[PD] question about clip and reverb

Andy Farnell padawan12 at obiwannabe.co.uk
Mon Jan 3 16:11:03 CET 2011


On Mon, 03 Jan 2011 14:34:08 +0100
IOhannes m zmölnig <zmoelnig at iem.at> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 01/03/2011 01:39 PM, ronni montoya wrote:
> > Hello,  i like the distorsion effect that i get when i  turn up the
> > amplitude of a freeverb~  until it clips the signal.
> > The sound is interesting but the problem is that since its  clipping
> > it affects all the other sound pd is generating.
> > 
> > My question is:
> > 
> > How can i replicate that distorsion effect in a way that it doesnt
> > affect the others sounds?
> 
> well, make sure that the effect appears with no other sound involved.
> e.g. what you are describing seems like you are saturating the [dac~].
> now what you are sending to the dac is a sum of all signals, and this
> sum is then saturated (if the absolute sum of your signals is >1 and the
> sound-infrastructure you are using cannot handle floating point signals)



The gotchya is that different sound systems handle >1(FSD) in differenmt
ways. So using [clip~], although offering portable results, is sometimes
not the same as overdriving the DAC. Symptom - you get a great sound
that using [clip~] cannot substitute.

But I learned that you must never rely on the results of overdriving
an unknown system back when moving a piece written on a linux-ALSA
386 PC with a soundblaster card to a Mac with an external firewire 
sound interface - totally different sound. I never did discover exactly
why, though I guess the signal, was finding its way into either signed 
or unsigned ints at different stages and getting wrapped or truncated
in either case.

The lesson is, make sure you get the sound exactly how you like with
plenty of headroom to spare.

> 
> therefore, the perceived effect depends on all signals sent to the dac.
> 
> the solution is of course to make the effect happen without involving
> all signals and then scale it back so that the sum of all signals does
> not trigger another distortion.
> 
> > Is there a external that recreates that type of distorsion?
> 
> no. you can do that with vanilla Pd objects, e.g. [clip~]
> Pd itself has plenty of objects to do most complicated stuff, often
> there is no need to resort to an external
> (otoh, even though you can do the requested thing with Pd-vanilla just
> fine, i'm sure that there is an external that does what you need)

Maybe looking at some distortion techniques using table lookups
might get you a similar sound.

 
> fgmsdrt
> IOhannes
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
> 
> iEYEARECAAYFAk0h0FAACgkQkX2Xpv6ydvQjdgCdH1bjG5MGZq/8vpmNpWHeHpDF
> 56wAnjef7EkkTeXLPmW2DY5fBvLQnPp6
> =pebv
> -----END PGP SIGNATURE-----
> 
> _______________________________________________
> Pd-list at iem.at mailing list
> UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list


-- 
Andy Farnell <padawan12 at obiwannabe.co.uk>



More information about the Pd-list mailing list