Specification of Nvidia GT212 Detailed

[drumzr]

Since ELSA revealed that GT212 would be another flagship card of NVIDIA, GT212 has been a focused topic. Being the top GPU this year, GT212 is coming to take the place of GT200(b) (such as GTX260/280) Series.

Our source indicates that the number of stream processers increases from 240 (GT200) to 384, and the number of TMUs also increases to 96. Such improvements are fairly significant, kind of like 128 SPs of G92 to 240 SPs of GT200. As the second-generation rendering structure, GT200 features 10 TPCs, and we’re wondering if GT212 would follow this structure. If so, GT212 will feature 16 TPCs, and there’re only 6 TMUs in each TPC, while there’re 8 TMUs in GT200. GT212 is very likely to adopt the third-generation rendering structure - each TPC contains four SMs and 8 TMUs, and there’re 12 TPCs, so there’re 12*4*8=384 stream processors in total.



Generally speaking, the stream processors of GT212 increase by 60%, but TMU just increases by 20%.

The memory interface of GT212 will reduce from 512bit to 256bit, which is quite similar to AMD RV770. GT212 will aslo use GDDR5 to make up the loss of memory interface. The memory clock frequency of GTX280 is 1107MHz. If the memory bandwidth of GT212 has to achieve that of GT200, its data rate must be above 4.5GHz.The new products are usually superior to previous ones in specification, so the data rate of GDDR5 is supposed to be around 5GHz. NVIDIA and AMD both use GDDR5 on their middle-end lineup, which makes us to worry about the productivity of GDDR5.

As to processing technology, GT212 undoubtedly features 40nm processing. Because of the sharp increase of stream processors, the number of transistors reaches 1.8 billion, while that of GT200 is 1.4 billion. Thanks to the new processing, the core die area is controlled within 300mm^2, and the power consumption can be also reduced greatly.

GT212 is scheduled to be launched in Q2 this year. As the coming flagship product from NVIDIA, it’s really worthy of our expectation.

( expreview )

[Wingit]

IMHO, this specification sounds reasonable. Using GDDR5 rated at 1250MHz, it would result in a bandwidth of 160GB/s. Im guessing that its shader domain is clocked at > 1.5GHz effectively making it 60% faster than the GTX285.

[opaper]

Expected in the second quarter of 2009, the flagship of the range, the chip NVIDIA GT212 will be the first engraved in 40 nm, but its other technical features were previously not known until the German online magazine News Hardware not publish today.

Carrying 384 stream processors, the GT212 should worthy successor to GT200, used by graphics cards GeForce GTX 280, which is content for its 240 units of calculation, an increase of 60%. The number of units of texture and transistors are also increasing, from 80 to 96 for the first, between 1.4 and 1.8 billion respectively. Earnings of fine engraving, the surface of the chip is greatly reduced from 576 mm ² on the GT200 engraved in 65 nm or 470 mm ² engraved in 55 nm to less than 300 mm ². Only flat this new chip, the width of the memory bus is reduced to 256 bit, 512 bit against before, but the benefit of the use of GDDR5 instead of GDDR3.

[Aikonik]

The first details on upcoming NVIDIA mobile chips were announced today by our colleague Anglophone VR-Zone. The entire range will suffer what appears to be a simple transition (die shrink) of 65 nm to 55 nm in the first half of 2009. The transition to 40 nm occur in the second half of 2009 according to a scheme based on the range of desktop chips. Different variations of a chip code-named N10E (E enthusiast) and based on the GT212 and GT215. Declensions of N10P (with the performance of P) are in turn based on the GT216. With entry-level finally, N10M (for mainstream) will use the GT218. Future NVIDIA mobile chips will follow the new nomenclature of naming when the chips are entitled Office and is called 180M/170M GeForce GTX, GTS 160M, GT 130M/120M or G 110M/105M.

[Golfert]

These choices ATI nvidia confirm that we just saw with the 4870: gddr5, bus 256-bit memory and technology to produce chips smaller and economically (for them, at least)

[Howerski]

These choices ATI nvidia confirm that we just saw with the 4870: gddr5, bus 256-bit memory and Aanzee technology to produce chips smaller and economically (for them, at least)

already ... the 512-bit bus I see it as too costly a monster with few benefits and performance... are doing some back, so good, the GPU to 40nm should be powerful

If the frequencies are fluctuating could go twice a current gtx280

for the bus if 256bit will do a few 384 and we do not talk, even to support 1.5GB of ram, with 256 because I think you do, or 1 or 2 GB, I do not think are possible middle way.