Phasex

[autostatic]

That would be great! Maybe you could take a look at Yoshimi which has a CLI -N option (--name-tag=) which enables you to set a different name for each instance. So phasex -N arpfiller would yield phasex-arpfiller as JACK inputs and outputs.

[disabled]

Working of multiple instances seems to be fixed.
Also changed scheme of instance naming - first instance will not get numeric suffix, any other instances with same name will get it starting from "01".
And, finally, added '-n' or '--name' option which is setting name of instance. (I think that changing full name instead of just name tag like in Yoshimi makes more sense. Anyway, it can be changed at any time.)

[autostatic]

Working of multiple instances seems to be fixed.

Great!

Also changed scheme of instance naming - first instance will not get numeric suffix, any other instances with same name will get it starting from "01".

So the first instance will be called phasex and the second phasex-01? That's a bit inconsistent and confusing imho.

And, finally, added '-n' or '--name' option which is setting name of instance. (I think that changing full name instead of just name tag like in Yoshimi makes more sense. Anyway, it can be changed at any time.)

That's just awesome, thanks a lot disabled! Compiling as we speak

[autostatic]

Hello disabled, just switched back to the unmodified version. Some of my patches sounded unusable and different and after closing Qtractor with two instances of phasex-dev open both those instances crashed, taking JACK along the way.

[disabled]

Hello disabled, just switched back to the unmodified version. Some of my patches sounded unusable and different

Although sounding was slightly changed by fixes in ADSR & chorus code, I didn't thought that it would be changed so heavily. Can i have an example of such patch?

after closing Qtractor with two instances of phasex-dev open both those instances crashed, taking JACK along the way.

More looks like JACK issue. I can't reproduce it.

So the first instance will be called phasex and the second phasex-01? That's a bit inconsistent and confusing imho.

I was heading mainstream at this - almost all synths does it that way.

[autostatic]

Although sounding was slightly changed by fixes in ADSR & chorus code, I didn't thought that it would be changed so heavily. Can i have an example of such patch?

I have a modified harsh-saw patch that displays this behaviour. I'll attach it to a new post or put it up somewhere.

More looks like JACK issue. I can't reproduce it.

It could be related to another application, I should test it some more. But the set-up I tested it on is rock solid and the only thing that changed was the new PHASEX version and my whole session crashed badly when I shut down Qtractor.

I was heading mainstream at this - almost all synths does it that way.

Then all those synths are wrong No, you're right, but I did like it that PHASEX doesn't do it this way.

[disabled]

Sorry for that "week of silence" - i had some technical problems.

AutoStatic, i tested many combinations of params and all default presets, but didn't find that terribly wrong sounding combination, so without example (you forgot to attach it?) there is nothing i can do.
Maybe you relied on original ADSR behavior which wasn't correct, with note duration less than attack + decay period it can give pretty noticable changes.

[disabled]

Meanwhile, tarball of first release - 'milestone 1' is ready for download.
Also, from now project has it's simple website: http://disabled.github.com/phasex-dev/index.html

[alex stone]

Can i ask that you make a small addition to the app, and make it jack_sessionable? It will help a great deal in big multi app projects.

Here's the info.

http://trac.jackaudio.org/wiki/WalkThro ... ackSession


Alex.

[disabled]

Can i ask that you make a small addition to the app, and make it jack_sessionable? It will help a great deal in big multi app projects.

As i already said, there is no stable release of JACK with jacksession, so this is not one of the primary goals, although maybe it will be done for Milestone 3.

[autostatic]

http://linuxaudio.org/

I'm going to check the different sounding patches today.

[autostatic]

Here's my modified harsh-saw.phx patch that sounds differently with phasex-dev:

Code: Select all

phasex_patch {
	info {
		version			= 0.12.0-pre1;
	}
	general {
		bpm			= 1;
		master_tune		= 64;
		portamento		= 0;
		keymode			= poly;
		keyfollow_vol		= 64;
		volume			= 100;
		transpose		= 52;
		input_boost		= 0;
		input_follow		= off;
		pan			= 64;
		stereo_width		= 112;
		amp_velocity		= 0;
	}
	filter {
		cutoff			= 76;
		resonance		= 58;
		smoothing		= 32;
		keyfollow		= keytrig;
		mode			= lp+bp;
		type			= 0;
		gain			= 110;
		env_amount		= 12;
		env_sign		= negative;
		attack			= 104;
		decay			= 10;
		sustain			= 127;
		release			= 56;
		lfo			= 0;
		lfo_cutoff		= 64;
		lfo_resonance		= 64;
	}
	delay {
		mix			= 0;
		feed			= 32;
		crossover		= off;
		time			= 1/2;
		lfo			= 0;
	}
	chorus {
		mix			= 0;
		feed			= 5;
		crossover		= on;
		time			= 13;
		depth			= 64;
		phase_rate		= 4/1;
		phase_balance		= 64;
		lfo_wave		= sine;
		lfo_rate		= 1/1;
	}
	envelope {
		attack			= 15;
		decay			= 45;
		sustain			= 10;
		release			= 0;
	}
	oscillator {
		modulation		= mix;
		polarity		= bipolar;
		source			= midi_key;
		wave			= saw;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 76;
		fine_tune		= 64;
		pitchbend		= 64;
		am_mod			= off;
		am_mod_amount		= 64;
		fm_mod			= off;
		fm_mod_amount		= 64;
		fm_mod_fine		= 64;
		pm_mod			= osc-3;
		pm_mod_amount		= 54;
		wave_lfo		= 0;
		wave_lfo_amount		= 64;
	}
	oscillator {
		modulation		= mix;
		polarity		= bipolar;
		source			= midi_key;
		wave			= saw;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		fine_tune		= 64;
		pitchbend		= 64;
		am_mod			= off;
		am_mod_amount		= 64;
		fm_mod			= off;
		fm_mod_amount		= 64;
		fm_mod_fine		= 64;
		pm_mod			= osc-4;
		pm_mod_amount		= 74;
		wave_lfo		= 0;
		wave_lfo_amount		= 64;
	}
	oscillator {
		modulation		= mod;
		polarity		= bipolar;
		source			= midi_key;
		wave			= saw;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 76;
		fine_tune		= 64;
		pitchbend		= 64;
		am_mod			= off;
		am_mod_amount		= 64;
		fm_mod			= off;
		fm_mod_amount		= 64;
		fm_mod_fine		= 64;
		pm_mod			= osc-2;
		pm_mod_amount		= 36;
		wave_lfo		= 0;
		wave_lfo_amount		= 64;
	}
	oscillator {
		modulation		= mod;
		polarity		= bipolar;
		source			= midi_key;
		wave			= saw;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		fine_tune		= 64;
		pitchbend		= 64;
		am_mod			= off;
		am_mod_amount		= 64;
		fm_mod			= off;
		fm_mod_amount		= 64;
		fm_mod_fine		= 64;
		pm_mod			= osc-1;
		pm_mod_amount		= 92;
		wave_lfo		= 0;
		wave_lfo_amount		= 64;
	}
	lfo {
		source			= keytrig;
		polarity		= bipolar;
		wave			= sine;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		pitchbend		= 64;
	}
	lfo {
		source			= keytrig;
		polarity		= bipolar;
		wave			= sine;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		pitchbend		= 64;
	}
	lfo {
		source			= keytrig;
		polarity		= bipolar;
		wave			= sine;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		pitchbend		= 64;
	}
	lfo {
		source			= keytrig;
		polarity		= bipolar;
		wave			= sine;
		init_phase		= 0;
		rate			= 1/1;
		transpose		= 64;
		pitchbend		= 64;
	}
}

[disabled]

AutoStatic, this difference caused by fixes of envelopes calculation. To observe that bug in original phasex you can just change "Env Amt" param in 'Filter Envelope' section to 127, resulting behavior will be unstable and obviously wrong. So sorry, but this will be never the same again.

Anyway, thanks for reply. Also, while playing with your preset, I found another bug in polyphony behavior

[murks]

Hi. Seems like you're making quite some headway

Have you tried contacting William by other means than the sysex mail address? I think that was necessary before because the website was down for a longer time. The page seems up now, but who knows whether mail works.

I'm also in favor of jack-session support. I know that there's no jack release that includes it yet.
Here's a video of jack-session with pyjacksm and jack2: http://hochstrom.endofinternet.org/file ... cast01.ogv

The number of programs that support it has been increasing recently:
http://www.trac.jackaudio.org/wiki/Walk ... ck_session

[autostatic]

AutoStatic, this difference caused by fixes of envelopes calculation. To observe that bug in original phasex you can just change "Env Amt" param in 'Filter Envelope' section to 127, resulting behavior will be unstable and obviously wrong. So sorry, but this will be never the same again.

That's ok, phasex-dev has so much advantages, I'll just play around with that patch until it sounds usable again.

Anyway, thanks for reply. Also, while playing with your preset, I found another bug in polyphony behavior

Nice! I already ran into weird behaviour of PHASEX when using it with Hydrogen, I thought it was a Hydrogen issue but apparently PHASEX wasn't handling the MIDI events that well either.