I have to check ,I'm also looking at other possible solutionsurvieh wrote:ok. another idea is to reduce the number of oscillators.is there a way to detune the incoming event before it passes the oscillator?
Let's build a mega-Synth
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
Ok,I tried to find all the solutions, to have that kind of control the patch need 32 oscillators,so the setup you used for your patch is the only solution urvieh,we can still try to optimize the patch and see the result.
question, why you use those mux outside of polysynth?
Edit:I forgot to specify, only 32 oscillator for both layer,the event delay solution unfortunately not work in this case.
question, why you use those mux outside of polysynth?
Edit:I forgot to specify, only 32 oscillator for both layer,the event delay solution unfortunately not work in this case.
-
- KVRist
- Topic Starter
- 184 posts since 13 Apr, 2004 from germany
sadly i came to the same conclusion 
no other possibilities there.
the mux patch uses audio-event generators. they are not present at the synthpatch.
i now start again from the simplified patch. i have dropped a lot of amps behind the oscillators and could free up some cpu-power.
all layers wont work in one synth-patch, i think so. because of the given polyphonic limit (48). i will have another try and then i will upload again
no other possibilities there.the mux patch uses audio-event generators. they are not present at the synthpatch.
i now start again from the simplified patch. i have dropped a lot of amps behind the oscillators and could free up some cpu-power.
all layers wont work in one synth-patch, i think so. because of the given polyphonic limit (48). i will have another try and then i will upload again
-
- KVRist
- Topic Starter
- 184 posts since 13 Apr, 2004 from germany
hey taifunk,
i came from work, my wife and my little daughter are on the road to the grandparents...now it is time for mulab
i quickly added another oscillator and did some little improvements. You can also turn on/off the Oscillators from the front-panel.
now the result is as follow:
with the same settings on the silenth the cpu is nearly identical. For this test i used your step-sequencer from the modified first version.
Now by playing 3 keys (without stepsequencer) at the same time (8*2*3=48 Voices!), the mux eats a little bit more cpu ~1-2% than silenth. (silenth was 9-10%, mux has used 11-12%)
So i think we are on the way
the modulation hopefully doesn't eats the rest of the CPU 
i came from work, my wife and my little daughter are on the road to the grandparents...now it is time for mulab
i quickly added another oscillator and did some little improvements. You can also turn on/off the Oscillators from the front-panel.
now the result is as follow:
with the same settings on the silenth the cpu is nearly identical. For this test i used your step-sequencer from the modified first version.
Now by playing 3 keys (without stepsequencer) at the same time (8*2*3=48 Voices!), the mux eats a little bit more cpu ~1-2% than silenth. (silenth was 9-10%, mux has used 11-12%)
So i think we are on the way
the modulation hopefully doesn't eats the rest of the CPU
You do not have the required permissions to view the files attached to this post.
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
Simply perfect, well done urvieh For what I see we had the same idea 
Sorry for the delay ,at the time I do not have my pc,in a few hours I'm at home and begin to see what I can add
For what I see we had the same idea Sorry for the delay ,at the time I do not have my pc,in a few hours I'm at home and begin to see what I can add
-
- KVRist
- Topic Starter
- 184 posts since 13 Apr, 2004 from germany
thats great! nice that you think the same way 
i also think, we have to stop with osc1 and osc2 (+ the additional trigger model). the patch then can simply multiplied by everyone. what do you think?
i also think, we have to stop with osc1 and osc2 (+ the additional trigger model). the patch then can simply multiplied by everyone. what do you think?
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
Hi urvieh,
I'm trying different solutions for the various features,
ATM I added the filters,I used the switch technique suggested by bibz1st,works great.
I changed the color of the panel to work better, I'm a bit color blind and I have some problems with the colors,do not pay attention to the front panel is only a temporary layout.
I'm sorry if I answer late, but unfortunately my days are a mess.
I'm trying different solutions for the various features,
ATM I added the filters,I used the switch technique suggested by bibz1st,works great.
I changed the color of the panel to work better, I'm a bit color blind and I have some problems with the colors,do not pay attention to the front panel is only a temporary layout.
I'm sorry if I answer late, but unfortunately my days are a mess.
You do not have the required permissions to view the files attached to this post.
-
- KVRist
- Topic Starter
- 184 posts since 13 Apr, 2004 from germany
Nice Filters. Thanks...but not playable live it uses too much CPU while playing.
please do not stress yourself. Its all right. This project should be fun
i have just finished the oscillator-section. so we can simply do the rest.
i would NOT integrate an amp-distortion into this patch, because of its massive cpu-usage.
let us first keep it simple, please
here is a picture of the final oscillator-section. it works great an it is much more flexible than silenth:
you can also combine retriggered and free oscillators. The result is a sound with much more attack!
i also found out that the cpu-usage is now equal with the silenth one:
i alway forgot to boost the polyphony into silenth. now silenth uses more cpu. it has ever stolen my voices
BTW: sylenth was originally compiled with synthedit.
i also started with a background-skin. - sorry for the colors - this thing will be a real beast, when we are finished
So what to do?:
1. Modulation-matrix? yes/no (which destinations)
2. Fixed Filter-Envelope (yes/no)?
3. Portamento?
4. Noise-gen? (yes/no)
-> some days ago i have also replicate the lfo. how much lfo's should this one-layer-patch have?
it uses too much CPU while playing.please do not stress yourself. Its all right. This project should be fun
i have just finished the oscillator-section. so we can simply do the rest.
i would NOT integrate an amp-distortion into this patch, because of its massive cpu-usage.
let us first keep it simple, please
here is a picture of the final oscillator-section. it works great an it is much more flexible than silenth:
you can also combine retriggered and free oscillators. The result is a sound with much more attack!
i also found out that the cpu-usage is now equal with the silenth one:
i alway forgot to boost the polyphony into silenth. now silenth uses more cpu. it has ever stolen my voices
BTW: sylenth was originally compiled with synthedit.
i also started with a background-skin. - sorry for the colors - this thing will be a real beast, when we are finished
So what to do?:
1. Modulation-matrix? yes/no (which destinations)
2. Fixed Filter-Envelope (yes/no)?
3. Portamento?
4. Noise-gen? (yes/no)
-> some days ago i have also replicate the lfo. how much lfo's should this one-layer-patch have?
You do not have the required permissions to view the files attached to this post.
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
exactly, this is the reason why some of the things can not be the same,in fact I prefer to use only two filters in series.urvieh wrote:Nice Filters. Thanks...but not playable live
Thanks, totally agree!please do not stress yourself. Its all right. This project should be fun
No problem, I think it is unnecessary to add all the same functions,better something like "inspired by" not a real clone.i have just finished the oscillator-section. so we can simply do the rest.
i would NOT integrate an amp-distortion into this patch, because of its massive cpu-usage.
let us first keep it simple, please
Perfecthere is a picture of the final oscillator-section. it works great an it is much more flexible than silenth:
you can also combine retriggered and free oscillators. The result is a sound with much more attack!
i also found out that the cpu-usage is now equal with the silenth one:
i alway forgot to boost the polyphony into silenth. now silenth uses more cpu. it has ever stolen my voices
BTW: sylenth was originally compiled with synthedit.
surely, more voices = more cpu
No problem, that's greati also started with a background-skin. - sorry for the colors - this thing will be a real beast, when we are finished
-
- KVRian
- 877 posts since 28 Feb, 2015 from Interstella 5555
The modulation matrix is very difficult if not impossible without eat all the CPU,urvieh wrote: So what to do?:
1. Modulation-matrix? yes/no (which destinations)
2. Fixed Filter-Envelope (yes/no)?
3. Portamento?
4. Noise-gen? (yes/no)
-> some days ago i have also replicate the lfo. how much lfo's should this one-layer-patch have?
Maybe for modulations is better 1lfo+1envelope with different destinations(filter/pitch/volume)
you can always add more later.
A global portamento control without the other control.
A noise generator with an on / off switch connected to the same oscillator amplifiers should do the job:)
-
- Banned
- 1256 posts since 22 Feb, 2014
any chance to download patch showing in picture ?urvieh wrote:Nice Filters. Thanks...but not playable live
please do not stress yourself. Its all right. This project should be fun
i have just finished the oscillator-section. so we can simply do the rest.
i would NOT integrate an amp-distortion into this patch, because of its massive cpu-usage.
let us first keep it simple, please
here is a picture of the final oscillator-section. it works great an it is much more flexible than silenth:
you can also combine retriggered and free oscillators. The result is a sound with much more attack!
i also found out that the cpu-usage is now equal with the silenth one:
i alway forgot to boost the polyphony into silenth. now silenth uses more cpu. it has ever stolen my voices
BTW: sylenth was originally compiled with synthedit.
i also started with a background-skin. - sorry for the colors - this thing will be a real beast, when we are finished
So what to do?:
1. Modulation-matrix? yes/no (which destinations)
2. Fixed Filter-Envelope (yes/no)?
3. Portamento?
4. Noise-gen? (yes/no)
-> some days ago i have also replicate the lfo. how much lfo's should this one-layer-patch have?
-
- KVRAF
- 3139 posts since 28 Mar, 2008 from a Galaxy S7 far far away
I was thinking the same!
Would you mind sharing at this stage?
Others may be able to contribute to the design?
Would you mind sharing at this stage?
Others may be able to contribute to the design?
-
- KVRist
- Topic Starter
- 184 posts since 13 Apr, 2004 from germany
noisegenerator is ready.thank you so much taifunkTaifunk wrote: The modulation matrix is very difficult if not impossible without eat all the CPU,
Maybe for modulations is better 1lfo+1envelope with different destinations(filter/pitch/volume)
you can always add more later.
A global portamento control without the other control.
A noise generator with an on / off switch connected to the same oscillator amplifiers should do the job:)
my customized lfo is a desaster. it makes a heater out of my cpu
so i have to use the mux-lfo module but its toooo sloooooooow...
no matter
Last edited by urvieh on Wed Mar 22, 2017 8:38 pm, edited 1 time in total.