Friday, May 1st 2015

NVIDIA GeForce GTX 980 Ti Silicon Marked "GM200-310"

NVIDIA's upcoming high-end single-GPU graphics card, based on the GM200 silicon, which debuted with the GTX TITAN X, will feature a silicon marked "GM200-310." The SKU will be named GeForce GTX 980 Ti, and is more likely to be priced around the $600-650 mark, than replacing the $550 GTX 980 off the shelves. Going by the way NVIDIA re-positioned the GTX 780 to $499 with the introduction of the GTX 780 Ti, we imagine something similar could happen to the GTX 980. From what we gathered so far, the GTX 980 Ti will be based on the GM200 silicon. Its CUDA core count is unknown, but it wouldn't surprise us if it's unchanged from the GTX TITAN X. Its different SKU numbering shouldn't be an indication of its CUDA core count. GTX 780 Ti and GTX TITAN Black had different numbering, but the same CUDA core counts of 2,880.

The card will feature 6 GB of GDDR5 memory across the chip's 384-bit wide memory interface. It will feature five display outputs, similar to that of the GTX 980. Unlike with the GTX TITAN X, NVIDIA partners will have the freedom to launch custom-design GTX 980 Ti products from day-one. There are two theories doing rounds on when NVIDIA plans to launch this card. One suggests that it could launch in mere weeks from now, probably even on the sidelines of Computex. The other suggests that it will launch towards the end of Summer, as NVIDIA wants to make the most cash from its existing GTX 980 inventory.
Source: VideoCardz
Add your own comment

43 Comments on NVIDIA GeForce GTX 980 Ti Silicon Marked "GM200-310"

#26
HumanSmoke
INSTG8RWell I expect at least a "reveal" I mean why host a "PC Gaming Show" If your not gonna bring some new toys to play with. They certainly can't be promoting the 290X...
It might be a let down in any case, just as GM200 was " just" a scaled up GM204, it looks increasingly likely that Fiji is " just" a scaled up Tonga - Hence my comment above about GM200 and Fiji feeling like stopgap measures (or proof of concept at best), and the real cause for excitement being Pascal and Arctic Islands.
Posted on Reply
#27
bogami
All together more and more scent marketing to move. (Both want to keep using primarily the price for outgoing 28nm production which is becoming cheaper than anything else, what we see in AMD which lowered the price and the delay in 20 nm technology) are thus again thrown sand in the eye PLL. For thick profit.
It shows that the nvidia wait what AMD can currently offer in 20nm and then released her version of 20nm or are major problems?, were there again ' profit marketing tactic? We'll see.
Somehow not escape the feeling that nVidia is becoming thick pig who it must be maintain from the poor population vith low tricks . Horror where there are times when we get the strongest 4 GPU option for € 1,200 and now is 1 card is so expensive. Horror same manipulation : Market ,information ,people, ITD. They fell so low.
Posted on Reply
#28
64K
bogamiAll together more and more scent marketing to move. (Both want to keep using primarily the price for outgoing 28nm production which is becoming cheaper than anything else, what we see in AMD which lowered the price and the delay in 20 nm technology) are thus again thrown sand in the eye PLL. For thick profit.
It shows that the nvidia wait what AMD can currently offer in 20nm and then released her version of 20nm or are major problems?, were there again ' profit marketing tactic? We'll see.
Somehow not escape the feeling that nVidia is becoming thick pig who it must be maintain from the poor population vith low tricks . Horror where there are times when we get the strongest 4 GPU option for € 1,200 and now is 1 card is so expensive. Horror same manipulation : Market ,information ,people, ITD. They fell so low.
Nvidia is doing what all successful enterprises do. They charge what the market will bear. They do whatever they can to ensure profit for their investors/owners just like all successful businesses do.
Posted on Reply
#29
radrok
64KNvidia is doing what all successful enterprises do. They charge what the market will bear. They do whatever they can to ensure profit for their investors/owners just like all successful businesses do.
Which is a smart business path :)

I am pretty sure their next GPU will be a 6GB non cut down variant and we'll probably see a dual GPU earlier than expected, hopefull for a sane price this time around.
Posted on Reply
#30
arbiter
bogamiIt shows that the nvidia wait what AMD can currently offer in 20nm and then released her version of 20nm or are major problems?
20nm was a flop, amd fiji will be using 28nm.
Posted on Reply
#31
radrok
arbiter20nm was a flop, amd fiji will be using 28nm.
I read somewhere that 20nm was never meant for big chips, only for small socs etc?
Posted on Reply
#32
HumanSmoke
radrokI read somewhere that 20nm was never meant for big chips, only for small socs etc?
That would be correct. Ever since TSMC canned its CLN20G process in 2012, there hasn't been a viable 20nm process for large die/large power budget IC's. Even TSMC's own slides confirm CLN20SOC being low power optimized



...and as for Globalfoundries, they made no secret of the fact that 20nm LPM stands for Low Power Manufacturing, and moreover, GloFo's process page which I linked to here, has now disappeared entirely, nor that the company pretty much jettisoned it in favour of pouring funds into licence manufacture of Samsung's 14nm-XM process.
...and yet, some people just seem content to bury their heads in the sand and write their own narrative completely at odds with known fact and the foundries own literature.
Posted on Reply
#33
looniam
Captain_TomI like how everyone says that he 980 Ti will be "Cut down without the memory problem." Everyone here does realize that the architecture of Maxwell makes that inherently impossible right?

The memory and cores are directly linked. Disable any of the cores and a proportional amount of the memory will need to be disabled as well - which besides being wasteful, isn't that bad as long as they don't lie about it.
no they're not . . the 970 had L2 cache disabled for binning.
Posted on Reply
#34
HumanSmoke
looniamno they're not . . the 970 had L2 cache disabled for binning.
I think it was just a lazy attempt at trolling the thread TBH. Even a casual glance at the GM204 SKU's would show that the (for instance) GTX 980M with 1536 cores (or 128 fewer than the GTX 970) still maintains a 256-bit/ 2MB L2 cache/64 ROP specification of the full GPU (and 4 or 8GB of vRAM). Maybe Captain_Tom should do some further reading before putting forward his deeply flawed observation, which would then show that memory allocation is dependant upon the number of memory controllers, 1 (or 2 if in clamshell mode) memory IC's per 32-bit memory channel (two 32-bit memory channels per 64-bit memory controller) . Memory size then being defined by IC density.

@looniam
I don't know if the GTX 970's issue was what Captain_Tom was driving at, although you're right that the 970's logic segmentation caused the issue you're talking about- reducing ROPs and their attendant L2 without proportionally reducing the memory controllers as Nvidia did with the 970M and 965M.
Posted on Reply
#35
bogami
64KNvidia is doing what all successful enterprises do. They charge what the market will bear. They do whatever they can to ensure profit for their investors/owners just like all successful businesses do.
Are you very rich or oa blind .it is not a success but manipulation.
Posted on Reply
#36
matar
Now is the time Nvidia should fix the GTX 970 with a GTX 970Ti now that card at the same price $329 I would Buy.
Posted on Reply
#37
arbiter
bogamiAre you very rich or oa blind .it is not a success but manipulation.
Well Which of the 2 companies AMD/Nvidia is strong and which is pretty close to going under?
Posted on Reply
#38
64K
bogamiAre you very rich or oa blind .it is not a success but manipulation.
Neither but I can look at financial statements for the truth. Nvidia's net profit margins are average for the industry with the exception of AMD which is drowning in red ink.
Posted on Reply
#39
rtwjunkie
PC Gaming Enthusiast
Captain_TomI like how everyone says that he 980 Ti will be "Cut down without the memory problem." Everyone here does realize that the architecture of Maxwell makes that inherently impossible right?

The memory and cores are directly linked. Disable any of the cores and a proportional amount of the memory will need to be disabled as well - which besides being wasteful, isn't that bad as long as they don't lie about it.
You do realize this isn't a Maxwell problem? It was a GM204 problem when they disabled cores/memory controllers. THAT chip only. It is not an inherent problem in the other GM chip monikers, or at least no evidence has been submitted to support that assertion. Unless you have inside engineering knowledge of Nvidia's HQ.......?.

@bogami While I do appreciate your point, businesses are not in business to cater to people's ability to pay. That viewpoint is directly from an economic system that has proven it can only fail, wherever implemented (Socialism, for those that need help) -"From each according to his ability, To each according to his need (ability to pay)." This was popularized by Karl Marx, but originated with Louis Blanc, the French socialist in 1851.

I guarrantee if you had a business, you would charge what people are willing to pay. Why? You have bills to pay, and maybe a family to support, and would like to move up someday. They are a business, not a non-profit agency. All successful businesses charge what people are willing to py. Those that charge too little (sales and special events notwithstanding) to cover all their necessary research, development, production, personnel and marketing expenses do not profit, and eventually go under.
Posted on Reply
#40
Dalai Brahma
For me, a 960Ti 3/4GB 192bit (< $250) would be enough... and I would be so glad.
Release it, Nvidia!! and cut prices!
Posted on Reply
#42
arbiter
64KWe may have a reveal of the GTX 980 Ti coming in 2-3 weeks and possibly it will go on sale in stores right after that if the rumor about the 2015 Computex Trade Show is correct.

www.kitguru.net/components/graphic-cards/anton-shilov/nvidia-may-announce-geforce-gtx-980-ti-at-computex-next-month/

I figured the R9 390x would drop first but maybe not.
Nvidia will do some damage if they put that at 6-700$ if the recent rumors that 390x 4gb yes 4gb will start at 850$.
Posted on Reply
#43
64K
More rumors that the GTX 980 Ti will preempt AMD's Fijis.

videocardz.com/55566/nvidia-geforce-gtx-980-ti-performance-benchmarks

According to this site review sample 980 Ti have already been sent out and it looks like the rumor that it's going to be a cut down GM200 is being reinforced and it will probably be a little slower than a Titan X. My guess is that this can be made up for by increased overclocking potential due to superior non reference coolers which will be available.
Posted on Reply
Add your own comment
Aug 23rd, 2024 15:26 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts