Fermi gets cut back

Status
Not open for further replies.
Smells like BS and damage control to me. Not a single solid thing there.
 
- Fermi based GPUs will be the fastest solutions in the market;
- We are very happy to see internal test results of Fermi;
- Fermi architecture developing for every single segment, and dual GPU is also possible;
- We are very happy to work with TSMC but always looking for another opportunities;
- ION 2 and Tegra 2 are very near to launch.

In other words:
- Fermi's doing okay, honest.
- Look over there at that distraction! (ion/tegra)
 
The Turkish website Donanimhaber has recently published a new video interview with Luciano Alibrandi, the Director of PR at NVIDIA.

Here are some highlights from the video:
According to Alibrandi,
- Fermi based GPUs will be the fastest solutions in the market;
- We are very happy to see internal test results of Fermi;
- Fermi architecture developing for every single segment, and dual GPU is also possible;
- We are very happy to work with TSMC but always looking for another opportunities;
- ION 2 and Tegra 2 are very near to launch.
You may click here for the video.

src= "Expreview"

Nvidia could technically make a dual gpu card. However the maximum power consumption for a pcie card allowed by the ATX specs is 300w so to meet that restriction they would either have to further cripple the card by disabling more stream processors or drastically reduce clocks.

In other words:
- Fermi's doing okay, honest.
- Look over there at that distraction! (ion/tegra)

lol That sounds about right to me.
 
Update: Nvidia Says Fermi Not Delayed - Tom's Hardware
“We are excited about the attention surrounding GF100 as it forms the foundation for the fastest and most feature rich graphics cards ever designed," said Nvidia's Bryan Del Rizzo. "Great things come to those that wait, and the wait will be over in Q1 as we have said previously. The release date has not changed since our initial comments last November." - nvidia

same old, same old...
 
They are strongly pushing it's GPGPU powers, so have they given up on 3d graphics? I think Nvidia is heading in the wrong direction. Geforce is a consumer 3d graphics card brand, not a GPGPU brand. If anything it should be a secondary function, or better yet they should have a separate architecture directed at GPGPU. I don't want my fps so slip because they decided that a feature I will never use is more important.

That, and I smell more PR bs. And we already know Tom's bends over backward for Nvidia.
 
But would people buy an ungodly priced GPGPU card that either can't play games or just sucks at them? I like Folding, and chose the GTX 260 for that purpose, but I like being able to fire up a game and it look good and play smooth, too. If, by the time Fermi gets out the door, it has been crippled or does not deliver a LOT more performance that previous generations, Nvidia will be not only shooting themselves in the foot but nailing their own coffin to boot.
 
But would people buy an ungodly priced GPGPU card that either can't play games or just sucks at them? I like Folding, and chose the GTX 260 for that purpose, but I like being able to fire up a game and it look good and play smooth, too. If, by the time Fermi gets out the door, it has been crippled or does not deliver a LOT more performance that previous generations, Nvidia will be not only shooting themselves in the foot but nailing their own coffin to boot.

Agreed. IMO gpgpu is a fad and not the future. You can't have a card be great at both things. Either the trend will die out, or they will make dedicated chips specifically for this task.
 
I also think Nvidia is on shaky ground. They have the potential to best ATI and release some monster cards. But they're gonna lose pretty much all of their low/midrange market if they keep focusing on GPGPU's. Photoshop and Illustrator CS4 are the only programs I know of that use open GL, and I can't really tell the difference with it on or off soo.
 
zmatt said:
Either the trend will die out, or they will make dedicated chips specifically for this task.
Sort of like math co-processors? ROFL!
 
Sort of like math co-processors? ROFL!

lol, yeah that was a fad too. I think all of the physx and gpgpu stuff the pr department has been putting out is to cover a lack of solid 3d performance. HPC is a very flooded market, with every major architecture having at least one super computer. 3d graphics are different, for serious stuff the is Nvidia and ATI and that's it. nvidia needs to play where the money is.
 
Status
Not open for further replies.
Back
Top Bottom