<html><body>
<DIV> </DIV>
<BLOCKQUOTE style="PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #1010ff 2px solid">
<P>-------------- Original message -------------- <BR>From: "Mikhail Kuzminsky" <kus@free.net> <BR><BR>> In message from richard.walsh@comcast.net (Fri, 12 Oct 2007 20:50:08 <BR>> +0000): <BR>> >Mikhail, <BR>> >I am not sure I fully understand what you are presenting here, but I <BR>> >might say that yes, at the FPU unit level the 2222 series AMD <BR>> >Opteron/Barcelona and the Intel Core2/Clovertown (and also Harpertown <BR>> >at 45 nm) are now more largely equivalent -- that is they both can <BR>> >execute 2, double-wide (2x64 bit) floats in certain FMA situtations <BR>> >simultaneously and/or in a pipeline. <BR>> .... <BR>> >Regards, <BR>> >rbw <BR>> >-- <BR>> Sorry, now I'm misunderstanding :-) <BR>> I thought that Opteron 2222 don't have Barcelona microarchitecture <BR>> (all the Barcelona's are 23xx or 83xx) and therefore can't perform <BR>> 4*64 bit FP results per cycle. Am I wrong ? </P>
<P>Mikhail,</P>
<P>Whoops ... ;-( ... I should have said 23XX. That is the Barcelona generation. If you were talking about generation two, then what I said about a 64-bit serialize staging of the 128-bit wide SSE FPU operations applies. Sorry, if I caused confusion. My comments regarding near-floating-point functional unit equivalency should be confined to comparing the 23XX series Opteron with the Clovertown/Harpertown processors from Intel. The point about 3-wide versus 4-wide applies with all Opteron generations.</P>
<P>Dobray Utra,</P>
<P>rbw</P></BLOCKQUOTE></body></html>