My Amiga1200 had an accelerator board with an FPU socket, to which I attached a 16MHz 68882 (or thereabouts.)
I was using Lightwave3D when I noticed that all my models had suddenly become very blocky, but everything else worked fine.
I found that the FPU had become partially unseated (it was seated upside down), and I guess that a few least-significant-bit pins were unconnected - thus all fp calculation results were rounded:)
Actually a REYES-style renderer with buckets, such as PRMan, could take a different approach to this since it has the shading decoupled from image sampling. After subdivision and shading it produces 'buckets' or grids of sub-pixel-sized micropolygons which are then sampled to produce the actual image.
Perhaps a star topology with fat-piped supernodes around the world (maybe owned by pixar?) which would have the scene data and would do the shading - and then send micopolygon buckets for subnodes to render..
Probably the shading part outweighs the sampling part so I don't know how much this would help..:)
While water definitely is hard to do in 3d, for the reasons you stated, the toughest challenge definitely is making a realistic human being in 3d. For exactly the same reasons - people look at people a lot, and are extremely sensitive to the smallest nuances.
That's why Gollum is such an achievement, it's the closest thing to a realistic human actor ever achieved. (Compare e.g. to the burly brawl in matrix where it's painfully obvous when the actors are swapped to their digital counterparts, and there isn't even any real acting going on..)
I do not think the point here was about ditching paper, that transition has already mostly happened. (AFAIK)
The point was moving from drawing lines in 2d to modeling+animating+rendering in 3d. (although the result might sometimes look like hand-drawn, if wanted..)
Slashdot Mirror
I guess we found out who BadAnalogyGuy is in real life.
Well put in the inimitable Perry Bible Fellowship:r Wizard.jpg
http://www.pbfcomics.com/temporary/PBF020ADGramma
For more: http://www.pbfcomics.com/temporary/archive.html
i'm pretty sure that no such agency exists.
ee.
I guess it will be just a matter of time before they demand access to the US Army's own Internet..:) (Global Information Grid)
2 8
http://slashdot.org/article.pl?sid=04/11/13/18342
My Amiga1200 had an accelerator board with an FPU socket, to which I attached a 16MHz 68882 (or thereabouts.)
:)
I was using Lightwave3D when I noticed that all my models had suddenly become very blocky, but everything else worked fine.
I found that the FPU had become partially unseated (it was seated upside down), and I guess that a few least-significant-bit pins were unconnected - thus all fp calculation results were rounded
Actually a REYES-style renderer with buckets, such as PRMan, could take a different approach to this since it has the shading decoupled from image sampling. After subdivision and shading it produces 'buckets' or grids of sub-pixel-sized micropolygons which are then sampled to produce the actual image.
Perhaps a star topology with fat-piped supernodes around the world (maybe owned by pixar?) which would have the scene data and would do the shading - and then send micopolygon buckets for subnodes to render..
Probably the shading part outweighs the sampling part so I don't know how much this would help..:)
eetu.
Haven't you played Millennium 2.2 ? :)
(now that was a game..)
While water definitely is hard to do in 3d, for the reasons you stated, the toughest challenge definitely is making a realistic human being in 3d. For exactly the same reasons - people look at people a lot, and are extremely sensitive to the smallest nuances.
That's why Gollum is such an achievement, it's the closest thing to a realistic human actor ever achieved. (Compare e.g. to the burly brawl in matrix where it's painfully obvous when the actors are swapped to their digital counterparts, and there isn't even any real acting going on..)
eetu.
I do not think the point here was about ditching paper, that transition has already mostly happened. (AFAIK)
The point was moving from drawing lines in 2d to modeling+animating+rendering in 3d. (although the result might sometimes look like hand-drawn, if wanted..)
eetu.
Am I the only one finding it interesting that they are planning on manufacturing it in Taiwan?
:)
Is economy starting to outweigh politics even in China?
eetu.
> Sometimes people scratch their heads about benchmarks and wonder "how did they come up with that number?"
the formula is in the help files, if you care to take a look.
it's a weighted average of the game test framerates.
there is also a thorough one at Beyond 3D
> I've been reading about this, and the big rift seems to come down to this: the pixel/vertex shader programs are not optomized.
they are not optimized _for_nvidia_ - which is kinda what you would expect from an impartial benchmark..
there is some closed ecosystem research going on, check e.g. BIO-Plex at http://pet.jsc.nasa.gov/
i think at the moment they have some loops closed and a 50% closed food loop - first 100% closed test coming next year or so.
it isn't much, i agree, but at least something is being done.
eetu.
"I say we nuke them from orbit, it's the only way to be sure"
Actually, that is also a relevant question to those people with agp motherboards, who want to add a second display adapter to their systems.
The top-of-the-line adapters often have dual-head capability these days, but there are people who are interested in decent secondary display adapters.