<br><br><div class="gmail_quote">2010/12/13 Robin Gareus <span dir="ltr"><<a href="mailto:robin@gareus.org">robin@gareus.org</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div><div></div><div class="h5">
>> it is much easier to screw up a<br>
>> kernel by blindly applying patches and generating a .config if you don't<br>
>> know what you're doing (and sometimes even if you know what you're<br>
>> doing). Also, one needs additional tools (like rtirq) to make good use<br>
>> of RT-linux, while -bfs runs OOTB.<br>
>> However these days most distributions do that setup for the users.<br>
>><br>
><br>
> well, I always followed<br>
> <a href="http://www.alsa-project.org/main/index.php/Low_latency_howto" target="_blank">http://www.alsa-project.org/main/index.php/Low_latency_howto</a> and never<br>
> screwed up a kernel<br>
<br>
</div></div>Lucky you :) The information on the alsa-wiki is very good, but not<br>
every user is diligently following instructions as you do.<br>
It is also much easier to make a custom kernel that runs on one (your)<br>
system only, than tackling the task of compiling one that can be<br>
distributed.<br></blockquote><div><br>let's say I learned to be careful the day after I forgot to put the --initrd option in a make-kpkg command line statement and got a kernel panic after rebooting my first brand new... not-working-at-all-kernel :)<br>
</div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="im">>> I don't care so much about speed. The important issue in pro-audio is<br>
>> reliability. It's not the smallest possible latency that counts, but the<br>
>> max. latency of the system.<br>
>><br>
><br>
> I really did not understand this statement and anyway I would not agree...<br>
> why anybody should be safe knowing that his box max latency is 20ms instead<br>
> of 50ms or 70ms?<br>
<br>
</div>The max. system-latency determines the audio-latency at which you will<br>
never have any x-runs. (The minimal and avg. latency determines the<br>
speed and reactivity of your Desktop.)<br>
<br>
Without the RT-linux patch you may be good at 99.9% of the time, but<br>
since there is no guarantee for system-latency: there may be drop-outs.<br>
<br>
And Murphy says that this 0.1% chance will always become real when<br>
you're on stage in the middle of a performance. The resulting click will<br>
not only kill the PA but half the audience will sue you for becoming<br>
deaf ;-)<br></blockquote><div><br>you're right but as a statistician I prefer to call it <b>law of large numbers</b>... same results but more complicated to explain :)<br><br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
An other use-case is a studio: It's not about low-latency there but<br>
about no x-runs at all, they are just unprofessional. YMMV.<font color="#888888"><br></font></blockquote><div><br>I agree about unprofessionality of xruns but what about our ears?<br>I trust my ears and bless them, thanks to them I can appreciate music from Ella to Jimi and so on... but do they betray me when I realize I can't recognize an xrun in a mix? How long is an xrun? Can we hear it when it occurs only 3,4,5 times in a mix?<br>
I guess I am going off topic now but I am a fan of "if it's sounds good then it's ok"<br></div></div><br>Regards<br><br clear="all"><br>-- <br><i><span></span><span>L'unica speranza di catarsi, ammesso che ne esista una,
resta affidata all'istinto di ribellione, alla rivolta non isterilita
in progetti, alla protesta violenta e viscerale.</span></i><br>