How come there is no next generation Nvidia speculation?

Mendel

Mr. Upgrade
Veteran
I mean theoretically even if Fermi was late, that should not delay their next step because they are supposed to work on a pipeline of refreshes and architechtural steps that don't depend on each other.

But maybe it's because of the failure of smaller process technology and they could not maneouver around that like AMD seems to do with their confused-with-how-to-hold-compass islands?

Finally, is the common concensus that there will probably be no gtx5xx/gtx6xx/whatever next generation from Nvidia during 2011?
 
Cause they haven't even finished rolling out this generation yet! :) But there is some speculation in dribs and drabs, just not all in one place.
 
Its because Charlie made a very convincing article about how this was the last generation of Nvidia GPU hardware... :)
 
I mean theoretically even if Fermi was late, that should not delay their next step because they are supposed to work on a pipeline of refreshes and architechtural steps that don't depend on each other.

But maybe it's because of the failure of smaller process technology and they could not maneouver around that like AMD seems to do with their confused-with-how-to-hold-compass islands?

Finally, is the common concensus that there will probably be no gtx5xx/gtx6xx/whatever next generation from Nvidia during 2011?


Mostly because:

— No one really expects anything from NVIDIA before mid-2011,
— There hasn't been a single mention of anything anywhere, no code name, no rumor about anything, zilch!

There was a lot of speculation about [strike]GT300[/strike] GF100, but that's because at the very least, we knew NVIDIA was working on it, and we knew it was due in late 2009.

Obviously they're working on something new, but until we hear something…

And what trinibwoy said.
 
A 512 CC "GTX 490" with some clock bumps might get near 6870.
Based on what evidence?

What says the 6870 will be any sort of significant performance bump on top of the 5870? We know ATI are loath to design large chips these days, and the only way to add major amounts of performance barring process shrink (which they don't have at this time) is to increase die size - which is already on the largeish side of things.

Of course we're free to speculate, but seems to me there's no actual fact to support such speculation at this time, making said speculation even more dubious than regular speculation... ;)
 
NV is off any new architecture until 28nm process is ready for mass deployment, and that ain't coming soon enough. They simply exhausted the capacity of the currently available 40nm tech from TSMC and can't go any further.
AMD on the other hand still have TDP and die area to burn for their upcoming refresh line, on the same 40nm process.
 
Based on what evidence?

What says the 6870 will be any sort of significant performance bump on top of the 5870? We know ATI are loath to design large chips these days, and the only way to add major amounts of performance barring process shrink (which they don't have at this time) is to increase die size - which is already on the largeish side of things.

Of course we're free to speculate, but seems to me there's no actual fact to support such speculation at this time, making said speculation even more dubious than regular speculation... ;)

You must have missed this: http://we.pcinlife.com/thread-1498296-1-1.html

Besides, isn't it natural to expect the next generation to bring performance improvements?
 
Isn't the next nVidia part just gonna be a proper Fermi that isn't nuclear powered and hotter than a star? :|

GF104 isn't.. it has a much much better power/FLOP than GF100.

Speculating, a GTX495 with two GF104 chips would work very well both in terms of power (under 300 W) and performance. This is easy to see in benchmarks like Anand's GTX460 review where SLI GTX460s use less power and give more FPS than a single GTX480.
 
People still ignore dual-GPU boards. Many reviewers marched GTX480 against HD5870, despite the HD5970 was closer in price and power consumption...
 
GF104 isn't.. it has a much much better power/FLOP than GF100.

Speculating, a GTX495 with two GF104 chips would work very well both in terms of power (under 300 W) and performance. This is easy to see in benchmarks like Anand's GTX460 review where SLI GTX460s use less power and give more FPS than a single GTX480.
Yup, I can't see NV releasing anything else than dual-GF104 board to (try to!) counter R9xx. I wonder how high they could clock it while remaining below 300W with the full 384 SPs - probably not very high.

But chip-wise, I assume they've been dedicating most of their resources to 28nm for some time. Jen-Hsun clearly said we won't see an architectural refresh before 28nm, so I wonder if they'll skip the usual tick-tock and do both at the same time. The only codename we heard so far was GF119 from Charlie, but that doesn't necessarily mean it will be the first one to come out I suppose.
 
Maybe we should draw up a 20 item poll with famous scientists' names and get a vote going for the next code name.
 
Here's my next-gen Nvidia speculation: It will be late.
 
In retrospect, it is more than likely that the rumored "Fermi 2" i.e. the new NV35 was indeed GF104;
I guess GF104 shows everyone that AMD has a lot of headroom should it come up with something scaling only half as good as GF104's PE versus GF100's PE.
 
Because maybe both nvidia and amdati are hitting limits with thermal and powar consumption and prefer to wait for either a process shrink or major stepping revision...

Meh personally I would have preferred a 40nm GT200b revision to lower consumtion and thermals but people would either complain its last gen or just buy it to overclock it and post scores defeating the green purpose...

28nm is the first speculation.
 
Back
Top