<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote">On Wed, Jun 21, 2017 at 6:29 PM, Christopher Samuel <span dir="ltr"><<a href="mailto:samuel@unimelb.edu.au" target="_blank">samuel@unimelb.edu.au</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">I thought it interesting that the only performance info in that article<br>
for Epyc were SpecINT and (the only mention for SpecFP was for Radeon).<br></blockquote><br></div><div class="gmail_quote">As did I, but a little digging shows a STREAM benchmark (on AMD's page) showing +25% performance of a single-socket Epyc vs. a dual-socket E5-2690 v4 Broadwell system[1], and roughly 60% better specfp_rate2006 numbers when comparing socket-to-socket[2]. When I read stuff like this, I feel a little bit like Charlie Brown going to kick the football, and <i>hoping</i> it's not going to get whisked away by Lucy... but right now, this looks pretty impressive, especially given the memory bandwidth issues on Xeons on some codes.<br><br> [1] <a href="http://www.amd.com/system/files/2017-06/AMD-EPYC-SoC-Delivers-Exceptional-Results.pdf">http://www.amd.com/system/files/2017-06/AMD-EPYC-SoC-Delivers-Exceptional-Results.pdf</a><br> [2] <a href="http://www.amd.com/system/files/2017-06/AMD-EPYC-SoC-Sets-4-World-Records.pdf">http://www.amd.com/system/files/2017-06/AMD-EPYC-SoC-Sets-4-World-Records.pdf</a><br><br></div><div class="gmail_quote"> Since those are both AMD links, and not independent, take a grain of salt with them, but I harbor some optimism, and would love to get my hands on one. We'll see. <br><br></div><div class="gmail_quote"> Cheers,<br></div><div class="gmail_quote"> - Brian<br><br></div><div class="gmail_quote"><br></div></div></div>