By using this site, you agree to our Privacy Policy and our Terms of Use. Close
Soundwave said:
HoloDust said:

Actually, GT7900 is rated at around 800GFLOPS in FP32, and double that in FP16 (PS4 is 1.84TFLOPS FP32)...but more or less, I see what you're getting at, so I won't jump in anymore.

You wouldn't happen to know how many watts the GT7900 consumes in FP32 would you? 


No idea really. I know that Tegra X1 uses 10W for its peak performance of 512GFLOPS (FP32).

For example, X1 runs Manhattan with around 1.5W vs 2.67W of A8X for equal performance, both on 20nm. So I'm really interested to see what Imagination's Series 7XT are pulling at 14/16nm vs nVidia.