Friday, January 2nd 2015

Possible NVIDIA GM200 Specs Surface

Somebody sent our GPU-Z validation database a curious looking entry. Labeled "NVIDIA Quadro M6000" (not to be confused with AMD FirePro M6000), with a device ID of 10DE - 17F0, this card is running on existing Forceware 347.09 drivers, and features a BIOS string that's unlike anything we've seen. Could this be the fabled GM200/GM210 silicon?

The specs certainly look plausible - 3,072 CUDA cores, 50 percent more than those on the GM204; a staggering 96 ROPs, and a 384-bit wide GDDR5 memory interface, holding 12 GB of memory. The memory is clocked at 6.60 GHz (GDDR5-effective), belting out 317 GB/s of bandwidth. The usable bandwidth is higher than that, due to NVIDIA's new lossless texture compression algorithms. The core is running at gigahertz-scraping 988 MHz. The process node and die-size are values we manually program GPU-Z to show, since they're not things the drivers report (to GPU-Z). NVIDIA is planning to hold a presser on the 8th of January, along the sidelines of the 2015 International CES. We're expecting a big announcement (pun intended).
Add your own comment

80 Comments on Possible NVIDIA GM200 Specs Surface

#76
HumanSmoke
Sony Xperia S3DC info is old enough and we have new piece of data which changes initial plans.
Maybe initially Fiji had indeed been scheduled for production on 28nm with 4096 shaders, but afterwards it could be forward-ported to a more advanced manufacturing process, 20nm at GloFo.
No doubt the design can be ported to a 20nm based process in future (although it seems unlikely with 16nmFF/14nm-XM not too far off), but I still haven't heard a single logical explanation why AMD would go from building large GPUs on 20nm (and if they are being benchmarked now production started at least a quarter ago) then move to 28nm SHP this year. Seems ass-backwards.

What makes it even more unlikely is that AMD will be unveiling the Carrizo APU with the same GPU architecture using 28nm (very likely at this weeks CES). Wouldn't it make more sense to consolidate APUs and GPUs on the same process when the APUs use the same graphics cores as the discrete GPUs?
Sony Xperia SIn theory it would be ok but in practice, to me, releasing anything 28nm (even GM200) is purely a short-vision decision.
Graphics cards have a short shelf life. You build on the processes suited for the task and readily available.
Sony Xperia Sthe more likely those will use either 20nm or 16nm. :rolleyes:
Unless AMD are really late to the party, I think you're setting yourself up for disappointment. The time frame for Fiji (its test and validation boards being sighted on Zauba some months back) almost certainly seems to point to 28nm IMO. Bermuda hasn't been sighted, so if it's a further development for a later launch then a smaller node is a definite possibility.
Posted on Reply
#77
Lionheart
xenocideYou miss the days of Seronxadamus predicting AMD's new CPU's performing 60% better than Intels with perfectly linear scaling and everyone in the software design industry magically coding for 32-threads? The news posts have always been hit or miss, this one is just especially silly.
Uum no! o_O Got nothing to do with just AMD & Intel, I'm referring to the way ppl had discussions on this site, ppl actually helped each other without resorting to a fanboyish mindset, and there were rarely any trolls as well. But I'm going back to around 2008 - 2011....
Posted on Reply
#78
Sony Xperia S
HumanSmokewhy AMD would go from building large GPUs on 20nm
No, because Bermuda won't be so large in the first place.
HumanSmoke(and if they are being benchmarked now production started at least a quarter ago)
Not necessarily. You had a benchmarked engineering sample.
HumanSmokethen move to 28nm SHP this year.
I am still not quite sure about this.
HumanSmokeGraphics cards have a short shelf life.

Unless AMD are really late to the party, I think you're setting yourself up for disappointment.
No.
No.

My finances would be actually quite relaxed to wait till real 20nm or 16nm products. Short shelf life doesn't equate to short service in consumers' cases.
Posted on Reply
#79
HumanSmoke
LionheartUum no! o_O Got nothing to do with just AMD & Intel, I'm referring to the way ppl had discussions on this site, ppl actually helped each other without resorting to a fanboyish mindset, and there were rarely any trolls as well. But I'm going back to around 2008 - 2011....
Just for comparison, here's a direct comparison of the same situation ( Nvidia top-tier GPU release rumour) from 2009- the fanboy accusations started flying inside the first 2 pages.

EDIT:
Sony Xperia SNot necessarily. You had a benchmarked engineering sample.
HumanSmoke(and if they are being benchmarked now production started at least a quarter ago)
Fiji XT boards began shipping two months ago at least

From tape out to finished silicon takes 8-12 weeks ( Tape out > fabrication > die cutting > chip runtime testing > die packaging > board assembly). Eight weeks prior to 7 November makes it four months - a quarter of a year.
Posted on Reply
#80
Sony Xperia S
HumanSmoke...
The problem with your theory is that if Fiji has been shipping for so long time, there is still no physical evidence or proof, or even a slight hint about coming launches.

So, when will it be released? :D
Posted on Reply
Add your own comment
May 11th, 2024 23:47 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts