NVIDIA Likely to Start 40nm GPU Conquest with GT218

[btarunr]

As the industry moves closer to implementation of the 40nm silicon technology on graphics processors, some of the most complex electronic devices, both AMD and NVIDIA have their offerings in the works. While AMD already has prototypes taped out, NVIDIA seems to be taking its time, perhaps to perfect its products.

The green camp, as Hardware-Infos finds out, is likely to start its 40nm lineup with the GT218 graphics processor. With NVIDIA's internal coding scheme, the GT218 is an entry-level GPU. In its generation hierarchy, would be the GT215, GT216 (mainstream) and GT212 (high-end). The GT218 will succeed the G98 GPU that went into making the GeForce 9300 series. It will have 32 stream processors, 8 texture memory units, and 4 raster operations pipelines. It will continue to have a 64-bit wide memory bus. Despite the increase in transistor counts, the GPU ends up with a lower TDP, expected to be around 22W, as against the roughly 30W for the G98. The GT218 is slated for April, later this year.



View at TechPowerUp Main Site

 

[Selene]

bring on teh GT212 already, who cares about the GT218.
Normaly we get the top card first, then the lowend ones, not sure why its going the other way this time.
IF they wait to long on the highend card, the GTX300s will be to close behind to bother with.

 

[Weer]

They care. Because they're going to make more money off of the GT218 than the GT212.

 

[spearman914]

That's some low power consumption!! Can't wait till GT300.

 

[Jakl]

40nm now eh? I wonder how long till they bring out lower size chips than that from 65nm...

 

[DarkMatter]

Doesn't seem like a big improvement overall. 32 shaders will make this card a good second for PhysX though, maybe that's their goal.

 

[alwayssts]

I think it's so that when the GT212, 384sp/2TF, launches, they can say that have a product that is 12x faster than it's entry level, with every space filled in the center, including the presumably 6x faster GT216 (192sp) and 9-10x faster GT212gt (288-320sp?)...with maybe a nice little gt214 (96sp?) thrown in for good measure. That could be quite a well-spaced stack...in theory, ranging from 32-384sp, 8-96TMU, 64-bit - 384-bit, 14.4gbps-192gbps, 150gflops-2TF, and most importantly...chip sizes from 8^2 (64mm2) to 19^2 (361mm2).




Ah, I dream of a simplified product stack. How lovely it would be.

 

[Hayder_Master]

nvidia try catch ATI in GPU technology

 

[a_ump]

i think they're starting with entry level this time so that when they produce their flagship they can use 40nm with most if not all bugs figured out from producing a smaller chip(GTX218) first then moving up in chip size with the process, instead of starting out their flagship with a common manufacturing process or soon to be outdated process.

 

[DarkMatter]

nvidia try catch ATI in fab technology

Fixed. GPU technology?? are you serious?

 

[waver_01]

Fixed. GPU technology?? are you serious?

seeing how it's true, I hope he is

 

[FordGT90Concept]

NVIDIA actually might get to a lower process than Intel is at? That in itself is remarkable.

 

[btarunr]

NVIDIA actually might get to a lower process than Intel is at? That in itself is remarkable.

It's AMD before NVIDIA and Intel, but yes. Intel will be demonstrating 32nm SSDs at CeBIT though.

 

[DarkMatter]

To be more precise TSMC is, but yeah both Ati and Nvidia are apparently going to have a chip in a smaller fab process than 45nm before Intel. Probably Ati a little bit sooner.