JRPGfan said:
But again thats FP16 and not 32.... 2.3 Teraflops (FP16) from 2 x X1's, would be about 1.15 Teraflops of FP32 probably... and those X1's in the android TVs use more than 20 watts of power. So your looking at 1.15 Teraflops (FP32) @40watts+ (from 2 x X1) vs PS4 1.84 Teraflops (FP32) @90? watts.
also you know that X1 memory bandwidth is much much lower right? It cant do 176 GB/s memory bandwidth, without serious rework to the design and memory channels ect that the chip has. Which would make the chip draw more power than it already does. (or have really bad memory bandwidth and not be able to run 1080p) Plus the CPU inside the PS4, is faster than the CPU inside the X1 (I think).
MortienGerrux: read the first few lines of the article you linked:
Its FP16 (which is much easier to do than FP32). PS4 does FP32 when it says 1.84 Teraflops. |
Thanks for the explanation. Still pretty impressive and shows that mobile gaming in smartphones and tablets (maybe even super small consoles) has potential for the future. Don't forget a Tegra P1 is coming soon based on Pascal.







