Attack stage contour for the Amp Envelope

Attack stage contour for the Amp Envelope
« on: March 29, 2022, 11:49:09 AM »
I am guessing it is logarithmic because it seems to be super quick. Seeing as how this is a polysynth it seems a bit strange. I would really like to have more gradual attacks for my pads.

Anyone else feel the same way or have any thoughts on this?

kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #1 on: March 29, 2022, 06:57:58 PM »
I've noticed this too.  It's logarithmic, intending to mimic an analog envelope.  But it almost seems too logarithmic, as its slope seems to go up faster and then slow down more than the logarithmic attacks on my other synths.  With a slow attack and a slow decay, it seems to "go flat" at the top before it reaches the decay, or like an envelope that's clipping at the top of the attack.

I don't know what envelope they modeled the Take 5 envelopes after.  I also have an OB-6 but no Prophets.  Are the Prophet 5/10 envelopes like this too?

Maybe changeable slopes could be added in a future firmware update?  It would make a flexible sound design synth even better.

Elric

  • *
  • 67
  • Peak Oil
Re: Attack stage contour for the Amp Envelope
« Reply #2 on: March 29, 2022, 10:21:16 PM »
(Over from the Pro3 forum, but...  )

Can you use the Mod matrix to modify this?
One of my fav things to do on the Pro 3 is modify the (Filt?, any) Env Decay with the (same) Envelope Amount (Amt).
You can make Curved Decay. With plus (Convex?) or minus (Concave?).
It's not perfect, but I love having some form of contour control.
(I hope it works on the T5.)

On the Pro 3, for the Filter Envelope, it would be:

Env LPF   >>   +/-4   >>   Env Filter Decay 

("4" = between 1 and 16 or 20?   --  I really want more definition there.  The jumps for each digit are big.)
« Last Edit: March 29, 2022, 10:30:26 PM by Elric »
:Elric:
Kurzweil K, Pro3, TX81z, K1r, Triton w/MOSS, Wavestation in a box

Re: Attack stage contour for the Amp Envelope
« Reply #3 on: March 30, 2022, 05:00:55 AM »
I've noticed this too.  It's logarithmic, intending to mimic an analog envelope.  But it almost seems too logarithmic, as its slope seems to go up faster and then slow down more than the logarithmic attacks on my other synths.  With a slow attack and a slow decay, it seems to "go flat" at the top before it reaches the decay, or like an envelope that's clipping at the top of the attack.

I don't know what envelope they modeled the Take 5 envelopes after.  I also have an OB-6 but no Prophets.  Are the Prophet 5/10 envelopes like this too?

Maybe changeable slopes could be added in a future firmware update?  It would make a flexible sound design synth even better.

You are right on. If the attack is set at max and all others are at 0 you can see just how ridiculously long the attack contour is and it seems the volume maxs out quite quickly. "flat" as you called it" I hope we can get some more flexibility in a forthcoming update.


This is my first Sequential so I can't speak to that but perhaps its common of Sequentials?
My Deepmind 12 had super flexible envelopes but I wasn't happy with the overall synth so I got rid of it for the take5. My juno60 and various moogs also have pretty smooth attack contours.

Re: Attack stage contour for the Amp Envelope
« Reply #4 on: March 30, 2022, 05:04:44 AM »
(Over from the Pro3 forum, but...  )

Can you use the Mod matrix to modify this?
One of my fav things to do on the Pro 3 is modify the (Filt?, any) Env Decay with the (same) Envelope Amount (Amt).
You can make Curved Decay. With plus (Convex?) or minus (Concave?).
It's not perfect, but I love having some form of contour control.
(I hope it works on the T5.)

On the Pro 3, for the Filter Envelope, it would be:

Env LPF   >>   +/-4   >>   Env Filter Decay 

("4" = between 1 and 16 or 20?   --  I really want more definition there.  The jumps for each digit are big.)

This is an interesting concept and I found this over on the Pro3 board. I tried it with some interesting effects but do I really want to use up an envelope for this?

https://www.presetpatch.com/article/Recursive-Envelope-Modulation-on-Synthesizers

kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #5 on: March 30, 2022, 10:02:19 AM »
This is an interesting concept and I found this over on the Pro3 board. I tried it with some interesting effects but do I really want to use up an envelope for this?

https://www.presetpatch.com/article/Recursive-Envelope-Modulation-on-Synthesizers
I wonder if you'd have to use up an envelope.  Have the envelope modulate the attack stage of itself through the mod matrix, with a negative amount, so as the attack rises, the attack time is reduced, speeding up the attack as it reaches the top.  Depending on the amount, it should allow you to get something closer to linear (or maybe even exponential) curve of the attack stage.

So, as the output of the envelope rises (and wants to slow down based on its built-in slope), the modulation makes it go faster.  Next time I'm synthing I'll give this a try.

Re: Attack stage contour for the Amp Envelope
« Reply #6 on: March 30, 2022, 10:52:04 AM »
This is an interesting concept and I found this over on the Pro3 board. I tried it with some interesting effects but do I really want to use up an envelope for this?

https://www.presetpatch.com/article/Recursive-Envelope-Modulation-on-Synthesizers
I wonder if you'd have to use up an envelope.  Have the envelope modulate the attack stage of itself through the mod matrix, with a negative amount, so as the attack rises, the attack time is reduced, speeding up the attack as it reaches the top.  Depending on the amount, it should allow you to get something closer to linear (or maybe even exponential) curve of the attack stage.

So, as the output of the envelope rises (and wants to slow down based on its built-in slope), the modulation makes it go faster.  Next time I'm synthing I'll give this a try.

Interesting idea. I gave it a try. Check out the attached graphic.

The first is attack at 3pm all others on 0 with no modulation
The second is attack at 3pm all other on 0 with ENV2 modulating ENV2 Att at -127
The third is attack at MAX all others on 0 and no modulation

It looks like this idea prolongs the attack time rather than changing the curve.

kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #7 on: March 30, 2022, 11:24:23 AM »
Interesting idea. I gave it a try. Check out the attached graphic.

The first is attack at 3pm all others on 0 with no modulation
The second is attack at 3pm all other on 0 with ENV2 modulating ENV2 Att at -127
The third is attack at MAX all others on 0 and no modulation

It looks like this idea prolongs the attack time rather than changing the curve.
Looks like it had the opposite effect.  Try a positive modulation (+127 instead of -127).  I assumed a positive modulation equated to turning attack up meaning it slows down, and negative would speed it up, but it appears from your test the opposite is the case.

If that speeds it up without affecting the curve enough, try a lower amount of modulation.

Re: Attack stage contour for the Amp Envelope
« Reply #8 on: March 30, 2022, 12:01:07 PM »
Interesting idea. I gave it a try. Check out the attached graphic.

The first is attack at 3pm all others on 0 with no modulation
The second is attack at 3pm all other on 0 with ENV2 modulating ENV2 Att at -127
The third is attack at MAX all others on 0 and no modulation

It looks like this idea prolongs the attack time rather than changing the curve.
Looks like it had the opposite effect.  Try a positive modulation (+127 instead of -127).  I assumed a positive modulation equated to turning attack up meaning it slows down, and negative would speed it up, but it appears from your test the opposite is the case.

If that speeds it up without affecting the curve enough, try a lower amount of modulation.

I tried that as well. Its an interesting effect but well check it out...

The first is attack at 3pm all others on 0 with no modulation
The second is attack at 3pm all others on 0 with ENV2 modulating ENV2 Att at -127
The third is attack at 3pm all others on 0 with ENV2 modulating ENV2 Att at +127


kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #9 on: March 30, 2022, 12:52:39 PM »

I tried that as well. Its an interesting effect but well check it out...

The first is attack at 3pm all others on 0 with no modulation
The second is attack at 3pm all others on 0 with ENV2 modulating ENV2 Att at -127
The third is attack at 3pm all others on 0 with ENV2 modulating ENV2 Att at +127
Try a (much) lower modulation depth.  Experiment with it, there might be a happy medium that will give you the desired result.  Also experiment with the attack time.

When I get off work I'll try it on mine as well.



kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #10 on: March 30, 2022, 01:20:02 PM »
I was just experimenting with mine.  Attached is a screen shot of the envelope curve as follows:

Env2 Attack at max, Decay/Sustain/Release at zero.
Mod routing:  Env2 -> Env2 Attack.  Mod depths are 0, 32, 64, 80, 96, 127.  Depths over 64 start to get exponential.

The 2nd screen shot has the following routings:

Env2 -> Env2 Attack, depth 80.
DC -> Env2 Attack, depth -46.

The DC applies a negative offset to the attack, to make the effect on the attack time bipolar, which gets even closer to linear, and it also makes the attack time longer with higher Env2->Env2 Attack mod depths.

« Last Edit: March 30, 2022, 01:21:56 PM by kpatz »

kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #11 on: March 30, 2022, 01:55:44 PM »
Here's one more comparison, this time my Take 5's Env 2 slope compared to my OB-6 module, which is the only other Sequential synth I own.

Both synths on a Basic Program, env 2 with decay/sustain/release at 0.  OB-6 attack at max, Take 5 attack set to match as closely as possible (about 3:30 on the knob), as Take 5's envelope can go much longer than OB-6's.

Screenshot shows Take 5 at top, OB-6 at bottom.  Take 5 starts fading in instantly with a rapid attack that slows and then barely moves before it reaches the decay, while OB-6 is silent for a couple seconds before becoming audible and having an audible increase until it reaches the decay.

OB-6's curve is quite different... maybe to match classic Oberheim env curves.  I wonder how a Prophet 6's envelope attack curve compares.

Re: Attack stage contour for the Amp Envelope
« Reply #12 on: March 31, 2022, 05:01:25 AM »
I was just experimenting with mine.  Attached is a screen shot of the envelope curve as follows:

Env2 Attack at max, Decay/Sustain/Release at zero.
Mod routing:  Env2 -> Env2 Attack.  Mod depths are 0, 32, 64, 80, 96, 127.  Depths over 64 start to get exponential.

The 2nd screen shot has the following routings:

Env2 -> Env2 Attack, depth 80.
DC -> Env2 Attack, depth -46.

The DC applies a negative offset to the attack, to make the effect on the attack time bipolar, which gets even closer to linear, and it also makes the attack time longer with higher Env2->Env2 Attack mod depths.

Nice work here. I'll play around with it when I can. But that OB-6 curve would be nice to have access to on the Take5.

Re: Attack stage contour for the Amp Envelope
« Reply #13 on: April 01, 2022, 09:21:11 AM »
I've mentioned this a couple times before in other Seq instrument forums, but -

Feature Request: It would be awesome if Seq could develop Envelope ADR Shaping Controls for Take 5 (and all other Seq instruments going forward)   Based on what I've heard, it seems the envelopes are in the digital domain for the majority (if not all) recent synths?   Having controls to adjust the ADR slope/shape would be amazing -- I would think that any development in this area would be something that could be reused for future instruments... ie: It would be a great IP investment.

From a UI standpoint, it doesn't need a dedicated control/knob... it would be fine for it to just live within a menu, and controlled by a soft knob / encoder.   Just a single control for each of the ADR stages.  In the middle, the envelope would have the default slope.  Counter clockwise would make the envelopes progressively more convex/snappy, and clockwise to make them more concave/linear/progressive.   

This is one of the nicest features that Deepmind, Moog One and some other synths have available, that Seq doesn't currently.   If we have some controls for ADR contours, we can better emulate various synth sounds of past with snappier or smoother contours.... opening up more variety of plucks, leads and bass sounds, as well as smoother progressing strings, pads and evolving sounds.   

Envelopes are really the heart of sound design...  I would put this among the top priorities for feature requests for all existing and forthcoming Seq instruments.

Sequential Pro 3 and Prophet Rev 2, Deepmind, PolyBrute - Sound Sets, Patch Banks - Available on Sellfy: https://sellfy.com/sounddesign/

Elric

  • *
  • 67
  • Peak Oil
Re: Attack stage contour for the Amp Envelope
« Reply #14 on: April 02, 2022, 08:11:11 PM »
I was just experimenting with mine.  Attached is a screen shot of the envelope curve as follows:

Env2 Attack at max, Decay/Sustain/Release at zero.
Mod routing:  Env2 -> Env2 Attack.  Mod depths are 0, 32, 64, 80, 96, 127.  Depths over 64 start to get exponential.

The 2nd screen shot has the following routings:

Env2 -> Env2 Attack, depth 80.
DC -> Env2 Attack, depth -46.

The DC applies a negative offset to the attack, to make the effect on the attack time bipolar, which gets even closer to linear, and it also makes the attack time longer with higher Env2->Env2 Attack mod depths.


Good job kpatz!

It looks like you got from "here" (top image, default Take 2 attack),
 ...to "there" (bottom image, your [self modulated and offset] "tweaked" attack!)
(:Image 1: [your images], I just shrunk one to the scale of other.)

I'd love to see the later curves (eased, "s-curved", 80,96,127) pushed to longer times with something.  They could look more like the OB6 curves.
(:Image 2: [half of one of your images], I just scaled out the last few attack curves.)

- E
« Last Edit: April 02, 2022, 09:07:49 PM by Elric »
:Elric:
Kurzweil K, Pro3, TX81z, K1r, Triton w/MOSS, Wavestation in a box

kpatz

Re: Attack stage contour for the Amp Envelope
« Reply #15 on: April 03, 2022, 04:28:34 PM »
I'd love to see the later curves (eased, "s-curved", 80,96,127) pushed to longer times with something.  They could look more like the OB6 curves.
Using the DC mod source with a negative depth increases the length of the attack, so experiment with both modulation depths (Env2->Env2 Attack and DC->Env2 Attack) to find the optimum balance between attack time and curve.

A higher Env2->Env2 Attack makes the curve more linear or exponential, at the cost of a shorter attack time.  A negative DC offset makes the attack time longer again.

I don't think there's any way to reach the OB-6 curve with the T5, but this trick gets you closer.  Another method would be to use envelope 1 to modulate envelope2's attack, and adjust the attack time on Env1 to shift things (or invert the modulation and use Env1 Decay), but then you're tying up both envelopes.  I know some have done this on the Pro 3, but it has a 3rd envelope available to do this.

Perhaps the goal with the T5 was to have a snappier sounding envelope.  Which is great for snappy sounds, but having the option of a more linear envelope for pads would be nice too.
« Last Edit: April 03, 2022, 04:32:31 PM by kpatz »

Elric

  • *
  • 67
  • Peak Oil
Re: Attack stage contour for the Amp Envelope
« Reply #16 on: April 03, 2022, 06:18:19 PM »
Yea, I've been messing with this for a month or more. (But with the Decay curve.)

...I just did a quick test on the Pro 3. It looks like they are the same/similar envelopes. (That's just a crazy attack curve!)

Here's a quick, notated pic for Attack values and modifications on the Pro3.
(- I'm actually more interested in curving the Decay step of the envelope, so I'll go back to that now. And back to the Pro 3 subforum...  Cheers!)

[Edit:  I wish I had a Take 5 on which to play around with these parameters, but I bet you can get to this eased curve on one of those too.]

[Edit2:  And hey! Look at that!  The Decay curve can snap like nobody's business!!  (Image2, from a Pro3) ]
« Last Edit: April 03, 2022, 07:07:55 PM by Elric »
:Elric:
Kurzweil K, Pro3, TX81z, K1r, Triton w/MOSS, Wavestation in a box

Elric

  • *
  • 67
  • Peak Oil
Re: Attack stage contour for the Amp Envelope
« Reply #17 on: April 07, 2022, 02:40:16 AM »
I think I have a solution/definition for what's going on here, and it's a little bit disappointing.
They are just cutting off the curve (with a quick ease).

It would take me a while to show you exactly what I mean*, but...
It works really well for the Decay, but not so well for the Attack.
(Basically, it's the same ease-timing, but, "sharp then smooth" works much better for decay than attack.)

*I might get back to that.

- E
:Elric:
Kurzweil K, Pro3, TX81z, K1r, Triton w/MOSS, Wavestation in a box

Re: Attack stage contour for the Amp Envelope
« Reply #18 on: April 07, 2022, 11:00:05 AM »
I was just experimenting with mine.  Attached is a screen shot of the envelope curve as follows:

Env2 Attack at max, Decay/Sustain/Release at zero.
Mod routing:  Env2 -> Env2 Attack.  Mod depths are 0, 32, 64, 80, 96, 127.  Depths over 64 start to get exponential.

The 2nd screen shot has the following routings:

Env2 -> Env2 Attack, depth 80.
DC -> Env2 Attack, depth -46.

The DC applies a negative offset to the attack, to make the effect on the attack time bipolar, which gets even closer to linear, and it also makes the attack time longer with higher Env2->Env2 Attack mod depths.

I just wanted to follow up here to say that this really helped me. Thanks again!