KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
Although it was expected that Nvidia Corp.’s upcoming GeForce GTX 980 Ti graphics adapter will be powered by the very same chip as the GeForce GTX Titan X, it now turns out that the new graphics board will be based on a version of the GM200 graphics processor with a different configuration of execution units.
The Nvidia GeForce GTX Titan X graphics card features the GM200-400 graphics chip with 3072 stream processors, 192 texture mapping units, 96 raster operations pipelines as well as 384-bit memory interface. The chip operates at 1000MHz/1075MHz base/boost clock-rate. By contrast, the upcoming GeForce GTX 980 Ti will use the GM200-310 graphics chip, reports VideoCardz web-site.
Nvidia marks graphics processing units with dissimilar configurations differently. For example, both GeForce GTX 970 and GeForce GTX 980 are powered by GM204 GPUs, but the number of activated SPs, TMUs and ROPs is different on these two graphics cards. As a result, the GTX 970 chips are marked as the GM204-200, whereas the GTX 980 processors are marked as the GM204-400.
The GM200-310 marking may indicate that the configuration of the graphics processor of the GeForce GTX 980 Ti will not be the same as that of Nvidia’s flagship graphics card. One thing that is clear at the moment is that the GeForce GTX 980 Ti graphics board will carry 6GB of GDDR5 memory.
Nvidia did not comment on the news-story.
Discuss on our Facebook page, HERE.
KitGuru Says: It is clearly logical for Nvidia to use a cut-down version of its GM200 for the GeForce GTX 980 Ti. The company has a lot of GM200 chips with defects and it needs to sell them somehow. At the same time, the company's product lineup requires a graphics solution that will be faster than the GeForce GTX 980, but slower than the GeForce GTX Titan X.
5.5Gb
GM200 and 204 are done corectly, it will not have 5.5/
6GB!
well the 204 was 3.5, how do you know they wont give 980 ti the same “feature”
Gb != GB
Because they know better.
plenty.
they faced a lawsuit, so I doubt they’ll even think of it as an option anymore.
kkkkk
no, 5,5GB! kkk
Ok !
Because its based similarly on the 980 which has full 4 gb, not the 3.5. Also the GM200-310 just means that it has newer tech that may allow full DX12 resource tiering, since only the 960 has this newer tech in it, while 970 and 980 are not full dx12 unlocked.
Geforce Cards use Cuda Cores. AMD cards use Stream Processors.
Plenty of hitching and stuttering
It’s fucking 6gb. You have to REALLY push to actually hit that.
Even the 980Ti won’t wet my appetite. No DisplayPort 1.2a or 1.3? NO THANKS!!
While there are some architectural differences, they are effectively the same thing. Nvidia used to call their cuda cores, stream processors. They changed name once the company moved toward marketing and leveraging the compute power of their stream processor by way of software initiatives.