iiwusynth-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[iiwusynth-devel] Re: CVS update


From: Josh Green
Subject: [iiwusynth-devel] Re: CVS update
Date: 19 Dec 2002 02:32:09 -0800

On Thu, 2002-12-19 at 01:25, M. Nentwig wrote:
> > My understanding of tuning in SoundFont files is that the
> > root note (set
> > from the sample layer or from a Root Key override generator at the
> > instrument layer) should always play the sound at its
> > original sampling
> > rate (taking into account other tuning adjustments, fine tune
> > and coarse
> > tuning at the inst layer and fine tune at sample layer). At
> > any rate, it
> > should not be the fixed value C3 but whatever the root note happens to
> 
> > be.
> 
> I think this won't work, at least not with the sample layer.

Huh?? There is no pitch scaling at the sample layer, its a generator.

> The root note is set at the sample layer, OK. Now there are several
> samples for different ranges of the keyboard.
> Each of them keeps its pitch at its root key. Then the pitch jumps at
> transitions between samples.
> Maybe this would work with the root key on instrument / preset level
> only. But I think then scale tuning would have to be handled by the
> SFloader (because the information is not available anymore to the
> synth).
> 

Am I missing something? Are you refering to what happens if one adjusts
the scale tuning on an existing split of samples?

It would seem that setting a scale tuning parameter for a multiple
sample split would cause them to no longer overlap correctly. I
confirmed this behavior with the OSS SB Live driver. I used an
instrument with different samples split across the keyboard (SAWTOOTH in
Vintage Dreams). Playing the root note of the lowest sample (which is
not set at C60) and changing the Scale Tuning had no effect on the
pitch. Playing any non-root note and changing Scale Tuning caused pitch
to change. Changing Scale Tuning on the global zone, DID cause the
splits to no longer overlap correctly (so one would then need to adjust
the root note and fine tuning).

> Cheers
> 
> Markus

Lates.
        Josh




reply via email to

[Prev in Thread] Current Thread [Next in Thread]