[PlanetCCRMA] Ardour changing pitch

[Stephen Crane]

On Wednesday 30 November 2005 09:49, Jan Depner wrote:
> On Wed, 2005-11-30 at 13:36, Fernando Lopez-Lezcano wrote:
> > On Wed, 2005-11-30 at 11:14 -0800, Stephen Crane wrote:
> > > I've been having trouble with Ardour changing the pitch of my tracks
> > > after I save, shut down the computer and come back to it a week or so
> > > later.
> > >
> > > First I recorded about a dozen tracks. When I came back to this project
> > > a week or so later, all the tracks had gone a little over a half step
> > > flat, but they were all in tune with each other. So I re-recorded my
> > > tracks, leaving the old, out-of-tune ones alone. When I came back a
> > > couple of weeks later, the newer tracks were out of tune (flat), again
> > > all by the same amount, but the old tracks that were out of tune
> > > before, were now perfectly in tune.
> > >
> > > Here's my specs:
> > > Fedora Core 2 - CCRMA Edge kernel
> > > Latest updates from CCRMA (a couple of days ago)
> > > M-Audio Delta 410 (PCI card)
> > > Sample Rate - 44100
> > >
> > > Any ideas on what could be causing this? I've searched all over, but
> > > the only thing I found even close to this problem was this post:
> > > http://ccrma-mail.stanford.edu/pipermail/planetccrma/2005-July/009835.h
> > >tml but it doesn't appear to be the same problem to me.
> >
> > This sounds to me like a mismatch at some point in sampling rates (44100
> > vs. 48000). Hard to say _how_ exactly it happens - but at the core of
> > the problem you will probably find something like that happening.
> >
> > 48000/44100=1.0884353...
> > 1/2 step in a 12 tone equal tempered scale = 1.059463...
> > -- Fernando
>
>     This is what I expect is happening as well.  Run envy24control and
> make sure you are set to 44100 when recording.  It's the Master Clock
> setting in the Hardware Settings tab.

Ok, I'll check this when I can get to that machine, but IIRC it was 44100. 
Even if it wasn't, why would some of the tracks be on pitch now, when they 
weren't before?

Thanks for the pointers, I'll look further into the resampling.

Thanks,
Stephen