silent way voice latency question

[james_holden]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

First up, really impressed with silent way, it works great with my MFB Kraftzwerg and a bunch of stuff in my modular. I'd love key and velocity to be able to modulate envelope parameters, but..

The problem I have is in Live 8.0.4 when I'm trying to use the voice module polyphonically - using Midi Yoke and a max patch I knocked up to share notes between n midi channels. I'm finding that with several silent way voice plugins my timing is off. I realise that the extra variables place this outside straight 'support' but I'm 100% sure the midi my max patch is sending is in time and correct.

My question is basically: can you suggest a way/workaround so that ableton's automatic external instrument delay compensation can be used with silent way?

Here're the details of my setup..

The patch is here: http://www.jamesholden.org/downloads/po ... r%20v2.zip (http://www.jamesholden.org/downloads/polyphoneriser%20v2.zip)
Midi Yoke is free.

In my Live set I have several channels of silent way, receiving midi that the max patch spits out via Midi Yoke's virtual loopback inputs. There is an 'external instrument' track in Live that is sending midi to the max patch, and receiving the mixed audio from the modular.

The problem: is that live seems to compensate for input/output latency on the first silent way plugin, but not the others. Dragging the first one to the right causes it to play all the voices in time again. Live's 'track delay' for the trigger midi channel doesn't work for positive values, so I've resorted to delaying all the other parts of the track, but I have to adjust this delay every time I add a plugin...

And if you were to release a polyphonic version of the voice plugin I'd pay extra for it BTW...

Thank you for your time and consideration in this matter.

[james_holden]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

i've been investigating this further..

it seems i can make it work in Live 7 - see this set (which uses the polyphoneriser max patch linked above receiving on midi yoke 1 and sending on midi yoke 2.) https://rcpt.yousendit.com/711448846/22 ... fdbd2a5361 (https://rcpt.yousendit.com/711448846/22264da61182f7f5324a4bfdbd2a5361)

i've had to predelay the midi tracks feeding the 2 silent way voices by a few ms to get it in time, but this delay stays constant no matter how many plugins are increasing the overall latency in ableton.

however in live 8 this set does not work at all - the timing of one of the sw voices is out by a different amount to the other, and neither is in time.

so it turns out that a large slice of the problem is ableton - i've mailed their support.


suggestions:
-could you add a 'latency' control to the sw voice vst? - similar to your latency fixer plugin (which is mac only)
-could you make a polyphonic vst? then at least the voices of an instrument would be in time with each other..

[os]

Hi James,

I think this is doomed to failure, or at least, to be a complete nightmare. I can only guess at how timing data is going to flow from Live, via MidiYoke, through Max, back through MidiYoke, and back to Live. <shudder>

What you want is a VST that processes MIDI to achieve the same effect, so you can ditch the whole MidiYoke thing. Maybe one of these (the first particularly):
http://www.kvraudio.com/get/3735.html
http://www.kvraudio.com/get/2969.html
http://www.kvraudio.com/get/3263.html

In fact, what you really want is Max for Live. When that's out? Soon?

It is of course also possible for me to make a polyphonic version of Silent Way. It's a reasonable amount of work though and I don't imagine that demand for it would be terribly high.

How many voices of polyphony are you using?

cheers,
os.

[james_holden]

"I think this is doomed to failure,"

you're probably right, it does sound like a nightmare. but it did all work 100% perfectly with my 2 outboard midi-cv modules instead of silent way. and now i've clearly established it's live 8's fault and have my song working great in live 7.

so - very sorry for saying it was your bug when it wasn't..

i'm still having to hand adjust the latency correction on the midi tracks that feed the silent way vsts, but it's consistent and doesn't change when live's latency increases due to more plugins being loaded. if you could find a way to preset the latency adjustment into silent way that would solve this problem.

i'm using 3 voice polyphony at the moment, but have had 5 going (using 2 self-oscillating filters as well as my oscs) and just ordered another oscillator.. it's really fun to have polyphony from a modular, treating each voice differently etc.. and polyphony does seem to be one of the popular features of volta.. i can empathise if you think it's a can of worms though.

one bug i spotted - if you duplicate a silent way voice controller track in live the vst may have issues showing the interface - you get a blank gui window and a slight ui freeze. saving/exit/reload solves this.

and: have you considered letting velocity/key modulate env parameters? or is this doable with scripting?

[os]

Velocity handling is in the frame for an update sometime soon. Surprised in retrospect that I overlooked it!

[WMP]

A little vertical velocity slider next to the envelopes would be great. So you can assign how much the velocity affects the envelope.