<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">2014-02-03 Reinhard Tartler <span dir="ltr"><<a href="mailto:siretart@gmail.com" target="_blank">siretart@gmail.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div class="im">On Mon, Feb 3, 2014 at 2:01 PM, Jaromír Mikeš <<a href="mailto:mira.mikes@gmail.com">mira.mikes@gmail.com</a>> wrote:<br>
<br>
> Please go ahead ... It looks like debian-devel would be right place.<br>
> It sound strange that sse is redundant in amd64.<br>
<br>
</div>Please elaborate, why is this strange?<br>
<div class="im"><br>
> Some time ago I played with sse optimization flags and packages build with<br>
> and without them<br>
> got 30-40% difference in performance in my cases.<br>
><br>
> Would be great to make it clear.<br>
<br>
</div>Unfortunately not. Please clarify what exact packages did you compile<br>
with what additional see flags on what architecture to get the 30-40%<br>
difference.<br></blockquote><div><br></div><div>I did test with jconvover ... I had to dig in old emails exchanged with Free Ekanayaka.<br></div><div>I was 4 years ago! ;)<br>Used flags -mmmx -msse -mfpmath=sse<br></div>
<div>Improvement in performance wasn't so huge ... it was about 15% ... sorry for my bad memory. <br></div><div>Testing also haven't been very scientific just running as much as possible instances of jconvolver until first xruns in Jack. <br>
</div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Yes, this can make a huge difference, but using those flags is like<br>
firing a shotgun: it may solve some symptoms, but is likely to cause but is likely to cause<br>
collateral damage.</blockquote><div><br></div><div>Sure, that's why I would like to know more.<br>Not make big disaster instead of little improvement.<br><br></div><div>mira<br></div></div></div></div>