AnsweredAssumed Answered

Nvidia GTX 1090 Ti & AMD R9 495X Benchmarked & Reviewed on April 1

Question asked by kingfish on Apr 1, 2016
Latest reply on Apr 1, 2016 by black_zion

"Both featuring the world’s most advanced FinFET Graphics Processing Units boasting the highest levels of performance we’ve seen yet."

First off we have the GTX 1090 Ti, Nvidia’s latest Pascal flagship. Featuring a full fledged GP102 GPU with 30 SPMs, Streaming Pascal Multiprocessors, for a total of 3840 CUDA cores. In addition to a 256bit GDDR5X memory interface, with 7.5GB ( don’t ask why ) of 12Ghz GDDR5 memory for a total of 384GB/s of memory bandwidth.

"The GTX 1090 Ti operates at base clock of 1212Mhz and a boost clock of 1372Mhz. Making this is by far the highest clocked graphics card Nvidia has ever introduced. The GP102 GPU is the most complex chip the company has ever made packing more than 10 billion transistors.  Nvidia is positioning this card as a direct replacement to the GTX 980 Ti. However, thanks to the next generation FinFET process the GP102 GPU is significantly smaller than GM200 and significantly more power efficient. Additionally, thanks to its smaller size Nvidia can price it more aggressively. Which brings us to the most exciting part about this new product and that is its price,. At $499 the 1090 Ti is not only faster, more power efficient and smaller but also cheaper."

"Moving on to the Red Team. This is the first time that we’ve ever got to see the R9 495X. This is the company’s flagship single GPU 400 series Polaris based. It Features a Polaris GPU that we haven’t heard of before today, called Polaris 12. Polaris 12 features 5120 GCN 4.0 cores and 256bit wide GDDR5X memory interface feeding 8GB of 12.0Ghz GDDR5X memory chips. The result is 384GB/s of memory bandwidth, exactly the same as Nvidia’s GTX 1090 Ti. So it seems both company’s decided to opt for smaller memory interfaces for their high-end designed, off-set by the significantly faster GDDR5X memory from Micron."