Re:The Supercomputer In Your Pocket: Metal and Swift
by stephencanon on Friday May 29, 2015 @04:42PM
(#9627236)
"LINPACK" (without qualification) is a double-precision benchmark, and doesn't measure raw compute flops, but what is actually achieved solving a linear system. K1 does double-precision at a rate of 1/24, and good libraries typically achieve ~90% of rpeak on similar hardware, so something around 13.7 gflops is a reasonable estimate.
[ Parent
| on HN ]
There are no comments on this comment yet.