<html><head><meta http-equiv="Content-Type" content="text/html charset=iso-8859-1"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">ok, I forget that<div><br><div><div>On Feb 19, 2013, at 7:50 PM, Eliot Miranda <<a href="mailto:eliot.miranda@gmail.com">eliot.miranda@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><br>Hi All,<br><br> kudos to Nicolas for posting some useful numbers in that they<br>provide some context, in this case the other VMs running on the same<br>machine. But wrist slaps to all of you for not specifying:<br><br>1. which OS<br></blockquote><div><br></div><div>osx 10.8</div><br><blockquote type="cite">2. what hardware<br></blockquote><div><br></div><div>i7 8gb</div><br><blockquote type="cite">3. what C compiler was used to compile the VM<br></blockquote><div><br></div><div>4.6.3</div><br><blockquote type="cite"><br>Further kudos for indicating what kind of load the machine is under<br>(one has to run benchmarks on a relatively unstressed machine, even if<br>multicore), and, *really usefully*, what a previous version's<br>benchmark score is on the same machine.<br></blockquote><div><br></div><div>no matter the load. </div><div>It is a comparative analysis: stackvm pre-merge with latest: 500 msends, after merge: 5msend... since I did not changed my mac for an ipad, the sends cannot be right. Also... 10% passive cpu usage is wrong, no matter the machine load. </div><div><br></div><br><blockquote type="cite">Nicolas' results, Cog ~= 6.5x Interpreter, Stack ~= 1.75x Interpreter<br>are exactly what one should expect for nfib (the sends/sec part of<br>tinyBenchmarks) with the current Cog architecture.<br></blockquote><div><br></div><div>yep... so where is the stack vm made with latest sources? (it is not in <a href="http://www.mirandabanda.org/files/Cog/VM/VM.r2678/">http://www.mirandabanda.org/files/Cog/VM/VM.r2678/</a>)</div><br><blockquote type="cite"><br>On Sun, Feb 17, 2013 at 9:21 AM, Nicolas Cellier<br><<a href="mailto:nicolas.cellier.aka.nice@gmail.com">nicolas.cellier.aka.nice@gmail.com</a>> wrote:<br><blockquote type="cite"><br>This is not confirmed in regular svn cog branch<br><br>1 tinyBenchmarks<br> '380 669 144 bytecodes/sec; 10 473 620 sends/sec' Interpreter VM<br> '371 014 492 bytecodes/sec; 18 512 525 sends/sec' Stack VM<br> '656 410 256 bytecodes/sec; 67 802 547 sends/sec' Cog VM<br><br>Nicolas<br><br>2013/2/15 Igor Stasenko <<a href="mailto:siguctua@gmail.com">siguctua@gmail.com</a>>:<br><blockquote type="cite"><br>On 15 February 2013 14:49, Esteban Lorenzano <<a href="mailto:estebanlm@gmail.com">estebanlm@gmail.com</a>> wrote:<br><blockquote type="cite"><br>Hi,<br><br>I just compiled a cog vm and a stack vm with latest sources.<br>While the cog works well, the stack vm is having a serious performance drawback, tinyBenchmarks is giving me 5m bytecodes/s, while it should be around 500m... also I see the cpu charge increases to double.<br><br>Do you tried to compile a stack vm lately? any idea where to start look for bugs?<br><br></blockquote><br>just tried on my machine.. the results is discouraging:<br><br>1 tinyBenchmarks '4648460 bytecodes/sec; 337199 sends/sec'<br><br><blockquote type="cite">thanks,<br>Esteban<br></blockquote><br><br><br>--<br>Best regards,<br>Igor Stasenko.<br></blockquote></blockquote><br><br><br>-- <br>best,<br>Eliot<br></blockquote></div><br></div></body></html>