NVIDIA GT200 Rumours & Speculation Thread

[pso]

And no way the GTX 280 will get 43 fps in 19x12 very high settings...

It's not Very High settings, only High with no AA.

 

[pso]

Apparently Charlie from Inq has GTX 280 benchmarks...

Take with a grain of salt, especially coming from him.

http://www.theinquirer.net/gb/inquirer/news/2008/06/14/complete-gtx280-scores-here

 

[trinibwoy]

Those aren't too bad. Looks like reviewers will have to crank those CPU speeds. 3Ghz Penryn's don't seem to be cutting it.

 

[pjbliverpool]

Those aren't too bad. Looks like reviewers will have to crank those CPU speeds. 3Ghz Penryn's don't seem to be cutting it.

Nehalem will go nicely with this I think

 

[A.L.M.]

Not fake according to spajdr:



http://www.nvnews.net/vbulletin/showpost.php?p=1681935&postcount=355



http://www.nvnews.net/vbulletin/showpost.php?p=1681939&postcount=356

How can he say that the table is true or not if he didn't posted it?
Beyond that, W1zzard said that there's no relevant performance increase coming with 177.34.

 

[Arty]

Nehalem will go nicely with this I think

By the time Nehalem hits the stores, we'll probably have the GT200b.

 

[igg]

I wonder how soon Nvidia will release the GT200b: Smaller, faster due to higher clocks, less noise.

This time I hope The Inq speaks the truth and it's coming in Sep .

 

[Slappi]

I wonder how soon Nvidia will release the GT200b: Smaller, faster due to higher clocks, less noise.

This time I hope The Inq speaks the truth and it's coming in Sep .

Looks like Charlie is using the GT200b as his saviour for bashing the GT200 so bad.

The GT200b is being fast tracked. It will be here ASAP (3 months) barring bad luck.

 

[igg]

The GT200b is being fast tracked. It will be here ASAP (3 months) barring bad luck.

Have you this information from Inq other from somewhere else?

 

[Arun]

3 months? Depending on how fast tracked it is, then that doesn't exclude the possibility they don't have the chips back yet and are hoping they can ship it as an A11... (there's a pretty good possibility they could, mind you, but heh)

 

[Lukfi]

Doesn't nVidia ship every chip in the first revision, unlike ATi?

 

[igg]

If the GT200b is so close, I wonder why there isn't more information available yet.
- 55 nm (likely)?
- Higher clocks (likely, but how much)?
- Performance increase (maybe it will catchup to the R700)?
- Other improvements (I've read about 384 bit memory interface + GDDR5, but I doubt that)?

Oh, and according to Nordic Hardware it's scheduled for November, not September:

Skipping the rest of the bias, GT200b will not hit the market for another couple of months, most likely for the annual NVIDIA November refresh, so anyone buying a GT200 card will not end up with a dead duck.

http://www.nordichardware.com/news,7815.html

 

[nicolasb]

If the GT200b is so close, I wonder why there isn't more information available yet.

Fear of the Osborne Effect.

 

[Razor1]

Doesn't nVidia ship every chip in the first revision, unlike ATi?

hmm most of the time they don't, they seem to have better results at sending chips out with first revisions though.

 

[igg]

Fear of the Osborne Effect.

From Nvidias point of view, yes.

But GT200 related NDA info has been leaked since months, even here.

All I could find about the GT200b:
- According to CJ it's coming (he said so in this thread)
- Different release dates: Inq (Sep), Nordic Hardware (Nov), Fudzilla (2009)

 

[MfA]

It doesn't even outperform the GX2 in Crysis

I find nill AA/anisotropic low setting crysis benchmarks less than convincing ... I dunno what's going on there but I think it's by far the least relevant of the benchmarks.

 

[Arnold Beckenbauer]

http://www.tgdaily.com/content/view/37931/113/



...

The unveiling of the client is set for the next week as part of the launch of Nvidia’s GT200 GPU series. Owning such a card will have its benefits in Folding@Home and will outrun Radeon 3870 cards. The new GeForce cards are expected to hit more than 650 nanoseconds of protein simulation in a single day, while the Radeon HD 3870 is stuck at about 170 ns. The Playstation 3 is able to produce "only" 100 ns of simulation, while a quad-core CPU creates an output of just four nanoseconds. For those who are keeping count: The GeForce GPU will be about 163 times faster than a quad-core processor in this specific application

http://foldingforum.org/viewtopic.php?f=16&t=3117#p28792

PS Note that Tom's H numbers are a bit misleading. We're not getting 650ns/day (yet) on the gtx280 (more like 550) and we're now getting 250ns/day on 3870's in the lab (perhaps 300ns/day in time), and the 3870's are the previous gen ATI cards. The gtx280 is going to be really great at folding though.

Nanoseconds...

 

[Arun]

Awesome, given where TGDaily's data comes from this is rather annoying...

 

[igg]

Another review: http://www.hardware-aktuell.com/viewtest.php?article=51&seite=4
It's german but I'm sure everybody will understand the graph

 

[INKster]

Frankly, i have trouble believing that a "GT200b" would come out so soon.

November ? Maybe (would shoot for March 2009 myself -followed by D12U in November 2009-, but i'm not ruling the other out either).
September ? Somewhat unlikely.

Sure, the core is huge, but, as G80 proved before, it only sold in high margin products, so the production cost wasn't that critical as, say, a G92 in a 8800 GT.
Average power consumption seem to be at G80 levels so far, that's not it either (in fact, being able to idle with G94-levels of electricity is excellent for the GT200 core, methinks).

RV770 is not (and was never meant to be) a threat to the GTX 200 series, but to the current G92/G92b, so that's where i think Nvidia will aim their guns at next.
And G92 will be more than one year old come November, isn't that convenient ?