the phone system. Both of which reach much further than the internet does.

I actually got up and asked her about all this 'value-added' crap we're being sold; asked her exactly why it's being sold to people who've shown no interest in it whatsoever. Eh, it was more than a year ago, and the exact memory grows faint. Point is, I did a Q at a several thousand person Q&A, and pimp-slapped Esther Dyson.

Bah. She never seemed relevant to me. A remainder from the era of 'cyber' and 'HotWired' and the internet buzz pre-1997.

--grendel drago

There are two distinct groups developing and commercializing similar technology.

The previously-posted story was about a walk-thru screen developed at Tampere University of Technology, Finland, demonstrated at SIGGRAPH 2003, which is being commercialized by FogScreen, Inc.

In the current story, the technology was developed at MIT, demonstrated for the media, and is being commercialized by IO2 Technology".

Both systems appear to use a particle wall or sheet, onto which video is projected. Neither is anywhere close to "holographic," so I'm afraid those late-night session "learning Vulcan" with Virtual T'Pol are still a few years off.

You may remember some of Glassner's excellent books on computer graphics here and here.
I sat at a table at the computer graphics conference Siggraph 2003 while Andy caught up with some of his friends (cough, not me, cough).
So I was interested to hear what graphics oriented tasks he had been doing.
He talked about writing a script for a short for, oh - about a year. And then talked about the challenges of making short films (not feature-length) on a working-class-salary-sized budget. He considered it a sabbatical.
No progress in writing new CG books <<sigh>>, but interesting to see what he's been up to.

You're absolutely right: Sony is in it for the longterm.

This is exactly what Masahiro Fujita (Sony Digital Creatures Lab) confirmed at the Siggraph 2003 - Robotics: The Android Dreams Special Session. They want to get a feel for the industry first by making entertainement robots. This is a safe learning ground as these robots are not safety critical so there is room for error/approximations. If AIBO doesn't understand your voice command or bumps into a wall, it's no big deal. When these robots become reliable and more intelligent, Sony plans to move into more safety critical and potentially huge markets: replacing labor-intensive jobs, helping handicaps etc...

The Japanese goverment just awarded $400M to research in the field of robotics. They already know this is the next big -if not biggest- thing. The US and EU should invest a lot more in robotics.

You should be looking at Siggraph which has a good history section, unfortunatly it's buried somewhere on that site. If you read the first 10 years of Cinefex magazine you'll find what you're looking for.

• Stanford Movement Publications (2002 and before) - especially Kathy Pullen's paper
  
• Motion Synthesis from Annotation
  
• SIGGRAPH 2002 Publications
  

We're working on using video see-through head mounted displays to overlay real time CG, something we call "mixed reality". It's like traditional VR, but we add a view of the real world, special effects like smoke, compressed air and interactive lighting, spatialized surround sound, etc. Take traditional VR and add in a whole bunch of technologies and techniques from theme parks and you get what we do.

One of our current research projects is for the Army to simulate urban combat training. We were also showing an entertainment version of the system on Display at SIGGRAPH in San Diego this past week.

The graphics engine for our system runs on Linux, using OpenGL and GLUT, written in C++. Control systems for special effects and point source sound are written in Java, which run on Linux, OSX, and WinXP (whatever platform supports harware interface drivers). We couldn't do what we do without OSS, and hope to release some components to the open source community once they get a little more mature.

See Our Website for more info.

Ya, it's mostly about graphics, but technically they bill it as "The World's Largest Marketplace of Computer Graphics and Interactive Techniques" (quoted from the conference webpage). So, I'm guessing this fits under "interactive techniques".

I also found this brief overview on the conference website for those that are interested.

OK /. How far away is a system like this from real-time photorealistic rendering?

I agree with other posters, who say that film resolution is a few times higher resolution than typical gamer's setup.
There are other barriers. Much of visual effects rendering involves sophisticated per-pixel programs called shaders. These often are processed in Rendermn [pixar.com]. These shaders perform generalized floating point operations, if-then-else structures, etc.
Newer generations of graphics cards are implementing floating point pixels and enough processing generality, that this kind of shading can be done on the graphics card

Before we get too excited about a scene running in/near realtime, remember that there can be a need for 100s of MB of data for 1 second of footage. Plus, lots of processing is done before rendering -- simulation of forces, cloth, hair/fur, smoke/fire, etc. Also, the composite process is performed after rendering to combine many layers (100+ for "hero" shots, like far-away harbor scenes in Pearl Harbor).

In the past few Siggraph [siggraph.org] conferences have shown prototypes of Nvidia hardware rendering movie scenes. Two years ago was a multi-character shot from Final Fantasy. Last year was a battle scene from LOTR. In each, lots of precomputing determined the geometry. While rendering, many many passes were made and the result was about a 5hz update rate.
(IMHO neither looked just like the movie -- perhaps the rendered images were adjusted in the composite phase, or they ran out of time for the demo).
The floating point operations occur so much faster on a GPU (Graphics) than a CPU, that speed-ups compared to software rendering of ~100x were not uncommon.

So, there are lots of precomputed items to rendering a scene in "realtime". And these scenes are optimized for just what you see -- nothing extra to slow down the process. There's less detail for smaller things than close things.
Do you want to fly around and explore a scene from a movie? Well, that might work -- but you won't see much detail.

Those that work on movies do not want to work in realtime -- they push their systems to the limit to maximize what they can do. Jurassic Park had scenes that took hours for one processor to render each frame. So do today's movies. Maybe with 1000s of processors, you can push it to 10+ hours per frame.

SIGGRAPH (The ACM Special Interest Group in Computer Graphics, etc) kind of performs some of this function. Not quite so glitzy, but with most of the major players who supported the 3D awards, and more besides.

Yes this is old news.

As far as the leading edge I guess having the biggest R&D dept. doesn't count, or having another paper (besides other presentations) atthis years SIGGRAPH doesn't count. Much of the technique for Gollum's skin (sub-surface scattering) came from an idea ILM made public at a RenderMan Users Group Meeting a few years ago. I guess those Sci-Tech Oscars a couple years ago, or something like the release of OpenEXR, were just figments of our imagination ;-):

Smoke Simulation For Large-Scale Phenomena
RenderMan, Theory and Practice
Creatures, Critters & Clones: Styles and Techniques Unique to Industrial Light + Magic
Academy 2001 Sci-Tech winners
ILM Sci_Tech awards

Yes this is old news.

As far as the leading edge I guess having the biggest R&D dept. doesn't count, or having another paper (besides other presentations) atthis years SIGGRAPH doesn't count. Much of the technique for Gollum's skin (sub-surface scattering) came from an idea ILM made public at a RenderMan Users Group Meeting a few years ago. I guess those Sci-Tech Oscars a couple years ago, or something like the release of OpenEXR, were just figments of our imagination ;-):

Smoke Simulation For Large-Scale Phenomena
RenderMan, Theory and Practice
Creatures, Critters & Clones: Styles and Techniques Unique to Industrial Light + Magic
Academy 2001 Sci-Tech winners
ILM Sci_Tech awards

Yes this is old news.

As far as the leading edge I guess having the biggest R&D dept. doesn't count, or having another paper (besides other presentations) atthis years SIGGRAPH doesn't count. Much of the technique for Gollum's skin (sub-surface scattering) came from an idea ILM made public at a RenderMan Users Group Meeting a few years ago. I guess those Sci-Tech Oscars a couple years ago, or something like the release of OpenEXR, were just figments of our imagination ;-):

Smoke Simulation For Large-Scale Phenomena
RenderMan, Theory and Practice
Creatures, Critters & Clones: Styles and Techniques Unique to Industrial Light + Magic
Academy 2001 Sci-Tech winners
ILM Sci_Tech awards

The two main standards currently under development are OpenGL ES by the Khronos group and the JSR-184 headed by Nokia. If you read through the list of participating companies, you'll notice a good bit of overlap; we can expect the two APIs to play quite nicely together.

Mobile 3D hardware will also be coming probably sooner than what most people think. Some Ericsson researchers will be giving a SIGGRAPH talk on the subject ("Graphics for the Masses: A Hardware Rasterization Architecture for Mobile Phones") even if nothing more than the title is known at this time.

While all mobile devices will have to make their own compromises on functionality, battery life, weight and cost, almost all of them are capable of running 3D graphics when the software is carefully constructed. Many modern software rendering techniques are based on dynamically generated/compiled code, and the processes are very similar to what happens inside 3D hardware. As these libraries will also be fairly small, they will not add cost or weight to the devices themselves. 3D chips will be reserved to those more keen on playing games on the road.

The technology is definitely coming, now all we need to do is invent the killer application. Ideas anyone?

Jouni

A grad student whose claim to fame was having the moderator of alt.sex.stories forget to remove the student's name on a posting came up with our project name: Whip Me.

We Handle Interactive Pattern Mapping Efficiently.

I would like to point your attention to one nomination in the animated short category: The Cathedral.

It's a really nice short, loosely based on a story by Jacek Dukaj, directed by a Polish animator, Tomek Baginski. It won the best animated short award at SIGGRAPH 2002.

You probably won't get a chance to see it in a movie theater (it ran for a some time in a few Polish cinemas before Minority Report and Signs), but you can download a trailer here: hi-res Divx (15 MB), low-res Divx (8 MB), low-res MPEG (9 MB).

Here is the author's page about the film (flash required).

-jfedor

I would like to point your attention to one nomination in the animated short category: The Cathedral.

It's a really nice short, loosely based on a story by Jacek Dukaj, directed by a Polish animator, Tomek Baginski. It won the best animated short award at SIGGRAPH 2002.

You probably won't get a chance to see it in a movie theater (it ran for a some time in a few Polish cinemas before Minority Report and Signs), but you can download a trailer here: hi-res Divx (15 MB), low-res Divx (8 MB), low-res MPEG (9 MB).

Here is the author's page about the film (flash required).

-jfedor

I would like to point your attention to one nomination in the animated short category: The Cathedral.

It's a really nice short, loosely based on a story by Jacek Dukaj, directed by a Polish animator, Tomek Baginski. It won the best animated short award at SIGGRAPH 2002.

You probably won't get a chance to see it in a movie theater (it ran for a some time in a few Polish cinemas before Minority Report and Signs), but you can download a trailer here: hi-res Divx (15 MB), low-res Divx (8 MB), low-res MPEG (9 MB).

Here is the author's page about the film (flash required).

-jfedor

It's been tough all around. SIGGRAPH has scaled back but at least they saved enough and reorganized so that the Conference could go on. They actually mention the situation Comdex and others are going through:

No, Centropolis. 885 baby Godzillas driven by a flocking algorithm and some AI.

In fact, much of the "CG" in Tron was hand-animated by some outsourced firm in Asia. The first movie to have "realistic CGI" was The Last Starfighter, with 27 minutes of CGI. Tron, except for the "light cycle" scene, did not have significant CGI.

Read this history of the field.

No need to apologize, I didn't find you rude or anything, except maybe passionate. As I said on re-reading my first post I think I didn't do a good job of expressing my ideas. It's not that I don't think Carmack will be right eventually (I just disagree on the timeline) or that many FX are rushed (I sure know several examples), I just couldn't find the right words :-).

As far as people from ILM I've met plenty at SIGGRAPH (attendee since 95), usually I've spoken with a few of them for a few min. I've got an acquaintance there but we only email evry so often since he is to busy. Heck I interviewed once there but I didn't get the job, oh well. Also I've inherited a page about ILM, The Unofficial ILM website:

The Unofficial ILM Website

Because of that I now have a relationship with ILM PR and they are sometimes very gracious in providing info and scoops. They even introduced me to a fwe people where I almost had a heart attack but that's beside the point.

My best suggestion is to get involved with SIGGRAPH (if you have a local chapter). If not, start saving and go to next year's SIGGRAPH, in San Diego. That's the most important event for VFX and the biggestrecruitment event for ILM and other houses. You can certainly makes losts of contacts ther and actually talk to all these people and they can give you great advice. First time I went my eyes opened, I really didn't know how much was FX all about. In case you haven't gone to one I have a primer on the conference:

SIGGRAPH 2003
SIGGRAPH Primer

If not you can contact me at the site forums.

If you want to see a video with the Virtual Chanbara (sword fighting) in action, to to this SIGGRAPH Emerging Technologies page. Actually, a coworker of mine is a committee member. Lucky bastard.

Click on the video stream towards the top of the page for audio visual enjoyment (which includes the virtual sword fighting and much more). I *so* wish I was there.

A very Quick Summary of the Virtual Chanbara is also available. Trust me. The video does a much better job.

If you want to see a video with the Virtual Chanbara (sword fighting) in action, to to this SIGGRAPH Emerging Technologies page. Actually, a coworker of mine is a committee member. Lucky bastard.

Click on the video stream towards the top of the page for audio visual enjoyment (which includes the virtual sword fighting and much more). I *so* wish I was there.

A very Quick Summary of the Virtual Chanbara is also available. Trust me. The video does a much better job.

This wasnt even close to the coolest thing at SIGGRAPH! Takeo Igarashi's work on predictive interfacing making easier 2d and 3d drawing tools was cooler. Digiplasty , a kind of 3d exquisite corpse as shown by Stewart and Makai was cooler. (For that matter the Studio, manned by Makai, Stewart, Scott and many others, where you could create 2d and 3d art and print 2d and 3d was AWESOME - you could work in there for hours, vs. the few seconds of playing with a silly virtual sword.) Scotts Dodecahedron was a wonderful example of taking something abstract and virtual and making it real and usable. Isa's overview of wearable tech and cyberfashion (she took out the notes, dammit!) was refreshing, if not so new to a frequent slashdotter. (She's a burner too!) Some of the mixed reality work being done at the University of Singapore was really neat. (This is an example of some of the most exciting stuff there. Several researchers showed some great work being done in augmented reality, and combining that with some of the reasonable priced wearable and wirelessable computing, we can see some real headway being made. One researcher even composites a virtual face back onto a fellow participant in the augemented reality environment, masking the HMD, even going so far as to track the eyes and simulate the gaze.) The results of last years meditation chamber research installation was an interesting and possibly VERY useful application of VR technology. W. Bradford Paley's work on applying alternative interfaces to explore other media was fascinating, where you can use this LARGE java tool named TextArc to examine graphically over 400 literary works. The Web3D Consortium's release of the final working draft of X3D (with tools) could end up being much more important than the newest video card from ATI. Dietmar Offenhuber's work on non-isotropic spaces at wegzeit was an interesting approach to mapping and representing real places. Zachary Simpson et al's delightfully simple shadow interactivity was many times more fun than the virtual swordfight. Fabric.ch's knowscape was also exciting, both for the viewers and the presenter, as he would find additions from his European counterparts each morning when he logged on to the shared 3d space. Kenneth Huff's beautiful art using maya was just one example of some wonderful digital work being done. Lastly, Michael J. Lyons soon-to-be-published research on the aesthetics of Tokyo's Kyoto Gardens was both informative and inspiring. And this is just a TINY PART of what happened there!

Really, SIGGRAPH was NOT just an exhibition floor with cheesey swag (although the little green LED lights were very nice) and some cool new toys. It was presentation after presentation by resesarchers, some barely able to speak engrish, but all excited about their work and open to collaboration. It was hours and hours of animation, some (Like Allain Escalle's "Le Conte du monde flottant") were so stunning as to make you forget where animation ended and life began. Disney's work on replacing one actors face with another, retaining ALL facial expression, was downright scary. And the Spiderman gag footage, his spidey-suit oddly replaced with a fully reflective silver surface, like most of the rest of SIGGRAPH'S less entertaining presentations, were surely an indication of things to come.
Take the time to go to SIGGRAPH2002 and look around. If you find something interesting, write the author. This is where the new VR and AR comes from - not ATI!

Yes it's mainly Maya and RenderMan. A good source of info is the Cinefex article. As another poster said, Shake is their main compositing app. They do develop propietary solutions, like complete apps like Massive that hook up to Maya to the standard use of MEL and plugins.

There will be some discussion of the work in 2 weeks at SIGGRAPH:

Course 30: Character Setup From Rig Mechanics to Skin Deformations: A Practical Approach
Maya Master Classes

Haha.. Kind of reminds me of Jim Blinn's Things I Hope Not to See or Hear at SIGGRAPH. Read it for a chuckle..

The color swatch at space.com is WRONG! The correct hex code taken from the academic page is #fff8e7 (which is gamma corrected assuming a display gamma of 2.2 which is only an average for various available monitors).

Here is the course from the Siggraph 2001 Conference page. And here's the researcher's home page. Looks cool, way above my head though.

media player is one of the few players that doesnt have a problem with using 3rd party codecs

I suppose that's technically a true statement, since there are only a few major media players total. For the record though, QuickTime is perfectly happy using a multitude of codecs.

(very disappointed that Discreet took down CodecCentral)

I know it seems foolish now, but I want to see the technology advancing at this rate in anticipation of real 3D displays. How many FPS do you think the NV25 could pull at 1200x1200x1200? There are a host of companies and researchers working on displays that have cubic resolutions such as this, and while the usual approach has been shining lasers at points in a plastic space treated with rare earth elements, I believe R&D at the major companies is focused on multi layering transparent LCDs (can't find any links just now :().

The demand for copious amounts of pixel data is looming and as such, there is a ways to go in the realm of graphics horsepower if we are ever going to get high definition 3D displays working. The notion of 3D displays made of many physical layers is not only feasable, but surely right around the corner. For the example of a 17" monitor, the form factor would almost be the same using a 1200-deep sandwich of pixel thick layers of LCD emulsion...

I suspect the material technology is available, but the datapath to drive such a display will have to be fast and wide in the extreme compared to the graphics systems available today. And the problem is made worse considering there will be no more backface culling to hide behind (yes, pun). Consider:

2d display, bits-per-second
1280 px X 1024px X 32bpp X 24fps = 1,006,632,960 (billion)

Volumetric 3D Display, bits-per-second
1280 px X 1024px X1024 X 32bpp X 24fps = 1,030,792,151,040 (quadrillion)

As you can see, your standard game of quake jumps two entire orders of magnitude more demanding on a video card by adopting a volumetric display strategy. That's some heavy minimum requirements. We may see the day when we all switch back to 8-bit diplays for a while as technology races to keep up with the next big leap in display technology.

Coupled with demanding framerate needs and the order of magnitude more polys that'll need rendering, the NV25 we drool over today will seem as crude and worthless as the CGA adaptor of only 10 years ago, and it'll all be here before you know it.

Actually while Red's Dream does involve a clown and a dream he might be referring to Bingo. It was produced internally by Alias/Wavefront under the direction of Chris Landreth, to test the newly created Maya. It premiered at SIGGRAPH 98 in Orlando. You can see it here:

You can also get it on video and DVD. It's included in the SIGGRAPH 98 video review of the Electronic Theatre, and at least in one of the Odysey 3D DVDs/videos (some others also included many classic animation and shorts by PDI among others):

Issue 125: SIGGRAPH 98 Electronic Theater Program
Computer Animation Marvels

Never played with magnetic fluid though.

Michael can't have been to a science museum of late. I can remeber seeing small tanks of magnetic fluid that allow you to wave magnets around near them to see what happes in museums a couple of years ago.

Oh, and Wired magazine had a lovely picture of magnetic fluid in a beautiful state that was to be shown at Siggraph (See wired for article check here for video)