Yes, however only until GPUs start taking advantage of the same technology and optimize it for graphics.
Check out Stanford's IMAGINE stream processor, and corresponding software implementation of OpenGL.
They took a massively parallel architecture and optimized the communication between the ALUs, resulting in performance which competes with older GPUs in terms of graphics performance, using a general purpose multiprocessor.
And this was two years ago, by a small crew of highly talented academics with limited resources.
Just think what big megacorps like Sony, and IBM could do....
I agree, and suggest that they look at the IMAGINE and MERRIMAC streaming architectures at Stanford.
Unfortunately, this is being proposed by 1 guy at a tiny company. They'd need major multi-million dollar investments to even get this thing off the ground.
Last year at GP2, David Kirk the CTO of NVidia, was asked why NVidia doesn't move to a more flexible programmable architecture such as an FPGA.
His response was, as soon as you switch to an FPGA, you get hit with a minimum of 10x performance *loss* compared to a traditional card.
His long term answer was that GPUs will become more programmable, but only gradually, since the optimizations that they've made use parallel pipelines for redundant tasks.
If you remove the redundant tasks with programmable units, you lose performance.
So, programmability will increase as chip speeds increase, since more speed can be sacrificed.
Unfortunately, the developers of Sh chose to publish the book which details features which are not yet supported in the library. Since the software is currently in the alpha stage of development and constantly changing, many of the examples described
in their book do not work as expected (the minimal GLUT example doesn't even compile).
In short, a lot of work remains before they meet their full project goals. The name, however, which is an abbreviation of Serious Hack, seems quite appropriate.
Since Lucas continues to deny that he's going to do all 9 films that he originally envisioned, one can only hope that there might be a slim chance that he would giveup his throne and let another filmmaker finish off the last three.
Spielberg, Jackson, or Fincher would be nice.
Hopefully Lucas will get a clue and redeem himself with The Revenge of the Sith, although apparently he's already back in the cutting room after screening it to a bunch of "close" friends.
It's really tragic, because he pushes ILM incredibly hard and uses them as a source of inspiration to refine and make changes to things he hasn't decided on. In the end, a bunch of work from ILM goes either goes unnoticed, or gets exposed in the confines of a short-sighted story element, reducing the effectiveness of the visuals.
I just hope the whole "surfing on a bed of lava" ending turns out better than it sounds.
Oops.
Maybe Indy 4 will be good enough to remove the predictably bad taste leftover from EP3.
Actually, this technique uses sub-surface scattering to accurately model the light transport characteristics of human skin.
Henrik has been a pioneer in developing efficient techniques for representing BSSRDF (bidirectional sub-surface scattering distribution functions).
This paper that he published in collaboration with other notable people at Stanford was among the first to describe methods of calculating the effects of sub-surface scattering.
Since their "solid" and "reliable" web-server can't seem to handle a minor posting to slash , I dredged the web [google.com] and came upon an older (circa '01) article with some screenshots.
Slashdot Mirror
Yes, however only until GPUs start taking advantage of the same technology and optimize it for graphics.
Check out Stanford's IMAGINE stream processor, and corresponding software implementation of OpenGL.
They took a massively parallel architecture and optimized the communication between the ALUs, resulting in performance which competes with older GPUs in terms of graphics performance, using a general purpose multiprocessor.
And this was two years ago, by a small crew of highly talented academics with limited resources.
Just think what big megacorps like Sony, and IBM could do....
I agree, and suggest that they look at the IMAGINE and MERRIMAC streaming architectures at Stanford.
Unfortunately, this is being proposed by 1 guy at a tiny company. They'd need major multi-million dollar investments to even get this thing off the ground.
Last year at GP2, David Kirk the CTO of NVidia, was asked why NVidia doesn't move to a more flexible programmable architecture such as an FPGA.
His response was, as soon as you switch to an FPGA, you get hit with a minimum of 10x performance *loss* compared to a traditional card.
His long term answer was that GPUs will become more programmable, but only gradually, since the optimizations that they've made use parallel pipelines for redundant tasks.
If you remove the redundant tasks with programmable units, you lose performance.
So, programmability will increase as chip speeds increase, since more speed can be sacrificed.
Here's your M5 for you, right here.
Entertainment science at it's finest.
I think his point is we might have intelligent humans by 2020.
Unfortunately, the developers of Sh chose to publish the book which details features which are not yet supported in the library. Since the software is currently in the alpha stage of development and constantly changing, many of the examples described in their book do not work as expected (the minimal GLUT example doesn't even compile).
In short, a lot of work remains before they meet their full project goals. The name, however, which is an abbreviation of Serious Hack, seems quite appropriate.
Since Lucas continues to deny that he's going to do all 9 films that he originally envisioned, one can only hope that there might be a slim chance that he would giveup his throne and let another filmmaker finish off the last three.
Spielberg, Jackson, or Fincher would be nice.
Hopefully Lucas will get a clue and redeem himself with The Revenge of the Sith, although apparently he's already back in the cutting room after screening it to a bunch of "close" friends.
It's really tragic, because he pushes ILM incredibly hard and uses them as a source of inspiration to refine and make changes to things he hasn't decided on. In the end, a bunch of work from ILM goes either goes unnoticed, or gets exposed in the confines of a short-sighted story element, reducing the effectiveness of the visuals.
I just hope the whole "surfing on a bed of lava" ending turns out better than it sounds.
Oops.
Maybe Indy 4 will be good enough to remove the predictably bad taste leftover from EP3.
Actually, this technique uses sub-surface scattering to accurately model the light transport characteristics of human skin.
Henrik has been a pioneer in developing efficient techniques for representing BSSRDF (bidirectional sub-surface scattering distribution functions).
This paper that he published in collaboration with other notable people at Stanford was among the first to describe methods of calculating the effects of sub-surface scattering.
Pretty innovative stuff....eh? (sarcasm).