Antidotes "special" envelopes

[satyr32]

antidote.jpg (144.08 KiB) Viewed 1072 times

I was making patches with different synths. When I fiddled around with Antidote and discovered a "strange" behaviour.
I expected the decay time to be the time the env goes from the peak to the sustain level which is also defined in the manual that way: "During the decay stage, the envelope falls back to a lower level, the sustain level. The DECAY control specifies the duration of the decay stage, i.e. how long
it takes to fall back to the sustain level."

But this is only true if Sustain is 0. I tested other synths and env generators (e.g. Thor, Movement, Ammo, ...) and all of them behave as expected.
I made a test (letting the key pressed) with A 0, D 3 seconds, S 50%, R 0 and with A 0, D 3 seconds, S 0, R 0, see the image attached. With Antidote if Sustain is NOT 0, the decay time is always shorter than the time you adjust. The higher the Sustain the shorter the Decay Time.

What do you think? Is this a bug or is this intended? If intended, why? I mean it is cool that you have different optiions, but it would be interesting what the developer had in mind.

[selig]

antidote.jpg

I was making patches with different synths. When I fiddled around with Antidote and discovered a "strange" behaviour.
I expected the decay time to be the time the env goes from the peak to the sustain level which is also defined in the manual that way: "During the decay stage, the envelope falls back to a lower level, the sustain level. The DECAY control specifies the duration of the decay stage, i.e. how long
it takes to fall back to the sustain level."

But this is only true if Sustain is 0. I tested other synths and env generators (e.g. Thor, Movement, Ammo, ...) and all of them behave as expected.
I made a test (letting the key pressed) with A 0, D 3 seconds, S 50%, R 0 and with A 0, D 3 seconds, S 0, R 0, see the image attached. With Antidote if Sustain is NOT 0, the decay time is always shorter than the time you adjust. The higher the Sustain the shorter the Decay Time.

What do you think? Is this a bug or is this intended? If intended, why? I mean it is cool that you have different optiions, but it would be interesting what the developer had in mind.

You are describing the difference between a Rate vs a Time based envelope. Antidote appears to use a rate, while the others use a time. What this means is that the slope (rate) with Antidote is always the same no matter the length. With the others, the time is the constant, meaning it will always take the same amount of time no matter the distance traveled.

This is somewhat like pan laws, where the only time you would notice the difference would be if the sustain was being modulated between key presses. Otherwise, you set the control to where the decay/release sounds right - I've never used the parameter values as absolute, speaking for myself only.

[Koshdukai]

I even compared the amplitude of the audio to be sure that the CV output was really representative of what's going on internally with the Envelope generator.

Release does respect the 3sec but Decay has indeed an unexpected behavior.

It's as if the Decay time is always calculated for Sustain=0 as reference but respecting the current Sustain level once the Envelope value reaches it.

Antidote Decay vs Others.png (124.8 KiB) Viewed 1046 times

So... the Decay value unit shouldn't be in absolute ms/sec but in % to "fix" this?

[dioxide]

You are describing the difference between a Rate vs a Time based envelope. Antidote appears to use a rate, while the others use a time. What this means is that the slope (rate) with Antidote is always the same no matter the length. With the others, the time is the constant, meaning it will always take the same amount of time no matter the distance traveled.

I don't get this. I'll have to have a mess around with this and try and figure it out. I've not heard of this before.

[satyr32]

Selig, I guess you are right. I tried out CV Mutant which is based on rate and it is a similiar behaviour. I always knew that something is different with Antidotes envelope, because I often design similiar sounds on different synths and Antidote was always the one with the most different results.

[bpmorton]

Rate/level envelopes were a major thing on the DX7. If you own PX7 you will find the same rate level envelopes. I personally hate them because if you decide to change the level then the time changes too and has to be adjusted.

[selig]

Rate/level envelopes were a major thing on the DX7. If you own PX7 you will find the same rate level envelopes. I personally hate them because if you decide to change the level then the time changes too and has to be adjusted.

Was just going to mention the DX series with it's rate/level envelopes. Ultimately having the option would be the best solution, as it is with glide IMO.

[selig]

Selig, I guess you are right. I tried out CV Mutant which is based on rate and it is a similiar behaviour. I always knew that something is different with Antidotes envelope, because I often design similiar sounds on different synths and Antidote was always the one with the most different results.

Envelope times, especially on amplitude, are never 100% accurate anyway. For amplitude release times the perceived time will vary depending on the volume you are monitoring at.

The envelope curve can also make a HUGE difference in the perceived time, skewing the time shown on the knob even more. I suggest the values are at best just a guideline, one I've never relied on for any purpose. This is definitely a case of "use your ears" IMO, as the values (even on a time based envelope) CAN be misleading in certain settings.

[Exowildebeest]

Rate/level envelopes were a major thing on the DX7. If you own PX7 you will find the same rate level envelopes. I personally hate them because if you decide to change the level then the time changes too and has to be adjusted.

Yeah those envelopes are a pain in the ass - so counter-intuitive if you've been conditioned with normal envelopes all your life.

[ScuzzyEye]

Yeah those envelopes are a pain in the ass - so counter-intuitive if you've been conditioned with normal envelopes all your life.

As someone who's recently been coding envelope controls, I understand Yamaha's motivation. When people say, that adjusting the level also means adjusting the time, that's exactly what my code for a time based envelope does. When the user changes the sustain level, the decay rate value has to be recalculated too.

[buddard]

It's as if the Decay time is always calculated for Sustain=0 as reference but respecting the current Sustain level once the Envelope value reaches it.

Yeah, that's what I'm seeing as well: In the scope image you can clearly see an angle where the decay curve meets the sustain level. If it was a "true" rate-based envelope it would converge to the sustain level (like it does when sustain is 0). So I think they always calculate the decay curve to reach sustain level 0 (or actually slightly below 0), and just wait for it to reach the sustain level before triggering the next stage of the envelope.

This effectively makes it a rate-based envelope of sorts, but there's a subtle difference here since the curve is exponential. So the higher the sustain level, the more linear the decay curve will appear...

This is actually a pretty common method for generating varying response curves when designing envelopes. By varying the "overshoot amount" (but usually compensating for time differences) you can smoothly vary the curve type between linear and exponential.

[selig]

Rate/level envelopes were a major thing on the DX7. If you own PX7 you will find the same rate level envelopes. I personally hate them because if you decide to change the level then the time changes too and has to be adjusted.

Yeah those envelopes are a pain in the ass - so counter-intuitive if you've been conditioned with normal envelopes all your life.

I've been "conditioned" by "normal" envelopes all my life, pretty much non-stop for 40 years now! Funny but even though I spend tons of time with Antidote I've NEVER been bothered by this. Never even noticed it (is that bad?). Must have created hundreds of patches by now. Guessing it's a personal thing?

[eXode]

I've been "conditioned" by "normal" envelopes all my life, pretty much non-stop for 40 years now! Funny but even though I spend tons of time with Antidote I've NEVER been bothered by this. Never even noticed it (is that bad?). Must have created hundreds of patches by now. Guessing it's a personal thing?

Never noticed it, or thought about it either.

[lowpryo]

interesting find, but yeah I would have never noticed! i don't think i've ever adjusted an envelope setting without also constantly mashing a key to hear the effect. visual feedback from knobs and sliders alone has never been enough for me.