I find it surprising that Sun claims that its 30 bit color is "what is likely the best color fidelity in the workstation industry". This is 10 bits per color channel and 8 bits for alpha. I'm sitting in front of an SGI Octane2, which has 12 bits for each of R,G,B,A (it costs around 3x more, but it's still a workstation, and a desktop machine at that).
Does 10 or 12 bits really make a difference over 8 bits? Of course it does. Most film work these days is rendered in either 12 bits, 10 bits logarithmic, or 16 bits. Think about it: in a dark movie theatre room, 256 levels of grey (for instance) is not a lot. And if that doesn't convince you, think about image manipulation: after a few multiplications and compositions, you'll end up with very little color resolution with 8 bits. And yes, these things are often done in hardware in the color buffer (eg flame).
and I can't imagine hundreds of thousands of kids staying up late at night with laptops under their covers instead of the far more traditional book & flashlight.
Well, provided that you're willing to spend 100 times more than you should, it allows you to use one less device! (in this case, the flashlight).
2001 is the year that the academic world finally takes computer and video games seriously.
Hmm, that's a bit harsh. Although games have never had a scientific conference or a peer-reviewed journal of their own, they have had their place in many journals and conferences for quite a while now. A few among the many: SIGGRAPH, which sponsors many conferences including of course SIGGRAPH 2001, GI the Canadian conference which often focuses on interactive rendering and animation, Eurographics, which sponsors many publications, journals, and conferences on rendering and animation, etc, etc. Gaming is one of the stronger motivations of all this research, and they do talk about other aspects of gaming. For instance SIGGRAPH had a course on game AI for at least the past 2 years, and often presents articles on 3D sound.
Just seems slightly sensationalistic to claim that the field has been ignored by the academia while it has been a driving force of so much research for at least 5 years, perhaps 10.
-- Eric Plante,
M.Sc. in CompSci on hair dynamics,
University of Montreal, 1999.
Contrarily to the guy who said that contrarily to other posts you can't get it, you _can_, in fact, get it. Geeze. In Montreal you can buy it at the grocery store ("3 Frères" on St-Laurent Street, among other places). An earlier post mentioned though that the recipe is adapted according to the country; I don't know about that, but I do know that the variety we have is not the same as in Germany; it's as much as I could gather from a German girl I met once who happened to be a sales rep for Red Bull.
And about mixing it with alcohol, well, everytime I'm at the Laïka I have at least one Space Drink, which is Red Bull and Vodka. Nice way to start the night!.
(The Laïka is one of the very recommended places to be at in Montreal if you like techno; it's a café with good DJs spinning every night, including DJ Maüs. After that you're ready for more fun at the Sona.
In Quebec, which is fairly social-democrat compared to the rest of North America, by law an employer must spend an amount corresponding to at least 5% of the salaries budget on employee training and education. This applies to any company with at least a certain number of employees (I believe the threshold is something like 10 or 20). This helps companies realize that there are benefits to letting your employees learn more. Many companies in the high tech business end up spending much more than that. For instance, it's funny to see how many people speak Canadian French at SIGGRAPH.
I can't believe, by reading people's comments, how many of you don't understand how this works. NO, Microsoft isn't modifying your web page ON YOUR SERVER, geeze. They're filtering whatever you send out through their browser, on the client side, as it comes in. Now, to come back to the issue at hand...
If I write a text, a copyrighted text that is, there is no way at all that the browser could legitimately modify this text by presenting it to the user. What is scary is that MS's lawyers probably thought of that, and my bet is that they're going to argue that the links are not part of the content, that they are not the text, and therefore more links can be inserted with no legal implications. But we all agree that links are part of the content, that they express as much as the text itself. I think Microsoft is prepared to challenge that if the need arises.
And that's not good.
Oh, and by the way, the possibility that you can insert a tag in your page to prevent IE from doing that is not making this feature any more or any less legal. If the copyright argument holds, a copyright notice suffices, I don't need to put a tag to prevent every possible copyright-infringing browser feature.
More exactly, the two main problems in computer animation are clothing and hair. Although hair is somewhat more difficult, they aren't really big rendering issues.
If you look carefully at Geri's clothing, you'll notice some odd quirks -- but I guess it's like watching a magician, you're not as impressionnable if you're a magician yourself.:-) Also keep in mind that in a motion picture production context, simulation is not your only resource; you actually have real people working for you:-). Sometimes you'll have small penetrations into the body or whatever, so you'll just pull a few vertices by hand or even load up the frames in photoshop to paint them...
As far as hair is concerned, Monsters inc (Pixar's next movie with Disney) doesn't show off any impressive animation. You can see stuff that's just as good, even better, in Mighty Joe Young (yes, most of the shots use a CG gorilla, complete with realistic-moving hair and all). The really challenging part about hair animation is the movement of long hair which interacts with itself. Otherwise, you just need to animate a few long chains (say, one per square inch), and interpolate the rest, and it looks fairly good.
Clothing animation is a very challenging problem. Efficient integration is of course an issue, but all methods should give similar solutions, if given enough time to solve the problem accurately.
A much more severe problem lies in self collisions and response. This has not been addressed at all in the article this story linked to. It has also been largely ignored in the litterature: for instance, in the Baraff & Witkin paper in siggraph 98, they describe their solution for this with a short paragraph which, unfortunately for them, clearly highlights that what they did doesn't work:-). For example, they start with the assumption that the initial position of the cloth (ie at the end of the last frame) is a legal state. This means that if something goes wrong, there's no hope of falling back on your feet. Furthermore, the solution at the last frame simply cannot be considered a legal state, because the body has moved, and something has to be done with penetrations with the body. Even if you move the body at the same time step as the simulation (a wise choice), you still end up with this problem. There are also obvious geometric problems with the approach they describe, but it is best described with both hands or at least illustrations so I won't get into that here.
To sum this up, the problem of cloth self-collisions is a very complex one. Pascal Volino has proposed a solution to this at this year's CGI (article in PDF here), but I have only skimmed through it myself, and others in the field have diverging opinions about it. Ironically though, he proposes a solution to this whole cloth self collision problem using the very same technique that Baraff used for the integration (but failed to use for self collisions).
No, unfortunately it is not even a question of computing time. The problem of clothing animation cannot in general give a correct solution, whatever the amount of CPU time you dedicate to it (within reasonable bounds... Because you can always resort to a purely probabilistic model:-) ). What makes this difficult is collisions of the cloth with itself. Maya's cloth animation has been criticized as much as any other (and BTW, Maya uses the technique by Baraff & Witkin (siggraph98) described the article this story links to).
Another one, on the Amiga 500's external hard drive (A520). If you opened the top cover, you could see that the power led was named Fred and the HD activity led was named Wilma (or was it the reverse). But then again, it was common on the Amiga to name hw parts.. But LEDs??
Make SURE you remove/usr/X11R6/lib/libGL.*, those are installed by xf4
and are not the ones you want. glplanet works fine here. pulsar -fps gives
wildly oscillating values between 230 and 370 fps. PIII 450 & GeForce.
The xmms opengl plugins have always crashed for me so nothing's new there.
Look at the other acronyms they use and how they conflict with other organizations/standards/etc
You didn't read MS's response. The NSA in "NSA key" actually means National Security Agency. Microsoft did not deny this, in fact they say so themselves. What they're trying to get us to believe is that it is called that way because the second ("backup" as they say) key was required for the NSA to approve the code, but it is not for NSA's use, thus the name. Yeah, right.
Why does it take up 7-8 MB of memory (right here now) running nothing except wmaker and a rxvt?
It's not running those programs. X is a server. You run applications that communicate with X by a socket or a port. So that communications are minimized, most things (pixmaps for example) are cached on the server side. That would explain some of the memory usage. Some more reasons include a complete protocol including authentication. You perhaps don't care about this, but that's what made X, and that's what allows me to run anything from school on my display at home (yes, I do this very often, and yes, I have a faster than 56kbps connection)
Oh and by the way, what are you comparing X to anyway? Add the memory size of X to that of your favorite free Unix-like kernel, and you're still way under NT's memory usage. And you don't get remote display capabilities from NT out of the box...
Yes, we know that, but the guy was having probs with anything over 2Gb, so we can assume he has a 32-bit platform. So what you're recommending is that he changes his machine? Nice. But we were concentrating on less expensive solutions.
If you split your backup in a few different files (say, backup ~ftp/pub/mirrors and then ~ftp/pub/linux or whatever) then you should be able to get away with tar.
This will change soon when SGI releases portions of xfs as open source, and when ext3fs is ready.
IMO too much computer animation is a bad thing. digital blubber just doesn't jiggle as naturally as good old foamed latex.
Well, it's supposed to jiggle as good, which is one of my disappointments about the computer animation in the film. JarJar's flaps seem to float around, and also many other organic creatures seem to lack weight, as if everything was moving in a pool of motor oil. This is inexcusable coming from ILM, they have the cash to do it right... Most of the non-organic stuff was impressive though... Also, notice the far away crowd in the pod racing arena's bleachers. Seems like they've just pasted a texture on there... In any case the far away characters weren't moving at all, that really caught my eye...
Just comment out ONE line in config.h
on
Red Hat 6.0
·
· Score: 1
utmpx and wtmpx exist on some systems (IRIX is theo only one I know of, there may be others) to solve the problem of utmp/wtmp logging of x terminal sessions by logging only the xdm login, not the subsequent shells you start in an xterm. Under linux & XFree it's irrelevant.
Slashdot Mirror
Does 10 or 12 bits really make a difference over 8 bits? Of course it does. Most film work these days is rendered in either 12 bits, 10 bits logarithmic, or 16 bits. Think about it: in a dark movie theatre room, 256 levels of grey (for instance) is not a lot. And if that doesn't convince you, think about image manipulation: after a few multiplications and compositions, you'll end up with very little color resolution with 8 bits. And yes, these things are often done in hardware in the color buffer (eg flame).
Well, provided that you're willing to spend 100 times more than you should, it allows you to use one less device! (in this case, the flashlight).
Judging by the Eden www site, its installation looks bigger than the one in Montreal, but still, here's a link to the Montreal Biodome's website.
Hmm, that's a bit harsh. Although games have never had a scientific conference or a peer-reviewed journal of their own, they have had their place in many journals and conferences for quite a while now. A few among the many: SIGGRAPH, which sponsors many conferences including of course SIGGRAPH 2001, GI the Canadian conference which often focuses on interactive rendering and animation, Eurographics, which sponsors many publications, journals, and conferences on rendering and animation, etc, etc. Gaming is one of the stronger motivations of all this research, and they do talk about other aspects of gaming. For instance SIGGRAPH had a course on game AI for at least the past 2 years, and often presents articles on 3D sound.
Just seems slightly sensationalistic to claim that the field has been ignored by the academia while it has been a driving force of so much research for at least 5 years, perhaps 10.
-- Eric Plante,
M.Sc. in CompSci on hair dynamics,
University of Montreal, 1999.
Contrarily to the guy who said that contrarily to other posts you can't get it, you _can_, in fact, get it. Geeze. In Montreal you can buy it at the grocery store ("3 Frères" on St-Laurent Street, among other places). An earlier post mentioned though that the recipe is adapted according to the country; I don't know about that, but I do know that the variety we have is not the same as in Germany; it's as much as I could gather from a German girl I met once who happened to be a sales rep for Red Bull.
And about mixing it with alcohol, well, everytime I'm at the Laïka I have at least one Space Drink, which is Red Bull and Vodka. Nice way to start the night!.
(The Laïka is one of the very recommended places to be at in Montreal if you like techno; it's a café with good DJs spinning every night, including DJ Maüs. After that you're ready for more fun at the Sona.
In Quebec, which is fairly social-democrat compared to the rest of North America, by law an employer must spend an amount corresponding to at least 5% of the salaries budget on employee training and education. This applies to any company with at least a certain number of employees (I believe the threshold is something like 10 or 20). This helps companies realize that there are benefits to letting your employees learn more. Many companies in the high tech business end up spending much more than that. For instance, it's funny to see how many people speak Canadian French at SIGGRAPH.
So it's like the pea plants. A purple flower is nothing a big can of white spray paint can't cure.
As long as you've got paint (or no phenylalanine in your diet), you're ok..
I can't believe, by reading people's comments, how many of you don't understand how this works. NO, Microsoft isn't modifying your web page ON YOUR SERVER, geeze. They're filtering whatever you send out through their browser, on the client side, as it comes in. Now, to come back to the issue at hand...
If I write a text, a copyrighted text that is, there is no way at all that the browser could legitimately modify this text by presenting it to the user. What is scary is that MS's lawyers probably thought of that, and my bet is that they're going to argue that the links are not part of the content, that they are not the text, and therefore more links can be inserted with no legal implications. But we all agree that links are part of the content, that they express as much as the text itself. I think Microsoft is prepared to challenge that if the need arises.
And that's not good.
Oh, and by the way, the possibility that you can insert a tag in your page to prevent IE from doing that is not making this feature any more or any less legal. If the copyright argument holds, a copyright notice suffices, I don't need to put a tag to prevent every possible copyright-infringing browser feature.
IMHO, India as a whole has much more severe human rights problems to take care of than anonymous access to cybercafes.
If you look carefully at Geri's clothing, you'll notice some odd quirks -- but I guess it's like watching a magician, you're not as impressionnable if you're a magician yourself. :-) Also keep in mind that in a motion picture production context, simulation is not your only resource; you actually have real people working for you :-). Sometimes you'll have small penetrations into the body or whatever, so you'll just pull a few vertices by hand or even load up the frames in photoshop to paint them...
As far as hair is concerned, Monsters inc (Pixar's next movie with Disney) doesn't show off any impressive animation. You can see stuff that's just as good, even better, in Mighty Joe Young (yes, most of the shots use a CG gorilla, complete with realistic-moving hair and all). The really challenging part about hair animation is the movement of long hair which interacts with itself. Otherwise, you just need to animate a few long chains (say, one per square inch), and interpolate the rest, and it looks fairly good.
A much more severe problem lies in self collisions and response. This has not been addressed at all in the article this story linked to. It has also been largely ignored in the litterature: for instance, in the Baraff & Witkin paper in siggraph 98, they describe their solution for this with a short paragraph which, unfortunately for them, clearly highlights that what they did doesn't work :-). For example, they start with the assumption that the initial position of the cloth (ie at the end of the last frame) is a legal state. This means that if something goes wrong, there's no hope of falling back on your feet. Furthermore, the solution at the last frame simply cannot be considered a legal state, because the body has moved, and something has to be done with penetrations with the body. Even if you move the body at the same time step as the simulation (a wise choice), you still end up with this problem. There are also obvious geometric problems with the approach they describe, but it is best described with both hands or at least illustrations so I won't get into that here.
To sum this up, the problem of cloth self-collisions is a very complex one. Pascal Volino has proposed a solution to this at this year's CGI (article in PDF here), but I have only skimmed through it myself, and others in the field have diverging opinions about it. Ironically though, he proposes a solution to this whole cloth self collision problem using the very same technique that Baraff used for the integration (but failed to use for self collisions).
No, unfortunately it is not even a question of computing time. The problem of clothing animation cannot in general give a correct solution, whatever the amount of CPU time you dedicate to it (within reasonable bounds... Because you can always resort to a purely probabilistic model :-) ). What makes this difficult is collisions of the cloth with itself. Maya's cloth animation has been criticized as much as any other (and BTW, Maya uses the technique by Baraff & Witkin (siggraph98) described the article this story links to).
Another one, on the Amiga 500's external hard drive (A520). If you opened the top cover,
you could see that the power led was named Fred and the HD activity
led was named Wilma (or was it the reverse). But then again,
it was common on the Amiga to name hw parts.. But LEDs??
Make SURE you remove /usr/X11R6/lib/libGL.*, those are installed by xf4
and are not the ones you want. glplanet works fine here. pulsar -fps gives
wildly oscillating values between 230 and 370 fps. PIII 450 & GeForce.
The xmms opengl plugins have always crashed for me so nothing's new there.
You didn't read MS's response. The NSA in "NSA key" actually means National Security Agency. Microsoft did not deny this, in fact they say so themselves. What they're trying to get us to believe is that it is called that way because the second ("backup" as they say) key was required for the NSA to approve the code, but it is not for NSA's use, thus the name. Yeah, right.
Why does it take up 7-8 MB of memory (right here now) running nothing except wmaker and a rxvt?
It's not running those programs. X is a server. You run applications that communicate with X by a socket or a port. So that communications are minimized, most things (pixmaps for example) are cached on the server side. That would explain some of the memory usage. Some more reasons include a complete protocol including authentication. You perhaps don't care about this, but that's what made X, and that's what allows me to run anything from school on my display at home (yes, I do this very often, and yes, I have a faster than 56kbps connection)
Oh and by the way, what are you comparing X to anyway? Add the memory size of X to that of your favorite free Unix-like kernel, and you're still way under NT's memory usage. And you don't get remote display capabilities from NT out of the box...
Yes, we know that, but the guy was having probs with anything over 2Gb, so we can assume he has a 32-bit platform. So what you're recommending is that he changes his machine? Nice. But we were concentrating on less expensive solutions.
If you split your backup in a few different files (say, backup ~ftp/pub/mirrors and then ~ftp/pub/linux or whatever) then you should be able to get away with tar.
This will change soon when SGI releases portions of xfs as open source, and when ext3fs is ready.
I hasn't been removed, it's still there.. 4069 votes and rising, linux games category down to 47% and losing ground.
/. has referred the page, just open up a new browser and type www.sierrastudios.com in the locator..
If you don't want them to be able to check that
Well, it's supposed to jiggle as good, which is one of my disappointments about the computer animation in the film. JarJar's flaps seem to float around, and also many other organic creatures seem to lack weight, as if everything was moving in a pool of motor oil. This is inexcusable coming from ILM, they have the cash to do it right... Most of the non-organic stuff was impressive though... Also, notice the far away crowd in the pod racing arena's bleachers. Seems like they've just pasted a texture on there... In any case the far away characters weren't moving at all, that really caught my eye...
utmpx and wtmpx exist on some systems (IRIX is theo only one I know of, there may be others) to solve the problem of utmp/wtmp logging of x terminal sessions by logging only the xdm login, not the subsequent shells you start in an xterm. Under linux & XFree it's irrelevant.