Page 6 of 7

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 10:17 am
by khz
Attack and Release currently react to changes in the MIDI notes (or velocity?) of the keyboard, not to MIDI CC?
Wouldn't it make more sense if <EDIT>Sustain and Release</EDIT> would only start to self-oscillate at 1.0 (maximum)?

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 10:26 am
by tramp
khz wrote:Attack and Release currently react to changes in the MIDI notes of the keyboard, not to MIDI CC?
Wouldn't it make more sense if attack and sustain would only start to self-oscillate at 1.0 (maximum)?
Please check out the latest commit, I've just pushed a fix for that 2 hours ago.
Also now detune is midi controllable by CC 0x5E / 94 Celeste Level or Detune.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 11:30 am
by khz
Now there is no sound with every setting. ;-)
@Carla When I close the instrument window and open it again, the settings are always set to default values, the instruments do not keep the changed settings.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 11:45 am
by tramp
Ups, a error in the ttl file slips in. :oops:
Carla is a bit picky about such.
Fixed now. Thanks for reporting.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 11:59 am
by khz
:D A wonderful instrument! :dance:
tramp wrote:Also now detune is midi controllable by CC 0x5E / 94 Celeste Level or Detune.
README.md

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Thu Dec 26, 2019 1:34 pm
by khz
Would an optional switchable FM synthesis - 12 voice operator - be possible, down right next to the effects?
Image

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Fri Dec 27, 2019 3:39 pm
by khz
What does the A(D)SR control? When the A(D)SR is completely closed you still hear the complete formant/vocal synthesis, no "zap" formant/vocal synthesis. The envelope seems to control something else.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Fri Dec 27, 2019 4:04 pm
by tramp
First is a NoteOn event open the gate, then attack steps in.
attack is attack time, when you set it to "zero" it attack immediately. When you set it to 1 (127) the attack time is the highest, means, the volume turns on sloooolllyyy, when you keep the note pressed. When attack have reach the full level, the sound play's as long you press the note. After a NoteOff event , the sustain steps in, and keep the level for the given time, here as well, as higher the sustain is set, as longer the tone will sustain. When sustain time is over, the release steps in. It's again a timer, which, in opposite to the attack will smooth down the volume since a 'zero' level is reached.
At any time you could modulate the sound with the Vowel or the detune.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 9:26 am
by khz
The envelope does not control the amplitude (volume) (ENV AMT)?
It normally relates to the amplitude (volume), but it may also involve elements such as filters (frequencies)
khz wrote:What does the A(D)SR control?
Maybe I'm thinking wrong, but when the envelope is completely closed there should be nothing to hear.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 9:39 am
by bluebell
tramp wrote:First is a NoteOn event open the gate, then attack steps in.
attack is attack time, when you set it to "zero" it attack immediately. When you set it to 1 (127) the attack time is the highest, means, the volume turns on sloooolllyyy, when you keep the note pressed. When attack have reach the full level, the sound play's as long you press the note. After a NoteOff event , the sustain steps in, and keep the level for the given time, here as well, as higher the sustain is set, as longer the tone will sustain. When sustain time is over, the release steps in. It's again a timer, which, in opposite to the attack will smooth down the volume since a 'zero' level is reached.
At any time you could modulate the sound with the Vowel or the detune.
This is different to most other implementations:

Attack: Time to reach a fixed maximum Level after note on
Decay: Time to reach the level that's defined by Sustain (key still pressed)
Sustain: Level that is reached when Decay time ends, range mostly between 0 and max (key still pressed)
Release: Time to reach zero level after note off
ADSR_parameter.svg.png
ADSR_parameter.svg.png (52.04 KiB) Viewed 1124 times
If Attack is 0 then the sound starts immediately with full volume.
If Release is 0 then the sound ends immediately after key note off.
If the key is released in Attack phase then Decay and Sustain have no effect, only Release.
If the key is released in Decay phase then Sustain has no effect anymore, only Release.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 9:56 am
by tramp
bluebell wrote:This is different to most other implementations:
Attack and Release been implemented exactly this way.
Difference is, here we don't have a Decay.
Difference is, Sustain is implemented as Hold, means it's a time duration were the sound play after the key has released. The release then steps in after the sustain time is over.
When Sustain (hold) is set to 1 (127) the sound will play forever. This allow to create a sound Floor, were the oldest Note get replaced by the newest.
When sustain is set to zero, Attack and Release behaviour is like in any other synth.
khz wrote:Maybe I'm thinking wrong, but when the envelope is completely closed there should be nothing to hear.
I don't understand what you mean by this.
When the envelope is closed, you hear nothing. I may rename the Sustain to Hold and implement a real Decay and sustain?

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 10:32 am
by khz
When an ENV AMT is completely closed, no sound can be heard. IMHO
tramp wrote:I may rename the Sustain to Hold and implement a real Decay and sustain?
For testing it would be interesting to see how the instrument behaves and how it sounds.
:-)

&&
khz wrote:Would an optional switchable FM synthesis - 12 voice operator - be possible, down right next to the effects?
Image
formant/vocal/FM synthesis ;-)

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 11:50 am
by bluebell
tramp wrote:I may rename the Sustain to Hold and implement a real Decay and sustain?
That would be fine. You can omit Hold (time). The Korg MS10 had it but I never used it. It doesn't make much sense

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sat Dec 28, 2019 11:10 pm
by Baggypants
ADSR is only one way to do an envelope though, and was mostly decided on because of price. The Korg EX800 had a 5 stage envelope which allowed for a second "attack" phase after the decay. For a vocal synth this would be more interesting as voices can vary their dynamic on a single note.

Image

Image from https://www.soundonsound.com/techniques ... -envelopes

The Casio CS range had an 8 stage envelope of attacks and decays.

Re: [LAA] Xmonk.lv2 a little sound generator

Posted: Sun Dec 29, 2019 10:18 am
by khz
@ADBSSR :idea: :-D
@ viewtopic.php?p=113188#p113188 With the 12 voices/operators of XPolymonk `SFFormantModelBP` from physmodels.lib.
Sixfold polyphony should be guaranteed at a minimum. You can optionally select 5 - x circuit models, similar to the effects. ((Yamaha) ~FS1R https://www.amazona.de/wp-content/uploa ... 30x207.jpg - Image )