Hi all,

I'm in the process of finishing up a sound toy for Android. I use a pd patch with libpd, which only supports vanilla objects. I like the way freeverb~ sounds, so I would like to replicate it sonically.

I'm looking to patch a good reverb, since the app is ambient by nature. Can anyone give me a general process to follow? I would use someone's abstraction, but I am selling the app, and it wouldn't feel right.

Thanks

Schroeder type topologies generally aren't very good. Freeverb is well tuned, and probably about the best you can expect to get from that kind of algorithm, but there's better stuff out there. Pd has a few decent example FDN type reverberators (Miller Puckette happens to have published the first paper that explicitly describes them), you can use them as is or add your own enhancements. I have some pretty good algorithms that I'll try to post soon.

Here's a pretty good reverb, it should be a "real" Lexicon algorithm (Concert Hall from the 224X, allegedly), but I can't confirm that. I've added a few enhancements. All necessary abstractions are included (and they may be useful on their own). It's definitely a step up from Schroeder-type algorithms. It's very synthetic sounding, but in a flattering way.

Here's the original: http://www.convexoptimization.com/wikimization/index.php/Dattorro_Convex_Optimization_of_a_Reverberator

It's not packaged as well as it could be, mainly because I'm trying to keep it in line with the original diagram. Of course it's possible to extend it further, but I'd rather start over with a new algorithm. This one has been taken about as far as is practical without sacrificing basic authenticity.

There's a lot of expr stuff, a couple non-Vanilla objects (lp8_cheb~ from iemlib), and possibly an implementation of a still valid patent (5553150, though I think there's some significant prior art that may render it unenforceable). And I don't guarantee that it won't kill your family. Use it at your own risk.

I have some fancier algorithms, but this one is nice sounding and reasonably efficient, so it gets a fair amount of use.

http://www.pdpatchrepo.info/hurleur/reverbCH.zip

that should help ypu too, considering the title :

http://www.uni-weimar.de/medien/wiki/PDCON:Conference/Reverb_Design

hi,

fdn~ is cool.....

allthebest

flo.

@BerengerRecoules said:

that should help ypu too, considering the title :

http://www.uni-weimar.de/medien/wiki/PDCON:Conference/Reverb_Design

Is he making the PD stuff available anywhere, or just talking about it?

Here's another one, it's a clone of the Ursa Major SST-282. I've based the design on the patents, user manual and service manual, but tuned it myself, and again added some enhancements. I'm not emulating the glitchy, non-interpolated delay modulation and I'm not including the selection of "rooms" (I don't have any information on the delay times here, and they don't seem extremely useful anyway). It's a very idiosyncratic algorithm, and certainly not an all-purpose thing since it turns into weird slush for long decay times (it actually has to, to prevent unstable feedback).

With some modification, it could work well as an initial stage for a more advanced reverberator. You can get away with a large amount of "feed forward" modulation, and the sound can be made extremely rich if it's then passed through some APF diffusion stages, or FDN or whatever. If a signal recirculates through a modulated delay (as in the reverbCH algorithm I posted earlier), it becomes increasingly detuned over time, so this limits the maximum modulation amount versus decay time. So if the goal is to be "lush" sounding, it's possibly better to keep modulated delays out of the main recirculating part.

I've got a selection of original algorithms too, but they're continual works in progress, and much more computationally expensive...

http://www.pdpatchrepo.info/hurleur/reverb282.zip

I'd love to try the reverb282 patch but cant get anything out of it, can someone post a patch with it connected to audio input and output?

@soapdodger That took a few minutes to figure out.......!
The [reverb282~] object takes an argument that sets the length of the delay line that is being tapped.
Here I have set it to 5 seconds. Maybe 20 would be better for a seriously long RT...... so [reverb282~ 20000].
Without the delay set, it just outputs an insane level (chronic immediate feedback) that overloads the soundcard and sounds like........ silence.....
working.zip
David.

@whale-av

thanks!

@acreil said:

creasingly detuned over time, so this limits the maximum modulation amount versus decay time. So if the goal is to be "lush" sounding, it's possibly better to keep modulated delays out of the main recirculating part.

great sound, thanks!

it might be worth consideration (www.reddit.com/r/reverbs)

@whale-av

In the reverb282~ object there is a filter~ object that does not seem to exist.

Is there an updated reverb unit that is preferred?

Katja Vetter created a vanilla abstraction that simulates [freeverb~]. I can only find a fork by Derek Kwan, but it's the same abstraction. It's here https://github.com/derekxkwan/pd-vfreeverb

@raynovich I don't have that error even in pure vanilla, and a search doesn't find the object Where do you see the object in the patch?
David.
P.S...... did you unzip the folder?

Hey,
So, I am not using vanilla pd. That must be it.