Engadget article on nVidia

[WelshBloke]

When you look at things like PhysX and TWIMTBP. I bet they invest more into gaming than anyone else. In fact it's the main reason I have stuck with them since 3DFX died.

Yeah they've been investing heavily in gaming, but that's because that's where they make their money.
If they are going to be making most of the profit from Tegra and Tesla (as per article) I would expect them to put most of their development funding there.

 

[thilanliyan]

Fermi will most likely be the most powerful gaming card and support both DX11 and PhysX.

What are you basing this on? I haven't seen anything mentioned about gaming performance but I may have missed something.

 

[Darkstar757]

The biggest questions is with 40nm production so flawed is this going to effect Femi?

 

[Wreckage]

What are you basing this on? I haven't seen anything mentioned about gaming performance but I may have missed something.

http://www.techreport.com/articles.x/17815

 

[lopri]

I miss Anand in 1999.. well, switch that to 2002. (That's when I joined this forum) Anand in 2009 is just.. so different.

 

[Idontcare]

I miss Anand in 1999.. well, switch that to 2002. (That's when I joined this forum) Anand in 2009 is just.. so different.

lopri are you drunk-posting again

 

[Grooveriding]

If Fermi turns out to be a monster and now with Larabee dead on arrival. NVIDIA may have no competition next year and beyond.

Considering the March/April release. They have 5 or 6 months to sell their new card until the 6XXX series comes out in September next year.

 

[Idontcare]

Considering the March/April release. They have 5 or 6 months to sell their new card until the 6XXX series comes out in September next year.

Nvidia's refresh could be out then too. Could be they only have 5-6 months to sell the card period before the refresh is released.

 

[Grooveriding]

Nvidia's refresh could be out then too. Could be they only have 5-6 months to sell the card period before the refresh is released.

I don't see nv's next series coming out six months after this one... a refresh a la 65nm gt200 to 55nm gt200 sure.

But ATI is releasing their next -series- in September 2010, not a refresh.

 

[thilanliyan]

http://www.techreport.com/articles.x/17815

All that did was give estimates of what the THEORETICAL numbers COULD be. The 5970 could (and probably will) still be faster. It would be amazing if nV pulled a rabbit out of the hat and destroyed the 5970 but that IMO is unlikely. Single GPU most likely it will be faster than the 5870 but it still may not be the "most powerful gaming card" as you stated.

 

[Wreckage]

All that did was give estimates of what the THEORETICAL numbers COULD be. The 5970 could (and probably will) still be faster. It would be amazing if nV pulled a rabbit out of the hat and destroyed the 5970 but that IMO is unlikely. Single GPU most likely it will be faster than the 5870 but it still may not be the "most powerful gaming card" as you stated.

You asked, I responded. Maybe your dream will come true, maybe it won't.

:\

 

[dguy6789]

The odds of Fermi being faster than the 5870 are pretty good(it better be faster for how late it is), but the odds of it being faster than the 5970 are pretty much zero. Never in the history of GPUs has there ever been that kind of jump from one generation to the next. Fermi beating the 5970 would be like skipping a generation performance wise, which I seriously doubt will happen. Going from Geforce 6800 to Geforce 8800 in one jump is the kind of jump it would have to be over the GTX 285 to beat the 5970.

 

[Wreckage]

I'm sure they will have a X2 as well. Seeing as how they pretty much invented that.

 

[Idontcare]

I'm sure they will have a X2 as well. Seeing as how they pretty much invented that.

True, but if AMD is struggling to fit 4.3B of TSMC's 40nm process tech transistors into a power-footprint of 300W for the ATX spec I can't imagine what kind of clockspeed reductions Fermi will have to take on in order for Nvidia to shoehorn 6B xtors from the same process tech into the same 300W footprint...

 

[Wreckage]

True, but if AMD is struggling to fit 4.3B of TSMC's 40nm process tech transistors into a power-footprint of 300W for the ATX spec I can't imagine what kind of clockspeed reductions Fermi will have to take on in order for Nvidia to shoehorn 6B xtors from the same process tech into the same 300W footprint...

No idea. I doubt they would need to put 2 of their high end chips into a X2. They could use lower clocked striped down chips.

We won't know till next year I suppose.

 

[extra]

If Fermi turns out to be a monster and now with Larabee dead on arrival. NVIDIA may have no competition next year and beyond.

Uhhh, what kind of dream world are you living in?

 

[jvroig]

True, but if AMD is struggling to fit 4.3B of TSMC's 40nm process tech transistors into a power-footprint of 300W for the ATX spec I can't imagine what kind of clockspeed reductions Fermi will have to take on in order for Nvidia to shoehorn 6B xtors from the same process tech into the same 300W footprint...

We don't even know the clockspeeds yet. Maybe Fermi's clocks don't have to be so high? Also, I can't remember, but does Fermi also have something designed to make it consume less power, idle and loaded, like Cypress/Juniper?

Whenever Fermi is out, I hope TSMC has 40nm yields up to snuff already. If they have ~50% right now, I hope they up that to ~70-80% by march (is 10% improvement per month too much?), so that Fermi doesn't suffer too much production problems, and the rest can be harvested. Perhaps that's one strategy nVidia will use to be able to roll out a complete Fermi line of graphics cards in spite of TSMC's 40nm problems (if they do still exist at that point in slightly less measure).

 

[thilanliyan]

nm