[Ardour-Users] The future?

[Brett Clark]

I researched RT OS's a few years back and it seems that performance hits
doing regular tasks is "par for the course".  The reasoning behind it was
that the kernel is forced to handle interrupts within a specified period of
time as opposed to being able to stack the interrupt until it reaches a more
optimal time to handle it.  This means that all interrupt handlers have to
be written so that another interrupt can occur while the current one is
being processed, which causes overhead.  It also means that all interrupts
will interfere with the time slice spent running userspace program code.

>From a high level it does seem a bit counter-intuitive that an RT system
will run more sluggish than a non-RT, but the RT system is more about
consistency than performance bursts.

--Brett




On Thu, Aug 12, 2010 at 9:11 AM, david headon <davidheadonuk at yahoo.co.uk>wrote:

>   I'm not sure why. I find similar non-functionality with the Ubuntu repos
> and their 'realtime' kernel..
> Longer boot times, and no discernible performance improvements...
> Perhaps even worse!
>
>
> > Also, I've
> > always found the RT kernels provide by J.eng to be quite stable and have
> > decent performance - in my experience better than any Ubuntu kernels I've
> > tested, though not quite on par with Fernando's usually excellent CCRMA
> > kernels.
>
> Do you have a feeling for why this is? Is it due to the versions being
> packaged or are Ubuntu doing something wrong?
>
>
>
>
> _______________________________________________
> Ardour-Users mailing list
> ardour-users at lists.ardour.org
> http://lists.ardour.org/listinfo.cgi/ardour-users-ardour.org
>
>
> _______________________________________________
> Ardour-Users mailing list
> ardour-users at lists.ardour.org
> http://lists.ardour.org/listinfo.cgi/ardour-users-ardour.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ardour.org/pipermail/ardour-users-ardour.org/attachments/20100812/c307ce48/attachment-0002.htm>