NVIDIA GF100 & Friends speculation

I don't see how any attempt at "blindsiding" ATI is going to work for nVidia. I'm sure when Cypress hit the ground running back in September ATI were designing their "5890" with the belief that Fermi was going to be very fast.

I doubt they have been slacking just because Fermi looks like being a bit of a lemon. That is the kind of behaviour that got nVidia into this mess in the first place, and it's not like ATI haven't made plenty of mistakes to learn from themselves.
 
trinibwoy said:
That sounds like wishful thinking. If they wanted to blindside ATi on launch day why are they hyping geometry and tessellation performance so much?
Click to expand...
Kinda like a wishful thinking post on [H] that said something to the effect that maybe Nv is trying to blindside ATI with their 6 month late card by releasing it in 5 months. :)
 
jimbo75 said:
I don't see how any attempt at "blindsiding" ATI is going to work for nVidia. I'm sure when Cypress hit the ground running back in September ATI were designing their "5890" with the belief that Fermi was going to be very fast.
Click to expand...

They don't want a spoiler of their launch. AMD has the whole option: faster cards, lower price or both. But without information about nVidia's line-up is a bit diffecult for them. AMD did the same with the rv770. :LOL:
 
Sontin said:
They don't want a spoiler of their launch. AMD has the whole option: faster cards, lower price or both. But without information about nVidia's line-up is a bit diffecult for them. AMD did the same with the rv770. :LOL:
Click to expand...
With the various German blogs reports of only 5000 worldwide cards at launch, availability will be it's own spoiler.
 
IIRC, Catalyst 10.3 are supposed to give a good boost in the Unigine demo (which drivers Nvidia used for 5870 enchmarking?), and looking at the graph it can be seen that there is a good difference between the green and red line only in the central part of the demo, while at the extremes performance seems to be on the same level.
 
Malo said:
Anything new? Looks like the same info released awhile ago by nvidia when they first demoed Fermi on Unigine.
Click to expand...

He explains that the 480 doesn't have much performance difference with the 5870 before you come to the tessellation part. Tessellation is a good purchase point IMO, so its a bonus if this represents ingame as well.

Later was more interesting. Doesn't seem like bezel management was ready yet when they showed Nvidia surround, but they advertise that its supposed to have bezel management. Lets hope its ready by launch and that they choose AMD's method by adding pixels instead of Matrox where they add black bars instead on the sides. :)
 
leoneazzurro said:
IIRC, Catalyst 10.3 are supposed to give a good boost in the Unigine demo (which drivers Nvidia used for 5870 enchmarking?), and looking at the graph it can be seen that there is a good difference between the green and red line only in the central part of the demo, while at the extremes performance seems to be on the same level.
Click to expand...


wasn't that for xfire not for single cards.....
 
leoneazzurro said:
IIRC, Catalyst 10.3 are supposed to give a good boost in the Unigine demo (which drivers Nvidia used for 5870 enchmarking?), and looking at the graph it can be seen that there is a good difference between the green and red line only in the central part of the demo, while at the extremes performance seems to be on the same level.
Click to expand...

It looks more and more that nVidia wants a real kick-ass DX11 card. I really interessted in their compute speed.
 
Razor1 said:
I know in 10.2 its only dx 9 and crossfire modes. And there were supposed to be the same gains in 10.3 was my understanding.

Here is a 10.3 preview

http://www.rage3d.com/articles/ati_catalyst_10/index.php?p=6

(for unigine in dx11)
Click to expand...

There are other beta leaks, I think, as some users in the forums worldwide saw some gains in unigine (please feel free to look in Google). In an Italian forum an user said to have a gain from 26 to 31 FPs in average in the DX11 demo. Also, 10.3 seem will make the whole 5xxx series performance better in many games, with ot without CF.
But, we will see on 26th and following.
 
Malo said:
Anything new? Looks like the same info released awhile ago by nvidia when they first demoed Fermi on Unigine.
Click to expand...


That is what's the 2x faster come from.
nv-tesselation-benchma0m5f.jpg
 
snapshot2010.png


heres the full picture.. the sections they use are kind of misleading.. heck at 2 areas ATI is actually faster (using old drivers and doubtfully the updated uengine).

reminds me when I used to race an aquintance .. my 72 Super Beetle (1600 CC 35HP - 44 with header, mechanical distrib.) would wasted his Buick with a 400 CI big block.. 0-30 .. on gravel ;-)
 
You must log in or register to reply here.

Similar threads

M
Interesting old GPU architecture
Replies
17
Views
3K
Ailuros
Ailuros
T
GPU Stability testing for Nvidia cards - PowerSweep
Replies
0
Views
440
T2098
T
J
Graphic card/CPU configuration for best memory graphics in Flight Simulator 98?
2
Replies
37
Views
5K
homerdog
homerdog
R
HDR settings uniformity is much needed
2 3 4
Replies
68
Views
9K
DavidGraham
D
D
Speculation and Rumors: Nvidia Blackwell ...
45 46 47
Replies
928
Views
103K
Kaotik
Kaotik
Back
Top