Page 1 of 2

new synth: vModSynth

Posted: Mon Feb 11, 2013 5:42 pm
by wolftune

Re: new synth: vModSynth

Posted: Mon Feb 11, 2013 6:00 pm
by wolftune
Well, the answer is clearly to not use it yet. Bug the dev to support JACK. It's a new project, he'll probably be happy to improve it.
EDIT: I already replied and mentioned KXStudio on that blog in response to someone talking about making it work with Pulse (ugh). But you probably have more clout than I do to tell the dev to support JACK…

Re: new synth: vModSynth

Posted: Mon Feb 11, 2013 6:24 pm
by wolftune
Yeah, I'd be happy with trying it just as an LV2 plugin potentially without a standalone option.

Re: new synth: vModSynth

Posted: Tue Feb 12, 2013 12:54 am
by GraysonPeddie
Audio latency of 3 to 5+ seconds (margin for error +/- 75%)... Yeah, RIGHT! :(

Re: new synth: vModSynth

Posted: Tue Feb 12, 2013 3:42 am
by tux99
GraysonPeddie wrote:Audio latency of 3 to 5+ seconds (margin for error +/- 75%)... Yeah, RIGHT! :(


Looks like you need a more powerful PC... what cpu are you using?

Re: new synth: vModSynth

Posted: Tue Feb 12, 2013 9:13 am
by werock
The documentation (which is very good BTW) quotes an expected latency of 6ms, and probably less if you use a low latency/real time kernel.

It looks like a very well thought out design, hopefully it proves more reliable than AMS.

I can't wait to try it out, I love modular synths :)

Re: new synth: vModSynth

Posted: Tue Feb 12, 2013 7:17 pm
by ssj71
hopefully it proves more reliable than AMS


Just curious, is AMS notoriously unstable? I admit I've only done relatively simple patches, but its been really good for me.
This project does look cool though. His patch cables defy gravity!

Re: new synth: vModSynth

Posted: Tue Feb 12, 2013 7:21 pm
by GraysonPeddie
tux99 wrote:
GraysonPeddie wrote:Audio latency of 3 to 5+ seconds (margin for error +/- 75%)... Yeah, RIGHT! :(


Looks like you need a more powerful PC... what cpu are you using?

Wrong assumption, cause when I play with ZynAddSubFX, it worked fine. Very little latency, if at all.

But if you want specs, how about just an AMD A10-5700 Accelerated Processor Unit (APU) and 8GB of RAM? The latency is about 2 seconds since when I've counted.

MIDI Input's first pitch jack to oscillator's 1st pitch jack
MIDI Input's gate to ADSR's gate
Oscillator's sine jack to amplifier's signal input #1
ADSR's envelope to amplifier's control input #1
Amplifier's "Output" (not "-Output") jack to Audio Out's Mono Out jack.

Re: new synth: vModSynth

Posted: Sat Feb 16, 2013 4:16 pm
by werock
ssj71 wrote:
hopefully it proves more reliable than AMS


Just curious, is AMS notoriously unstable? I admit I've only done relatively simple patches, but its been really good for me.
This project does look cool though. His patch cables defy gravity!


It was never reliable for me - it would usually crash before I could get any tunes going with it.

Re: new synth: vModSynth

Posted: Sat Feb 16, 2013 11:49 pm
by aMused
From the comments:
Rafał Cieślak says:
"Thank you for your tip. I am aware of JACK’s advantage over ALSA, and if only I will continue to develop this project, getting it use JACK for audio output will be one of my priorities."


...... :)

Re: new synth: vModSynth

Posted: Sun Feb 17, 2013 4:59 pm
by linuxdsp
From the comments:
Rafał Cieślak says:
"Thank you for your tip. I am aware of JACK’s advantage over ALSA, and if only I will continue to develop this project, getting it use JACK for audio output will be one of my priorities."


In response to this I believe... :)

linuxDSP Says:
February 11, 2013 at 10:36 pm

Can I recommend that you consider using JACK instead of connecting to the soundcard directly via ALSA. JACK is the accepted standard for *all* pro-audio applications on linux, and allows much easier integration with existing e.g. DAW software, allowing audio / MIDI to be routed to and from compatible applications, in addition to the physical I/O interface


Although I see later in the thread..

Sadly some of the dsp code is incompatible with realtime constraints (use of std::queue and locks.) An LV2 plugin would be nice, but that would require complete separation of UI and dsp state which would be a major refactor for this.


Which, unfortunately, if true, indicates the design is fundamentally wrong - irrespective of whether it is LV2, standalone JACK, VST, ALSA or whatever else.

Re: new synth: vModSynth

Posted: Tue Feb 19, 2013 12:42 am
by aMused
Think I will make sure that this ".... :) " is clear.
I smiled because there is the potential of new synth. But more importantly a receptive developer who might take to heart the involvement of experienced eyes.