I applaud the 9th Circuit Court of Appeals for this courageous decision. It must have been very hard for them to separate their disgust for the Neandertals who run the site from their respect for the principles of the First Amendment. I don't think that the courts can make value judgments on speech without degenerating into the worst form of censorship.
The cure for bad speech is truly more speech. This ruling will make it much harder to quash unpopular or politically incorrect speech, which is great. Kudos, again, to the 9th Circuit!
I've read this argument many times, and each of the authors harkens back to days of his youth when 'everybody' watched the same TV shows last night, and when you went to school or work the next day you all had something to talk about. Or, you all watched the same news shows, so people had a common reference as a framework for discussion.
I think that this was a complete, and fortunately short-lived, disaster. I cannot imagine a worse tyranny than that of corporate-media consensus driven into everybody's skull. This existed from the beginning of mass media in the first days of nationwide network radio until the decline of the networks during the nineties. Before network radio, people lived in fairly cloistered environments [you can say Balkanized if you felt like it.] Even the churches were pretty decentralized -- there were common themes and dogma, but without mass communication each church had autonomy.
I love the Slashdot moderation system. While there are definitely some common themes that get moderated up or down -- causing some bias (biases that agree with mine for the large part) moderation has the intended effect of letting you see well-written or at least well-reasoned points of view. It certainly influences my writing; and I write differently for Slashdot than I do for USENET because of the moderation quality-filter.
The thing is, that if you want to find divergent viewpoints, you can just talk to other people. I don't know why the author of this book doesn't realize this. Whenever I'm with my friends, we talk about what we've been reading; and I find out wonderful things that I wouldn't find on my own. I'd much rather do this than be forced to skip over links of alternative viewpoints [alternatives selected by whom? on what agenda? under what supervision?]
The author's vision of mandated consensus is truly insane. The number of viewpoints on any issue of any import is nearly infinite; there is no way that you can force them all to be 'carried' under some 'must carry' rule.
Finally, a proof that consnesus is impossible is to extend the author's argument just a little further. Why should this glorious mandated consensus just be limited to Americans? If it's good for all (US) Americans to have this consensus, wouldn't it be even better for Canadians to be included too? And Mexicans? Russians? Chinese? Martians? Diversity exists, and it's a powerfully good thing. Command consensus won't work any more than command economies -- the marketplace of ideas is fluid and efficient.
Done by ILM, of course. Casper and Dragonheart both had CG title characters, arguably the most important characters of those films.
With good direction and animation, there's no reason that CG characters can't be every bit as compelling as real-time humans; Jar-Jar notwithstanding. He would have been just as annoying as a guy in a rubber suit with the same lines.
Give Lucas and his animators a chance before you slag them. They've got a lot of work ahead of them; and I'm sure that most of that work will end up on the screen. It all comes down to story, as it always does.
From what little I could glean from the NY Times article, this sounds like a dataflow machine; that is, a machine when the various units 'fire' when all of their inputs are present. The idea is that each functional unit of the machine could be running in parallel, asynchronously, without any of the complexity that EPIC, say, imposes.
Unfortunately for Sutherland, there's something called the PS300.
Back in the late 70's and early 80's, his company,
Evans and Sutherland, ruled the world of computer graphics with their very slick Picture System machines. These were peripherals to PDP-11s and VAXes, and were wonderfully programmable machines.
There was a fast interface between host memory and Picture System memory; letting you mess with the bits to your heart's content. We had a couple of them at NYIT's computer graphics lab; and did a lot of great animation with them.
E&S's next machine, though, was the PS300. This was a far more powerful machine, its first machine with a raster display. It was an advance in every way, except that it imposed a dataflow paradigm on programming the machine. You could only write programs by wiring up functional units. It was astonishingly difficult to write useful programs using this technology. Everybody I know that tried (and this was the early 80s, when people were used to having to work very hard to get anything on the screen at all); every one, gave up in frustration and disgust.
ILM got the most out of the machine; but that was by imposing their will on E&S to provide them with a fast direct link to the PS300's internal frame buffer.
Basically, dataflow ideas killed the PS300, which destroyed the advantage that E&S had as the pioneer graphics company, and they have never recovered from it. While the idea is charming, and to a hardware engineer it makes a lot of sense, programming them takes you back to the plugboard era of the very first WW-II machines. Nobody wants to do that.
thad
Ellen Wolfe got this just right
on
Linux in 3D
·
· Score: 2
It's nice to see such accurate and timely journalism. Indeed, there is a tidal wave of Linux boxes about to break over the visual effects industry.
People have said it before, but the major thing lacking for Linux in Visual Effects is good color management. Mac and SGI platforms have had strong color mangagement solutions for years, and if you are careful; what you see on the monitor screen is very close to what you'll see in the theater. So far, this is untrue for Linux tools, at least as far as I know.
It's a damn shame to see SGI on the losing end of this, but I do think that they've had their day in the sun; and now have to find a new niche or die.
The X33 had to die; this was an opportune time to weild the hatchet. The program was plagued with problems from the beginning, IMHO because of a lack of cohernce on the goals.
See, everybody wanted to build a single-stage-to-orbit spaceship. If you do the math, you'll find that it's just barely not impossible to do this; but to do it you have to cut every possible corner. So, fairly quickly it was determined that the X33 would not actually reach orbit, but would be a hypersonic technology demonstrator.
That would have been fine, if Lockheed had then gone to a more conservative design; but they didn't. The obvious straw that broke the camel's back was the LH2 tanks. Now, these could have been normal aluminum tanks (like the space shuttle
uses, say) but Lockheed insisted on making them out of a composite material -- and they failed more or less the way that most experts predicted they would. But, if it wasn't the tanks, it would have been something else.
Single-stage-to-orbit can work; and they will be built. It's will be one of the advantages of living on the Moon or Mars.
I can't tell you how surprised I was when I saw the designs for these ultra-long duration superpressure balloons. I had always thought that a sphere would be the strongest shape for a balloon, best able to resist pressure by having uniform stress throughout the shell.
It turns out that a pumpkin-shaped balloon, with strong cords in the seams of the pumpkin, does a much better job of resisting pressure for a given overall weight of balloon. Each gore between the ribs sees only the stress for a spherical balloon more or less the diameter of the space between the ribs, not the diameter of the whole big balloon.
What struck me was how the Greeks still influenced me thousands of years later with their ideas of 'perfect' shapes.
And, of course, this balloon didn't 'pop', it leaked...I'm sure that they'll get that fixed and try again; it's too good an idea to abandon.
Apparently there is precendent for patents being declared invalid if there was fraud of exactly this sort, in previous situations. Other companies have failed to disclose patents pending during similar meetings of standards bodies, and have had the patents thrown out.
I'm not taking the +1 bonus, because I can't find the links that I remember...oh well.
thad
Re:What ever happened to the ant movie?
on
Genetic Stone Soup
·
· Score: 1
*sigh* It's a long story.
At NYIT in the late 70's and early 80's, we had the first big CG production facility. Alex Schure, who ran the place, had gone to the University of Utah in the mid 70's, just as their funding was running out for computer graphics research. Alex brought Ed Catmull and his team from Utah out to NYIT to make CG movies.
Well, we worked developing tools for a few years, and Lance Williams wrote a script for a movie called The Works. Then we tried to make this movie; thinking along the lines of 'we have a story, we have some tools, we have some computers, let's make a movie!'. Well, it turns out that you need more than that; and between all of us we had exactly zero years of movie making experience; so we ended up basically spinning our wheels for the next couple of years.
On of those spinning wheels was Dick Lundin. He thought that the best approach to making the movie would be to do a few fully-realized scenes from the film. He took the models that were lying around, made some more, and wrote a bunch of dynamics-based tools for secondary motion, and made a couple of short films -- more or less scenes from The Works.
Unfortunately, we couldn't find a way to make the rest of the movie at the high level set by Dick Lundin's sequences; and the movie foundered in '82 and '83 -- slowly each of us realizing that it wouldn't happen.
Ed Catmull, of course, went on to found the Computer Division at Lucasfilm, and then split off to become Pixar, and the rest is history. Lance Williams is now at Disney's Secret Lab. The lab at NYIT eventually collapsed around 1989.
Alex was very protective of the work done at NYIT, and any video that remains from those days was spirited away in the middle of the night. Somebody went by the lab as it was closing and found all of the old 2" video-tapes stuffed into a dumpster -- but those tapes are all on an obsolete video format so they're probably not readable.
It was a great place, no question. Paul Heckbert keeps the best archive of NYIT stuff. if your curiousity hasn't been satisfied.
I first met Jim Kent on the stage at the Siggraph Technical Papers in 1992, he presented his 3D morph paper right after my 2D morph paper. Later, we hired him to come work with us at Pacific Data Images, where he worked with us in the R&D group. He's definitely a good guy to work with.
I left PDI six years ago to start my own company, and I've lost touch with Jim; I had heard vaguely that he'd 'gone back to school', but I had no idea that he was up to something this big. It's great to see an old friend make such a great contribution in a new field. Way to go, Jim!
Another old friend of mine, Carter Burwell, went the other way, from doing genetics with Crick at his Cold Spring Harbor laboratory, to working on early computer graphics at NYIT in the early 80s, to becoming one of the pioneers in digital music, and is finally now a leading composer for films such as the recent O Brother, Where Art Thou (somehow passed over by the Academy).
I read Aviation Week and Space Technology religiously; even though it's wretchedly conservative and pro-weapons (as you might expect) because occasionally they have remarkable images and articles.
There was one, a few years ago, taken by a guy in Houston of the night sky. He pointed the camera up, opened the shutter, and left it for a couple of hours. Of course, the image was dominated by star trails; caused by the earth's rotation.
But, dramatically, there was a string of pearls across the center of the frame, which were the geosynchronous satellites which don't move with respect to the earth. While they are way to dim to see normally, integrated over a few hours on film they showed up quite brightly indeed. One of those really amazing pictures that makes AvWeek worthwhile.
I'll leave the other responders to comment about Houdini, Maya, and the rest, but the GIMP development is near and dear to my heart.
It turns out that the big visual effects production company Rhythm and Hues (R&H) and compositing software developer Silicon Grail have each contributed the time of one programmer toward the task of extending GIMP to 16-bit-per-channel images. Way to go, John Hughes and Ray Feeney!
thad
My last steam-tunnel adventure
on
Infiltration
·
· Score: 1
So there I was at Johns Hopkins University; in the middle of what should have been my junior year; when I got the letter from the Dean suggesting that I continue my education elsewhere -- due to the fact that I was spending all of my nights hacking the steam tunnels and the newborn ARPAnet and sleeping through classes.
So, on my last day there, this only girl of our group of adventurers invited me down to the steam tunnels for one last walkthrough; with a gleam in her eye. We walked from one end of campus to the other, clambering through dark passages with perhaps a little more body contact than absolutely required. I was distraught, of course, about getting kicked out of school, but enjoying myself at the same time.
So we end up breaking into the computer room in Maryland hall at about 4:00 am. This was a common destination; as it let you combine both kinds of hacking in the same trip. When we got there, everybody else in our group had decorated the room for a surprise going-away party. I cried, of course -- what do you do? You never have friends again like those you had in school.
One thing that I really liked about the film is
that a lot of the martial arts looked like the
actors knew what they were doing; even if they
were assisted by wires.
Seeing the 'making of' videos afterward, I saw
that the actors trained for four months
to get ready for their roles...that kind of
dedication is rare in Hollywood films. Rarely do
actors get the time to get good at what they are
supposed to be doing.
Anyway, I have heard that Matrix II is really
Matrix 0; the story of the loss of the earth.
thad
We run a little digital film company here at Hammerhead Productions, and recently had a bunch of executives of a game company over for a tour, because they are interested in the convergence of film effects and games -- basically games machines will soon be able to do pretty seriously good pictures in real time; and they were looking to see if any of our filmmaking expertise could be useful to them. Several experience visual effects people have recently been hired by games companies; for just that reason.
The most interesting questions from my point of view were 1) What is the most important part of game design and 2) Who is the 'director' equivalent in a game.
We were having dinner at four separate tables, so we got four different groups of people answering the questions; but the answers were completely uniform. 1) The most important part of a game is the 'game-play', the way that the button presses influence the way the game works. Next is visuals, last is 'story'. 2) The 'director', as it were, are the game testers; the people who sit there and play the game all day, every day, as it is developed; to determine how it works and feels.
We were quite surprised, and asked the question a few different ways; but the answer was always the same.
Now, Microsoft is not writing the games (are they?) but are building the boxes, so testers would have a different role there. Still, it's a very important one.
thad
Re:Holy shit! This guy has the same lame idea I do
on
The Encryption Wars
·
· Score: 2
Toy Story was, in fact, made with tools that would shock you by their 'primitiveness'. All of the animation is done in spreadsheets and procedural languages. Even their lighting scheme in Toy Story is procedural and not interactive. It remains the best CG movie ever made, and I maintain that it is because of that proceduralism encoding the will of the animators and technical directors into the characters.
Now, it is true that other people at the apex of my computer graphics effects industry have a hard time accepting this; so I won't hold it against you. But, nevertheless, it's true. Toy Story, Toy Story II, A Bug's Life, and Antz were all animated with spreadheets. You can look it up.
Thunderbird and Duron CPUs did have this problem with RedHat 6.2 only; in that RedHat 6.2's installed kernel (as opposed to the installation kernel) tried to turn of the CPUID with a panic resulting.
The way to fix it is to boot this system at the lilo prompt with the x86_serial_nr=1 flag as in
the following
lilo: linux x86_serial_nr=1
Then, rebuild a kernel. The defaults, interestingly, don't enable the CPUID; so just making a kernel with all the defaults yeilds something that boots.
In the very best case, leaving the satellites up doesn't make their inevitable re-entry any safer. By any reasonable standard, it is more dangerous to leave them up. Right now the satellites have generous fuel margins and working control systems; and could be aimed with at least some precision. Waiting until they are derelict means giving up all that control.
Also, the Iridium orbits will decay into orbits that cross the path of the space station and other manned spacecraft; Iridium satellites are all in low polar orbits. As these orbits decay naturally, they will be at similar altitudes to the manned systems. While a 10lb titanium fuel tank might damage a car that it falls onto somewhat, a 1000lb satellite would vaporize (literally) the space station.
The bigger question is 'why were these satellites allowed to be launched at all?' This (imho flawed) analysis could have been done before the satellites were launched. It was obvious from the beginning that they would deorbit relatively soon. If it is an unacceptable risk now, it was certainly just as unacceptable then.
I really like these products from Fluke. They are not particularly inexpensive, but can be rented for a reasonable fee, and are extremely rugged. If you only need it for a limited time, why not rent something that you are sure will work?
Go watch Showgirls, and find the 20 visual effects shots that we did. Hey, find *any* of them! I agree that many times FX shots stand out in a movie, but I don't think that very often it's due to the resolution. Since the advent of the laser film recorders shooting onto IP stock, there isn't even much of a change in grain going down a digital generation.
Kodak's Cinesite likes to promote that they do 4k FX, but the huge majority of the shots that go through that facility are 2K (including the scanning/recording that they do as a service for others). Cinesite itself is strongly pushing a digital negative service that includes 2K (realtime, but still 2K) scanning.
What nobody has said so far here is that while there is no question that the theoretical resolution of film is on the order of 3000 lines, needing 6000 pixels to satisfy the Nyquist criterion, very little actual film has that kind of resolution. Practical (read somewhat blurry) lenses and fast (read grainy) negative filmstocks reduce the effective resolution to 2K. Doing effects shots at higher resolution makes them appear disturbingly sharp. Even at 2K, digital elements almost always need to be blurred to marry them into the live-action plates.
Now, I have to agree with you on the fact that none of my employees noticed that it was a digital print leads one to wonder about their ability to critically evaluate an image on the screen. It was quite disturbing to me; although perhaps it was that they were blinking back tears at the hilarious plot holes. But seriously, when it comes to our own shots; we are our own most severe critics, and while we are often bitterly disappointed at things that our clients never complain about; the resolution is just not one of those things.
It's true that Lucas is using a HDTV-resolution camera, but it is a special-built Panavision 24fps digital camera at HDTV resolution. HDTV resolution is good enough; almost every digital visual effects shot that you see on film today is scanned, calculated, and filmed out at 2048 pixels, just barely more than HDTV resolution, and the resolution isn't an issue. The exceptions are for things like starfields and credits; with super-sharp high-contrast features.
The TI DLP projectors are prototypes, and while they are indeed 1280x1024, the production ones will be 1920 pixels across. Even so, my digital film effects company went to see Mission to Mars on a digital screen, sitting in the fourth row of a huge screen, and only two of us noticed that it was digital.
Contrast and brightness are issues, and TI is working on them. Still, they are not issues for the huge majority of the viewing public. The striking quality advantages of digital -- no weave, no scratches, no projector changes, consistent vibrant color, and all the rest are true advances; and once the contrast, brightness, and resolution are improved it will be better in almost every respect.
The freedom to tweak the digital negative, which is part of the digital cinema paradigm, is a great deal. On film you have only the most ham-handed ways of adjusting color balance and brightness of a scene (and no way to adjust contrast or hue). Directors and cinematographers will gain tremendous abilities with digital negative.
I really like this. It's going to change my business in a hundred different ways; some bad, most good. The biggest change for us will be that we'll get to (ok, we'll have to!) work up until the very last day before a film is released -- right now it takes a couple of weeks to make prints and distribute them to the theater. I'm positive that some movies will take even longer; that some shots will change from Friday to Saturday, or the first weekend to the second, as last-second shots get added or audiences weigh in on the first screenings.
Interestingly, according to Sellers (arrested at the same time, based on evidence from the same officer, but whose case was dismissed yesterday for a lack of evidence despite a $1M bond based on massive evidence...) this was not a jury trial. Apparently what you get in Philadelphia, and maybe all of Pennsylvania, is a trial before a judge, and if you don't like it you can get another trial on the same charges in front of a jury.
It seems insane, but that is what was said. Perhaps this has to do with the 'speedy trial' provision of the law; perhaps not.
3DFX has always been in the forefront of 3D on Linux. From the Glide full-screen OpenGL to being among the first and best cards supported by the Precision Insight/XFree86 DRI.
It's really a shame to see somebody who cast their lot with the Linux 'market', and then lost. 3DFX has always been relatively forthcoming with the technical data needed to build drivers; in a way that n***** isn't.
Perhaps it just wasn't yet time.
thad
Re:Pixar short [deep shadow maps patent pending!]
on
Simulating Cloth in CG
·
· Score: 2
Note well that this technique is in the process of being patented by Pixar.
I have no idea how much they are trying to patent, when it may grant, and all of the other pertinent information. I do know that they make no mention of the patent in their Siggraph paper.
thad
We did the zero-g fire here at Hammerhead Productions. We had some fun doing research, zero-g fire is something that NASA and others are pretty interested in, as you might imagine.
Unfortunately, real zero-g fire pretty much goes out right away as, in general, there is no draft to bring fresh reaction products in. The only unintentional fire in a spaceship was the one on Mir, a few years ago, discussed in Linenger's book -- what was burning there was an 'oxygen candle', which of course could supply its own oxygen to the flame.
Of course, the studio just wanted something that looked cool, so we made some assumptions that made reasonable scientific sense and still looked cool -- the fire was usually a surface the moved through the atmosphere, the idea being that the fire would oxidize the fuel as it went; but it didn't just sit in one place and burn.
For the (other) nerds out there, the animation was all done with my own Z animation system and rendered with Pixar's Renderman on our Linux render garden (5 machines doesn't count as a render farm.)
Slashdot Mirror
The cure for bad speech is truly more speech. This ruling will make it much harder to quash unpopular or politically incorrect speech, which is great. Kudos, again, to the 9th Circuit!
thad
I think that this was a complete, and fortunately short-lived, disaster. I cannot imagine a worse tyranny than that of corporate-media consensus driven into everybody's skull. This existed from the beginning of mass media in the first days of nationwide network radio until the decline of the networks during the nineties. Before network radio, people lived in fairly cloistered environments [you can say Balkanized if you felt like it.] Even the churches were pretty decentralized -- there were common themes and dogma, but without mass communication each church had autonomy.
I love the Slashdot moderation system. While there are definitely some common themes that get moderated up or down -- causing some bias (biases that agree with mine for the large part) moderation has the intended effect of letting you see well-written or at least well-reasoned points of view. It certainly influences my writing; and I write differently for Slashdot than I do for USENET because of the moderation quality-filter.
The thing is, that if you want to find divergent viewpoints, you can just talk to other people. I don't know why the author of this book doesn't realize this. Whenever I'm with my friends, we talk about what we've been reading; and I find out wonderful things that I wouldn't find on my own. I'd much rather do this than be forced to skip over links of alternative viewpoints [alternatives selected by whom? on what agenda? under what supervision?]
The author's vision of mandated consensus is truly insane. The number of viewpoints on any issue of any import is nearly infinite; there is no way that you can force them all to be 'carried' under some 'must carry' rule.
Finally, a proof that consnesus is impossible is to extend the author's argument just a little further. Why should this glorious mandated consensus just be limited to Americans? If it's good for all (US) Americans to have this consensus, wouldn't it be even better for Canadians to be included too? And Mexicans? Russians? Chinese? Martians? Diversity exists, and it's a powerfully good thing. Command consensus won't work any more than command economies -- the marketplace of ideas is fluid and efficient.
thad
With good direction and animation, there's no reason that CG characters can't be every bit as compelling as real-time humans; Jar-Jar notwithstanding. He would have been just as annoying as a guy in a rubber suit with the same lines.
Give Lucas and his animators a chance before you slag them. They've got a lot of work ahead of them; and I'm sure that most of that work will end up on the screen. It all comes down to story, as it always does.
thad
Unfortunately for Sutherland, there's something called the PS300.
Back in the late 70's and early 80's, his company, Evans and Sutherland, ruled the world of computer graphics with their very slick Picture System machines. These were peripherals to PDP-11s and VAXes, and were wonderfully programmable machines. There was a fast interface between host memory and Picture System memory; letting you mess with the bits to your heart's content. We had a couple of them at NYIT's computer graphics lab; and did a lot of great animation with them.
E&S's next machine, though, was the PS300. This was a far more powerful machine, its first machine with a raster display. It was an advance in every way, except that it imposed a dataflow paradigm on programming the machine. You could only write programs by wiring up functional units. It was astonishingly difficult to write useful programs using this technology. Everybody I know that tried (and this was the early 80s, when people were used to having to work very hard to get anything on the screen at all); every one, gave up in frustration and disgust.
ILM got the most out of the machine; but that was by imposing their will on E&S to provide them with a fast direct link to the PS300's internal frame buffer.
Basically, dataflow ideas killed the PS300, which destroyed the advantage that E&S had as the pioneer graphics company, and they have never recovered from it. While the idea is charming, and to a hardware engineer it makes a lot of sense, programming them takes you back to the plugboard era of the very first WW-II machines. Nobody wants to do that.
thad
People have said it before, but the major thing lacking for Linux in Visual Effects is good color management. Mac and SGI platforms have had strong color mangagement solutions for years, and if you are careful; what you see on the monitor screen is very close to what you'll see in the theater. So far, this is untrue for Linux tools, at least as far as I know.
It's a damn shame to see SGI on the losing end of this, but I do think that they've had their day in the sun; and now have to find a new niche or die.
thad
See, everybody wanted to build a single-stage-to-orbit spaceship. If you do the math, you'll find that it's just barely not impossible to do this; but to do it you have to cut every possible corner. So, fairly quickly it was determined that the X33 would not actually reach orbit, but would be a hypersonic technology demonstrator.
That would have been fine, if Lockheed had then gone to a more conservative design; but they didn't. The obvious straw that broke the camel's back was the LH2 tanks. Now, these could have been normal aluminum tanks (like the space shuttle uses, say) but Lockheed insisted on making them out of a composite material -- and they failed more or less the way that most experts predicted they would. But, if it wasn't the tanks, it would have been something else.
Single-stage-to-orbit can work; and they will be built. It's will be one of the advantages of living on the Moon or Mars.
thad
It turns out that a pumpkin-shaped balloon, with strong cords in the seams of the pumpkin, does a much better job of resisting pressure for a given overall weight of balloon. Each gore between the ribs sees only the stress for a spherical balloon more or less the diameter of the space between the ribs, not the diameter of the whole big balloon.
What struck me was how the Greeks still influenced me thousands of years later with their ideas of 'perfect' shapes.
And, of course, this balloon didn't 'pop', it leaked...I'm sure that they'll get that fixed and try again; it's too good an idea to abandon.
thad
I'm not taking the +1 bonus, because I can't find the links that I remember...oh well.
thad
At NYIT in the late 70's and early 80's, we had the first big CG production facility. Alex Schure, who ran the place, had gone to the University of Utah in the mid 70's, just as their funding was running out for computer graphics research. Alex brought Ed Catmull and his team from Utah out to NYIT to make CG movies.
Well, we worked developing tools for a few years, and Lance Williams wrote a script for a movie called The Works. Then we tried to make this movie; thinking along the lines of 'we have a story, we have some tools, we have some computers, let's make a movie!'. Well, it turns out that you need more than that; and between all of us we had exactly zero years of movie making experience; so we ended up basically spinning our wheels for the next couple of years.
On of those spinning wheels was Dick Lundin. He thought that the best approach to making the movie would be to do a few fully-realized scenes from the film. He took the models that were lying around, made some more, and wrote a bunch of dynamics-based tools for secondary motion, and made a couple of short films -- more or less scenes from The Works.
Unfortunately, we couldn't find a way to make the rest of the movie at the high level set by Dick Lundin's sequences; and the movie foundered in '82 and '83 -- slowly each of us realizing that it wouldn't happen.
Ed Catmull, of course, went on to found the Computer Division at Lucasfilm, and then split off to become Pixar, and the rest is history. Lance Williams is now at Disney's Secret Lab. The lab at NYIT eventually collapsed around 1989.
Alex was very protective of the work done at NYIT, and any video that remains from those days was spirited away in the middle of the night. Somebody went by the lab as it was closing and found all of the old 2" video-tapes stuffed into a dumpster -- but those tapes are all on an obsolete video format so they're probably not readable.
It was a great place, no question. Paul Heckbert keeps the best archive of NYIT stuff. if your curiousity hasn't been satisfied.
thad
I left PDI six years ago to start my own company, and I've lost touch with Jim; I had heard vaguely that he'd 'gone back to school', but I had no idea that he was up to something this big. It's great to see an old friend make such a great contribution in a new field. Way to go, Jim!
Another old friend of mine, Carter Burwell, went the other way, from doing genetics with Crick at his Cold Spring Harbor laboratory, to working on early computer graphics at NYIT in the early 80s, to becoming one of the pioneers in digital music, and is finally now a leading composer for films such as the recent O Brother, Where Art Thou (somehow passed over by the Academy).
And I'm still just making pictures :) Oh well.
thad
There was one, a few years ago, taken by a guy in Houston of the night sky. He pointed the camera up, opened the shutter, and left it for a couple of hours. Of course, the image was dominated by star trails; caused by the earth's rotation.
But, dramatically, there was a string of pearls across the center of the frame, which were the geosynchronous satellites which don't move with respect to the earth. While they are way to dim to see normally, integrated over a few hours on film they showed up quite brightly indeed. One of those really amazing pictures that makes AvWeek worthwhile.
thad
It turns out that the big visual effects production company Rhythm and Hues (R&H) and compositing software developer Silicon Grail have each contributed the time of one programmer toward the task of extending GIMP to 16-bit-per-channel images. Way to go, John Hughes and Ray Feeney!
thad
So, on my last day there, this only girl of our group of adventurers invited me down to the steam tunnels for one last walkthrough; with a gleam in her eye. We walked from one end of campus to the other, clambering through dark passages with perhaps a little more body contact than absolutely required. I was distraught, of course, about getting kicked out of school, but enjoying myself at the same time.
So we end up breaking into the computer room in Maryland hall at about 4:00 am. This was a common destination; as it let you combine both kinds of hacking in the same trip. When we got there, everybody else in our group had decorated the room for a surprise going-away party. I cried, of course -- what do you do? You never have friends again like those you had in school.
thad
One thing that I really liked about the film is that a lot of the martial arts looked like the actors knew what they were doing; even if they were assisted by wires. Seeing the 'making of' videos afterward, I saw that the actors trained for four months to get ready for their roles...that kind of dedication is rare in Hollywood films. Rarely do actors get the time to get good at what they are supposed to be doing. Anyway, I have heard that Matrix II is really Matrix 0; the story of the loss of the earth. thad
The most interesting questions from my point of view were 1) What is the most important part of game design and 2) Who is the 'director' equivalent in a game.
We were having dinner at four separate tables, so we got four different groups of people answering the questions; but the answers were completely uniform. 1) The most important part of a game is the 'game-play', the way that the button presses influence the way the game works. Next is visuals, last is 'story'. 2) The 'director', as it were, are the game testers; the people who sit there and play the game all day, every day, as it is developed; to determine how it works and feels. We were quite surprised, and asked the question a few different ways; but the answer was always the same.
Now, Microsoft is not writing the games (are they?) but are building the boxes, so testers would have a different role there. Still, it's a very important one.
thad
Now, it is true that other people at the apex of my computer graphics effects industry have a hard time accepting this; so I won't hold it against you. But, nevertheless, it's true. Toy Story, Toy Story II, A Bug's Life, and Antz were all animated with spreadheets. You can look it up.
thad
The way to fix it is to boot this system at the lilo prompt with the x86_serial_nr=1 flag as in the following
lilo: linux x86_serial_nr=1
Then, rebuild a kernel. The defaults, interestingly, don't enable the CPUID; so just making a kernel with all the defaults yeilds something that boots.
thad
Also, the Iridium orbits will decay into orbits that cross the path of the space station and other manned spacecraft; Iridium satellites are all in low polar orbits. As these orbits decay naturally, they will be at similar altitudes to the manned systems. While a 10lb titanium fuel tank might damage a car that it falls onto somewhat, a 1000lb satellite would vaporize (literally) the space station.
The bigger question is 'why were these satellites allowed to be launched at all?' This (imho flawed) analysis could have been done before the satellites were launched. It was obvious from the beginning that they would deorbit relatively soon. If it is an unacceptable risk now, it was certainly just as unacceptable then.
thad
thad
Kodak's Cinesite likes to promote that they do 4k FX, but the huge majority of the shots that go through that facility are 2K (including the scanning/recording that they do as a service for others). Cinesite itself is strongly pushing a digital negative service that includes 2K (realtime, but still 2K) scanning.
What nobody has said so far here is that while there is no question that the theoretical resolution of film is on the order of 3000 lines, needing 6000 pixels to satisfy the Nyquist criterion, very little actual film has that kind of resolution. Practical (read somewhat blurry) lenses and fast (read grainy) negative filmstocks reduce the effective resolution to 2K. Doing effects shots at higher resolution makes them appear disturbingly sharp. Even at 2K, digital elements almost always need to be blurred to marry them into the live-action plates.
Now, I have to agree with you on the fact that none of my employees noticed that it was a digital print leads one to wonder about their ability to critically evaluate an image on the screen. It was quite disturbing to me; although perhaps it was that they were blinking back tears at the hilarious plot holes. But seriously, when it comes to our own shots; we are our own most severe critics, and while we are often bitterly disappointed at things that our clients never complain about; the resolution is just not one of those things.
thad
The TI DLP projectors are prototypes, and while they are indeed 1280x1024, the production ones will be 1920 pixels across. Even so, my digital film effects company went to see Mission to Mars on a digital screen, sitting in the fourth row of a huge screen, and only two of us noticed that it was digital.
Contrast and brightness are issues, and TI is working on them. Still, they are not issues for the huge majority of the viewing public. The striking quality advantages of digital -- no weave, no scratches, no projector changes, consistent vibrant color, and all the rest are true advances; and once the contrast, brightness, and resolution are improved it will be better in almost every respect.
The freedom to tweak the digital negative, which is part of the digital cinema paradigm, is a great deal. On film you have only the most ham-handed ways of adjusting color balance and brightness of a scene (and no way to adjust contrast or hue). Directors and cinematographers will gain tremendous abilities with digital negative.
I really like this. It's going to change my business in a hundred different ways; some bad, most good. The biggest change for us will be that we'll get to (ok, we'll have to!) work up until the very last day before a film is released -- right now it takes a couple of weeks to make prints and distribute them to the theater. I'm positive that some movies will take even longer; that some shots will change from Friday to Saturday, or the first weekend to the second, as last-second shots get added or audiences weigh in on the first screenings.
thad
It seems insane, but that is what was said. Perhaps this has to do with the 'speedy trial' provision of the law; perhaps not.
thad
It's really a shame to see somebody who cast their lot with the Linux 'market', and then lost. 3DFX has always been relatively forthcoming with the technical data needed to build drivers; in a way that n***** isn't.
Perhaps it just wasn't yet time.
thad
Note well that this technique is in the process of being patented by Pixar. I have no idea how much they are trying to patent, when it may grant, and all of the other pertinent information. I do know that they make no mention of the patent in their Siggraph paper. thad
Unfortunately, real zero-g fire pretty much goes out right away as, in general, there is no draft to bring fresh reaction products in. The only unintentional fire in a spaceship was the one on Mir, a few years ago, discussed in Linenger's book -- what was burning there was an 'oxygen candle', which of course could supply its own oxygen to the flame.
Of course, the studio just wanted something that looked cool, so we made some assumptions that made reasonable scientific sense and still looked cool -- the fire was usually a surface the moved through the atmosphere, the idea being that the fire would oxidize the fuel as it went; but it didn't just sit in one place and burn.
For the (other) nerds out there, the animation was all done with my own Z animation system and rendered with Pixar's Renderman on our Linux render garden (5 machines doesn't count as a render farm.)
thad