The average person's left hand does 56% of the typing
|
|
|
Nvidia F@H client MIA
If you remember what happened with the last NV killer app, GPU physics, the firm mouthed off about it, then realised it was losing badly to ATI. What followed was a swift shutting up, followed by a complete corporate forgetting of the concept.
Fast forward to GPGPU. NV launched it while avoiding the IT press like the plague while courting the non-technical press. That should tell you something there. They launched it quietly, then ATI launched theirs. Suddenly NV was loathe to so much as talk about it publicly. It kind of makes you wonder.
The most damning bit is the Folding@Home client, or lack thereof. I am told by coders, cuda devs and even ATIers that the F@H clients are almost trivial to write. ATI did it, every architecture under the sun has an optimized client, and it will even run on the PS3.
The one thing it won't run on is the G80, or any other NVidia card for that matter. I can't find a beta client, a project doing it, or anything else. With the supposed edge is GPU power, you would think NV would be hot to stuff ATI's face in the dirt with their superiority.
The fact that it didn't happen tells us there is an Achilles heel to the architecture, or it isn't nearly as general purpose as ATI's last gen parts. With the release of R600 in a month or so, one has to wonder if the concept of GPGPUs will be forgotten about at NV altogether.
The promised horde of happy clients didn't arrive, there wasn't even a trickle of mildly bemused shills, just deafening silence. Cuda devs fill my inbox with questions about why this 'crap' doesn't work, and ATI points to clients.
All this together points something seriously wrong with the NV architecture. G7x had great performance to a point, then once you hit a certain number of used registered, performance collapsed. You have to wonder what the straw is for the camel named G80? µ
Share this:
Share
