Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Edit: if you mean which CPU specifically, probably the E7-8890 v4. 24-core Broadwell at 2.2 GHz (this won't be the AVX speed but meh) should be good for over 800 GFLOPS (DP). It's expensive though.![]()
if you mean which CPU specifically, probably the E7-8890 v4. 24-core Broadwell at 2.2 GHz (this won't be the AVX speed but meh) should be good for over 800 GFLOPS (DP). It's expensive though.![]()
SP is twice the DP rate. 32 per cycle per core, peak.
so the E7-8890 v4 can get to 1.6TFLOPS then?
do we have any sources for this?![]()
Are you seriously thinking of buying one? It's £6,500.
If you need that kind of muscle you're probably better off with several 2P systems in a cluster, and/or some Xeon Phis.
Why don't you explain what you're after?
Im trying to make a graph to compare single precision FLOPS of the best CPUs and GPUs year year from 2000-2016
OK. Hmm.
Theoretical GFLOPS have always been a bit meaningless, but it's become a lot worse recently now that the high-core-count parts will adjust their clock speeds a lot depending on the workload.
So while the E7-8890 v4 has a base clock speed of 2.20 GHz (which in theory would yield an incredible 1690 SP GFLOPS), it would almost certainly reduce its clock speed when handling a high performance AVX code.
Also using theoretical numbers isn't much good for showing the difference between say Haswell and Skylake. Both have the same 32/cycle, similar core counts and clock speeds, but Skylake has various improvements that give it better benchmark performance that you won't see in theoretical peak numbers.
Not sure what to recommend, sorry. Curious what you come up with though.