ZynAddSubFx (PadSynth) - why does nothing come close? :(

[MaliceX]

The reason the padsynth algo sounds lush and natural is the constant cancellation and reinforcment of harmonics, it's done by basically "rendering" a big (100+) number of detuned oscillators into a sample. You can get similar results in any synth that supports big "unison"-counts. Try helix! where you can have 64 osc per voice.

Any of you heard what the man said? Try Helix. It can do some insane pads.

Cheers!
bManic

It can sound quite alike those N***s e********s MP3 demo sounds for synths if given the time. It's just that I feel there's still something missing from it sound-wise. (We need some good sound designers touching this!)

Other than that, yes, it can do very nice pads, it just takes quite a bit to program it real good. (And with tilter zipper issue, holds it back a bit.)

[grymmjack]

I wish I understood enough about DSP and Audio Engineering to know what is being explained by Paul Nasca here: http://zynaddsubfx.sourceforge.net/doc/ ... Dsynth.htm

All I do know right now is that ZynAddSubFx is capable of the most incredible lush pad sounds that I've heard out of anything in my arsenal (Sytrus, Toxic III, Morphine, Hydra, Wusikstation, etc.) and even a number of synths I've demo'd out of my pricerange.

wow, that's a damned bold statement fandang.

i know karmafx features the padsynth module. dunno if that helps or not.

[grymmjack]

Karma FX Synth Modular has a Padsynth module:

http://www.karmafx.net/

;( i fail.

[grymmjack]

As for KarmaFX's "PAD" module, while it sounds nice, it kills my CPU.

render the track to wav. no more cpu issue

[fandango]

I really mean it though! I love those synths for what they offer, in particular Toxic III which still amazes me with it's flexibility. However, there is a definate richness that suits ZynAddSubfx for Pads that give a really full and powerful sound without being dense or muddy which can be the case when using large amounts of chorus etc.

I can't put the sound down to just being a massive 'unison'. It seems the shaping of the unison the further it is away from the original sound is tapered to reign in the effect a little bit, making it sound more natural around the core note than just a massive equally spread unison.

It's time I stopped blaming my ignorance and start learning to program DSP and make a little program to create PadSynth waveforms that can be loaded into SFZ. I've been going through the C++ examples that Paul Nasca has kindly made available. I do programming for a day job (mostly Java and C), just nothing to do with sound programming or DSP so it's time to get down & dirty. :p

[speccyteccy]

I'm suprised we've not seen more synths using Pauls algorithm.

I'd guess it would be easy to implement as an SE or SM module.

Also agree that the zynaddsubfx vst is too much of pain to use regularly.

[Rock Hardbuns]

I can't put the sound down to just being a massive 'unison'. It seems the shaping of the unison the further it is away from the original sound is tapered to reign in the effect a little bit, making it sound more natural around the core note than just a massive equally spread unison.

It's actually closest to additative synthesis, except the partials cover a band rather than a single frequency. I guess you can think of it as additative+unison , but as you say, you can control the bandwidth/ frequency relationship which is kind of novel. I don't think I've seen that elsewhere in any form.

It's time I stopped blaming my ignorance and start learning to program DSP and make a little program to create PadSynth waveforms that can be loaded into SFZ. I've been going through the C++ examples that Paul Nasca has kindly made available. I do programming for a day job (mostly Java and C), just nothing to do with sound programming or DSP so it's time to get down & dirty. :p

Good luck! ( I assume, of course, you know you can get help in the DSP area here on KvR )

[tamahome]

I think it's basically 'infinite oscillators', where the loudest oscillator is usually right where the harmonics are. Paul added a patch to export the wavs, but I don't know if there's a compiled windows version with that patch, even standalone. But in Paul's synth, the wav has a random start location each time it plays. That mp3 at the very beginning of this topic is nice btw. Sounds like Clockwork Orange. I'd like to hear the rest of it!

[Numanoid]

I see Beat Mag showcase the synth on their Facebook page today:
https://www.facebook.com/beat.magazin

Edit: The synth is in fact updated as lately as 25 feb 2015, to 2.5.0

How did they get that black skin BTW ?

[discoDSP]

Discovery Pro can do PadSynth re-synthesis as well.

http://www.discodsp.com/discoverypro/

[Chris-S]

Edit: The synth is in fact updated as lately as 25 feb 2015, to 2.5.0

How did they get that black skin BTW ?

I think this skin is standard in the newest linux versions.

http://sourceforge.net/projects/zynadds ... naddsubfx/

[wagtunes]

Frustrating that this won't work on Windows 7 64 bit because it sounds excellent.

Oh well, such is life I guess.

[Numanoid]

Frustrating that this won't work on Windows 7 64 bit because it sounds excellent.

I got it working on Windows 7 64 bit, but I'm using it in FL Studio

[aMUSEd]

It's no good on Mac either except as a standalone using some ridiculous Jack server.

[wagtunes]

Frustrating that this won't work on Windows 7 64 bit because it sounds excellent.

I got it working on Windows 7 64 bit, but I'm using it in FL Studio

Well, I tried it a while ago in Cubase 7 and I couldn't even get the GUI to show up right. That was in 2013. I doubt much has changed since then.