Help me understand relationship between env amt and dest amt

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
RandomSkratch
Posts: 448
Joined: 10 May 2016

10 May 2016

I posted this in the Reason sub reddit but didn't get any replies (either no one knows the answer or no one cares?). Thought I'd post it on here just in case someone can answer it!

I've been actively trying to figure out filter envelopes for a few weeks now and I somewhat understand them but not fully. (Amp envelopes are easy for some reason because they don't have an amount knob). For this example I'll reference Subtractor. Attack/Decay/Release I get because they are time based (off topic I wish Props would update Sub/Mael to read like Thor because it uses ms values and not arbitrary 0-127 values). It's the Amount and Sustain knobs and how they relate to the existing filter settings that's what's confusing me. Here's what I know so far (whether it's accurate or not is tbd).

The setting on the filter is the base setting. The envelope cannot make it go below this. This is the new "0".

The envelope amount knob will set the upper range of the envelope. In other words, the new top "127" value. (First question, how do the number values relate to the filter values? If the filter cutoff is set to 50 and I apply an amount of 50, will the result be 100? Or is some form of scaling applied?)

Attack is the time to go from "0" setting to amount value + "0" setting. (easy)

Decay is the time to go from previous peak back down to "0" setting UNLESS sustain is set. (easy)

Sustain is the point the signal decays to but like the amount value, how does this translate to the set cutoff value? Is it added on top? I can't quite articulate this but like my example above, if the cutoff is set to 50 (with a previous total of 0-127), it's now 50-127. And the Sustain value can be 0-127, does the sustains 0-127 scale to fit within the existing cutoff 50-127? Does that make sense? Also, how does the amount value affect this sustain value?

Release is time to go from sustain value to "0". (easy).

Normally the Props write great manuals and I've gone over them a few times but cannot wrap my head around it. I wish they would implement a knob/slider animation option (switch on back of rack or something) so you could see how much your filter is being modified. Most of the time I know the sound I want to get based on number values and I can usually get it by ear with a bunch of fiddling, but if I knew the relationship between the values I could dial it in much quicker.

I also don't have much experience with other synths and I don't know if understanding the Reason way will help translate to other synths... (I have Massive/Monark but they seem to operate differently.. Massive's envelopes are a lot easier to dial in with the coloured range dials, easy to see!)

Thanks!

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

10 May 2016

As you say, attack, decay and release is TIME but sustain is a little different, it defines the LEVEL after the decay. So the signal goes to full level in [attack time] then goes to the sustain level in [decay time] and then when you release the key it goes down to zero in [release time]. The envelope amount only says what percentage of these values gets applied to whatever you apply the envelope to.

So if you apply 50% env to the filter and have the filter at zero in your patch it will open 50% when the attack time has passed. If the filter is at 50% in your patch it will open completely, even if you apply only 50% env.

Stranger.
Posts: 329
Joined: 25 Sep 2015

10 May 2016

ΣΣΣ
Last edited by Stranger. on 03 Jun 2016, edited 1 time in total.

RandomSkratch
Posts: 448
Joined: 10 May 2016

10 May 2016

normen wrote:As you say, attack, decay and release is TIME but sustain is a little different, it defines the LEVEL after the decay. So the signal goes to full level in [attack time] then goes to the sustain level in [decay time] and then when you release the key it goes down to zero in [release time]. The envelope amount only says what percentage of these values gets applied to whatever you apply the envelope to.

So if you apply 50% env to the filter and have the filter at zero in your patch it will open 50% when the attack time has passed. If the filter is at 50% in your patch it will open completely, even if you apply only 50% env.
You reference percentage but the Amount knobs are in values 0-127. Should I be converting this to percentage? Or is it numerical value based? Ie - set to 50 would mean to add 50 to the current value of whatever you're applying it to?

RandomSkratch
Posts: 448
Joined: 10 May 2016

10 May 2016

Stranger. wrote:
RandomSkratch wrote:(First question, how do the number values relate to the filter values? If the filter cutoff is set to 50 and I apply an amount of 50, will the result be 100? Or is some form of scaling applied?)

I also don't have much experience with other synths and I don't know if understanding the Reason way will help translate to other synths... (I have Massive/Monark but they seem to operate differently.. Massive's envelopes are a lot easier to dial in with the coloured range dials, easy to see!)

Thanks!
The filter cutoff is basically your frequency Hz..so you apply an amount of 50 @ say,50hz for eg( it is not 50hz though.)
Sadly reason has to scale everything-including the envelopes and mix faders-and some envelopes even seem to foldback ==== this is not good imo,and do not work like other daw's or vst's etc. :roll:
But on the subtractor it's not a frequency Hz value... it's 0-127..

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

10 May 2016

RandomSkratch wrote:You reference percentage but the Amount knobs are in values 0-127. Should I be converting this to percentage? Or is it numerical value based? Ie - set to 50 would mean to add 50 to the current value of whatever you're applying it to?
Its not really defined I guess. 0 would be 20 Hz and 127 20kHz I suppose, with a logarithmic scale. Easy to check with the analyzer in the EQ though. And yeah, 0 would be 0% and 127 would be 100%

Stranger.
Posts: 329
Joined: 25 Sep 2015

10 May 2016

ΣΣΣ
Last edited by Stranger. on 03 Jun 2016, edited 1 time in total.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

10 May 2016

Stranger. wrote:@normen- i suspect it's more exponential,rather than linear scaling in subtractor and some other devices also tbh.
Exponential? That doesn't make much sense, as I said its probably logarithmic, like most frequency displays are (and how the human ear actually perceives it), e.g. the analyzer in Reason.

This is the same frequency plot logarithmic and linear:

Image

RandomSkratch
Posts: 448
Joined: 10 May 2016

10 May 2016

normen wrote:As you say, attack, decay and release is TIME but sustain is a little different, it defines the LEVEL after the decay. So the signal goes to full level in [attack time] then goes to the sustain level in [decay time] and then when you release the key it goes down to zero in [release time]. The envelope amount only says what percentage of these values gets applied to whatever you apply the envelope to.

So if you apply 50% env to the filter and have the filter at zero in your patch it will open 50% when the attack time has passed. If the filter is at 50% in your patch it will open completely, even if you apply only 50% env.
I just reread what you wrote and I think you contradicted yourself (but at the same time enforced my assumptions about the behaviour of the envelope).

When you said if you apply 50% to the filter and have the filter at 0 it will open to 50% (That I always thought to be true). But when you said if the filter is at 50% to start, it will open completely if you only apply 50% envelope, but that wouldn't be true if it was a percentage based because it would only open to 75%. If it would open to 100% that means the numbered values of the sliders are being added at face value (64 + 64 = 128). Now I'm not a mathematician mind you and I could be completely wrong here...

I really wish I could see the amount being applied (or even have some feedback from a developer at Propellerheads...) Then there wouldn't be any room for interpretation!

RandomSkratch
Posts: 448
Joined: 10 May 2016

10 May 2016

Stranger. wrote:
RandomSkratch wrote:But on the subtractor it's not a frequency Hz value... it's 0-127..
Correct-
JFYI- with subtractor_ the filter envelope is always active,even with 0 amount being applied.

@normen- i suspect it's more exponential,rather than linear scaling in subtractor and some other devices also tbh.
What do you mean it's always active? As in it's always in the signal path? Or an amount is always being applied, even with 0? I don't understand...

User avatar
Benedict
Competition Winner
Posts: 2747
Joined: 16 Jan 2015
Location: Gold Coast, Australia
Contact:

10 May 2016

Benedict Roff-Marsh
Completely burned and gone

User avatar
selig
RE Developer
Posts: 11841
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

11 May 2016

RandomSkratch wrote:
normen wrote:As you say, attack, decay and release is TIME but sustain is a little different, it defines the LEVEL after the decay. So the signal goes to full level in [attack time] then goes to the sustain level in [decay time] and then when you release the key it goes down to zero in [release time]. The envelope amount only says what percentage of these values gets applied to whatever you apply the envelope to.

So if you apply 50% env to the filter and have the filter at zero in your patch it will open 50% when the attack time has passed. If the filter is at 50% in your patch it will open completely, even if you apply only 50% env.
I just reread what you wrote and I think you contradicted yourself (but at the same time enforced my assumptions about the behaviour of the envelope).

When you said if you apply 50% to the filter and have the filter at 0 it will open to 50% (That I always thought to be true). But when you said if the filter is at 50% to start, it will open completely if you only apply 50% envelope, but that wouldn't be true if it was a percentage based because it would only open to 75%. If it would open to 100% that means the numbered values of the sliders are being added at face value (64 + 64 = 128). Now I'm not a mathematician mind you and I could be completely wrong here...

I really wish I could see the amount being applied (or even have some feedback from a developer at Propellerheads...) Then there wouldn't be any room for interpretation!
There are REs that help visualize CV, or you can use Thor to convert to audio and record as an audio track (still my fave, because you can see as long a time frame as you need).

As for the 50% issue, what Normen is saying is that the "amount" value represents the total range, so this value would not change based on the starting point as you suggest.

In other words, and providing a more specific example… if the filter's range is 10 octaves, the the full mod amount is "supposed" to represent this entire range. For example: If you set the "amount" to 100% you'd get 10 octaves of modulation. If "amount" was set to 50%, you'd get half that amount, or 5 octaves. Doesn't matter what the starting frequency is, 50% modulation would (in this example) always give you 5 octaves of modulation no matter where the filter is set. Make sense?
:)
Selig Audio, LLC

Stranger.
Posts: 329
Joined: 25 Sep 2015

11 May 2016

ΣΣΣ
Last edited by Stranger. on 03 Jun 2016, edited 1 time in total.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

11 May 2016

Yeah as Selig said, percentage of the values generated by the envelope.

RandomSkratch
Posts: 448
Joined: 10 May 2016

12 May 2016

Thanks Benedict. I've recently found your site and enjoy reading the articles. The one linked, although helpful, doesn't really answer my questions about the relationship the amount knob has on the destinations - It would seem that the way things are being described using a percentage compared to what is represented in the software (0-127) is what's confusing. Should I forget the fact that it says 0-127 and think about it as 0-31-63-95-127 = 1-25%-50%-75%-100%? And do these get applied by simple addition? Or is it scaled based on initial value of filter cutoff?
normen wrote:Yeah as Selig said, percentage of the values generated by the envelope.
Wot? :?
Stranger. wrote: Here's some preeety pics to illustrate some subtee action...hope this helps.>
I wish they did! :shock:

Maybe if I rephrase my question the answers would be easier to understand?

For simplicity's sake, I'll use specific scenarios with easy numbers. If someone is able to answer these I think I'll understand the relationship better.

Scenario 1.
a. Filter Cutoff is set to 0. (0%)
b. Filter envelope amount is set to 63. (50%)
c. Envelope sustain is set to 0. (0%0

At the top of the peak of the attack stage, what value will the filter cutoff be at?
After decay, what value will the filter cutoff be during the sustain?

Scenario 2.
a. Filter Cutoff is set to 31. (25%)
b. Filter envelope amount is set to 95. (75%)
c. Sustain is 63. (50%)

Cutoff at top of attack?
Cutoff during sustain?

Scenario 3.
a. Cutoff is 63. (50%)
b. Envelope amount is 63. (50%)
c. Sustain is 0. (0%0

Cutoff at top of attack?
Cutoff during sustain?

Scenario 4.
a. Cutoff is 31 (25%)
b. Env amount is 31. (25%)
c. Sustain is 95. (75%)

Cutoff at top of attack?
Cutoff during sustain?

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

12 May 2016

1) 50/0
2) 100/75
3) 100/50
4) 50/100

You were asking what percentage the env amount refers to, it refers to percentage of the envelope values. The actual value depends on what you apply the envelope to (and the base value that is set on that parameter).

RandomSkratch
Posts: 448
Joined: 10 May 2016

12 May 2016

normen wrote:1) 50/0
2) 100/75
3) 100/50
4) 50/100

You were asking what percentage the env amount refers to, it refers to percentage of the envelope values. The actual value depends on what you apply the envelope to (and the base value that is set on that parameter).
Okay those answers confirm that all you need to do is add the values together to get the final value for the first question (How much the filter opens) but it leaves the second question still a mystery. Reason I say this is because audibly, the Sustain value is effected by the Amount knob but your answers don't factor this into the equation.

If Filter Cutoff is 0 and amount is 25% and sustain is 50%, by your math the sustaining cutoff freq will be at 50% however if you set amount to 50%, using the same math the cutoff freq should still be at 50% however if you listen to it, it sounds higher than that... (unless I'm imagining things.. also I don't have Reason in front of me at the moment, only going by memory).

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

12 May 2016

Oh yeah, I actually made a mistake there, of course the sustain would be modified by the amount percentage..

So sustain would be
2) 62,5
4) 43,75

Anyway, normally people listen and adjust :)

RandomSkratch
Posts: 448
Joined: 10 May 2016

12 May 2016

normen wrote:Oh yeah, I actually made a mistake there, of course the sustain would be modified by the amount percentage..

So sustain would be
2) 62,5
4) 43,75
Could you show me the math you used to get those? I got different values...
normen wrote:Anyway, normally people listen and adjust :)
:roll: I know I know I know... lol music is listened to with your ears so should be written with your ears right?

This is more of my ingrained urge to understand the how and why in everything I do!

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

12 May 2016

RandomSkratch wrote:
normen wrote:Oh yeah, I actually made a mistake there, of course the sustain would be modified by the amount percentage..

So sustain would be
2) 62,5
4) 43,75
Could you show me the math you used to get those? I got different values...
normen wrote:Anyway, normally people listen and adjust :)
:roll: I know I know I know... lol music is listened to with your ears so should be written with your ears right?

This is more of my ingrained urge to understand the how and why in everything I do!

For 2) it would be 50 x 0.75 + 25

RandomSkratch
Posts: 448
Joined: 10 May 2016

12 May 2016

normen wrote:
RandomSkratch wrote:
normen wrote: For 2) it would be 50 x 0.75 + 25
Oh I see it, yeah we did the same math but I used the numerical value of 63 instead of the percentage. It's the same equation I used.

Makes sense... thanks very much for helping me out normen! (Also thanks to everyone else who chimed in)

Stranger.
Posts: 329
Joined: 25 Sep 2015

12 May 2016

ΣΣΣ
Last edited by Stranger. on 03 Jun 2016, edited 1 time in total.

RandomSkratch
Posts: 448
Joined: 10 May 2016

12 May 2016

Stranger. wrote:
Stranger. wrote:Moving a combinaton of both the decay and sustain while a note is being triggered,will raise and lower the filtering amounts dynamically..but if the sustain raises or lowers more or less than the decay-it gets and new decay section back to sustain level (this seems confusing,but is not when you 'get it' ;)
Maybe your over thinking stuff- just listen and let that decide when your creating "music" :thumbs_up:
The most powerful filter is you-yourself.

PH handle the technical side-but choose not share 'how they roll' - only they have the answers you require in totality-the real math and equations used in the programme.
I probably am over thinking it but that's what I do and it's satisfying to get the answers.

I'm not sure the math would or should be considered top secret, it's kind of important in order to understand the product.

If you look at other softsynths (for example Massive) the way the envelopes are applied are shown in such a way that it doesn't leave any room for interpretation.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

12 May 2016

RandomSkratch wrote: If you look at other softsynths (for example Massive) the way the envelopes are applied are shown in such a way that it doesn't leave any room for interpretation.
Mh, often times its not really clear if theres logarithmic curves displayed linearly, linear curves displayed logarithmically or anything in between ;)

Stranger.
Posts: 329
Joined: 25 Sep 2015

12 May 2016

ΣΣΣ
Last edited by Stranger. on 03 Jun 2016, edited 1 time in total.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: luckygreen and 8 guests