ADSR and Filter Cut-Off

[Teoman]

Guys,

Suppose I decided to modulate the filter cut off frequency of the LPF on my synth using the Attack-Decay-Sustain-Release envelope structure. And suppose the filter cut off frequency is set to 150 Hz by default. Here is the question: does the filter cut off frequency start from zero or 150 Hz when the ADSR has just been applied to the filter. In other words, the cut off reaches a maximum when the attack phase is completed and decay just begins. So does this maximum mean 150 Hz or 150 Hz+envelope amount. Any explanation would be greatly appreciated.

 

One more thing: is the cut off frequeny the only parameter to be played with. Can I modulate the resonance(emphasis) by any means.?

[grumphh]

Wow. That is a very theoretical question to ask IMO

Personally i would just play around with the values to see what sounds good, instead of getting into tech details

 

But, without being an expert on the subject, going by my synths, to me it sounds as if the envelope amount is added to the base filter setting (i.e. 150 hz + EG amount) - but i could be totally wrong...

 

 

As for resonance - in most synths i have seen the resonance stays fixed for the duration of a note.

But in some synths (like my Matrix 6r) you can actually route an EG to the resonance to make it a dynamic part of the sound, so that depends on the flexibility of the particular synth you use.

[Yoozer]

Usually it's like this - LFO's are bipolar (means that they use the cutoff as the middle and sweep up, then go back, sweep down, and go up again). Envelopes are unipolar - means that they either only add or subtract on the trajectory they go through.

 

To properly test it - first, listen a patch where the envelope does nothing and the filter's set at 150Hz - then, let the filter envelope value go to the maximum, and give it a very high attack value. That way, the first 2 seconds should be pretty much equal to what you heard in the static sound.

 

Since only the Nord displays the cutoff, I'm assuming you are doing your sounddesign on that; and as far as I know you cannot separately modulate the resonance. The Virus should be able to, though.

[Teoman]

Well, I haven't made such a test with a high attack value. Such a test can obviously tell me if the origin of the ADSR vs time sketch is zero hertz or 150 hertz. But, my logic tells me that there should be a common understanding of ADSR modulation among the manufacturers and the attack part should take the default cut off frequency (150 Hz in my example) and carry it to 150+envelope amount after reaching the max of the attack line. As you said, it is unipolar either additive or subtractive so it may work both ways. LFO case is more obvious, it should start with 150 Hz and swing between let's say 100 and 200 Hz if the envelope amount is set to 50. But I will test it also Thanks for the comments.

[Qwave]

 

does the filter cut off frequency start from zero or 150 Hz when the ADSR has just been applied to the filter

 

 

The envelope control voltage (or its digital equvalent) is allways added to the cutoff set up before (assuming positive modulation). Therefor the filter will be at 150 Hz at the note start and then rise during the attack stage till the maximum set by envelope modulation. And this maximum is the point when the Attack stage ends and the Decay stage starts.

 

I never seen or played a synth that had a different way then this adding the envelope modulation to the cutoff controll value sheme.

 

And I had plenty of synths my self and still have more synths here then my car could transport.

[Teoman]

Thanks Qwave,

your comment made me confident about my guess. thnks mate.

 

Now a bigger question :

I have a longplay from the past belonging to BoneyM (album's name was bonononos if I am not wrong). There is a monophonic synth section in one of the songs. It goes like this: it is a simple two note sequnce like: C3-A3-C3-A3-C3-A3-C3-A3-......while this sequence goes on, the sound changes from a dull character to an extremely brighter tone. While this brightening phase, I guess the equalizer band pass filter gain is increased at high frequencies. This makes the sound evolving in time with a brighter and more treble tone. I hope I could explain what I mean. So, how can I accomplish such a sound? In this example, ADSR and LFO modulation obviously cannot be used for filter opening because these modulation sources are triggered at every key stroke which is not the case in my example. So, my guess is this: it was a sound playing the sequence with either constant cut-off frequency and an equalizer automation (to make the sound continuously more treble with time) or with a cut off modulation independent of the keyboard, but evolving linearly with time by some means and some equalizer gain increased parallel to the cut off. I hope I can get some comments to this fictitious example

[toorglick]

It depends on the manufacturer and the synth. I've only see it where the envelope's peak is the value you've set for cutoff (meaning the envelope starts at zero).

 

The same applies to amplifier envelopes and it's much easier to demonstrate why: the "output level" of the oscillator is the maximum the amp envelope will go, otherwise what would be the point of having an "output level" parameter? Without that kind of control over the peak level it would be much more difficult to create a patch.

 

LFOs are a different story, they, in most case, do indeed add a little beyond the threshold set by the filter cutoff or output level, but not too much.

 

I've used Korgs and Rolands, mainly, for years and this is how they all worked. My ESQ 1 operates the same way. My K-Station plugin works this way, as well as most of my other softsynths as far as I recall. I know the Darkstar did as well. I'm pretty sure it's common for the envelope's peak threshold to be set by the cutoff and output parameters (respectively).

[Qwave]

just move the cutoff while the sequence is played to get this FX. There were no automated EQs in the time of Boney M.

You may use a second envelope to cutoff modulation to automate this. Or use some very slow LFO to cutoff to get this changing sound. This is a very, very common sound programming for analog synths. And If you LFO maybe synced to MIDI clock, you get these special slow cutoff modulation (on top of the envelope modulation) synced to the sequencer tempo. Very nice thing to try.

[grumphh]

Originally posted by teoman Thanks Qwave, your comment made me confident about my guess. thnks mate. Now a bigger question : I have a longplay from the past belonging to BoneyM (album's name was bonononos if I am not wrong). There is a monophonic synth section in one of the songs. It goes like this: it is a simple two note sequnce like: C3-A3-C3-A3-C3-A3-C3-A3-......while this sequence goes on, the sound changes from a dull character to an extremely brighter tone. While this brightening phase, I guess the equalizer band pass filter gain is increased at high frequencies. This makes the sound evolving in time with a brighter and more treble tone. I hope I could explain what I mean. So, how can I accomplish such a sound? In this example, ADSR and LFO modulation obviously cannot be used for filter opening because these modulation sources are triggered at every key stroke which is not the case in my example. So, my guess is this: it was a sound playing the sequence with either constant cut-off frequency and an equalizer automation (to make the sound continuously more treble with time) or with a cut off modulation independent of the keyboard, but evolving linearly with time by some means and some equalizer gain increased parallel to the cut off. I hope I can get some comments to this fictitious example

If you have a slider for the cutoff frequency, you could just do it by hand? (Which is what i assume they did in the good old days before midi automation)

 

Second, again, if you have a synth that is flexible enough, you could route a slow free-running LFO with an upsaw shape (or on the Matrix a "ramp generator") to the cutoff frq to get the brightening effect over time.

[toorglick]

 

I hope I could explain what I mean. So, how can I accomplish such a sound? In this example, ADSR and LFO modulation obviously cannot be used for filter opening because these modulation sources are triggered at every key stroke which is not the case in my example.

 

 

Just modulate the filter with the LFO and turn off triggering.

[Qwave]

 

Originally posted by toorglick It depends on the manufacturer and the synth. I've only see it where the envelope's peak is the value you've set for cutoff (meaning the envelope starts at zero)...

 

 

it's not like this on these synths I played or owned once (not complete):

Moog Prodigy,

Moog Source,

Minimoog,

Memory Moog,

Moog Rogue,

PPG wave 2.2,

PPG wave 2.3,

Waldorf Q,

Waldorf WAVE,

Waldorf microQ,

Waldorf XT,

Waldorf microWave,

Modulus MonoWave,

Modulus MonoWave II Chameleon sound skin,

Alesis A6 Andromeda,

Alesis ION,

EMS VCS 3,

EMS synthi A,

SCI Prophet 5,

SCI T8,

SCI Pro One,

Poland Jupiter 8,

...

 

I am eager to know what synth is starting the cutoff from zero (by default or all the time) if envelope modulation is used ?

Toorglick please name one. I am willing to learn.

 

Actually a synth that would do so would be boring. All sounds would start with a very much filtered, and close to sine wave, sound all the time.

 

On real analog synths it is like this in terms of tech:

- the cutoff knob adds a positiv controll voltage to the cutoff modulation input point

- the envelope is also patched to this point

 

This results in:

- if the cutoof isn't set to zero, the filter will not drop below this set cutoff frequency undless you apply negative envelope voltage. By default all envelopes I ever encountered are producing positive volatges starting from zero to their max during the attack time.

 

So how to get the cutoff modulated to zero on note of without inverting the envelope?

[Qwave]

Originally posted by toorglick I have a big penis, yes, but, sure, yours may be bigger.

 

 

You are a childish, aren't you ? Maybe you are not even grown up yet.

 

Pleople posting these low level words should be banned. I wonder how you made it to become a SENIOR MEMBER by posting these rubish you just posted.

[toorglick]

Okay, sorry, lack of sleep clouds my mind.

 

You're right, what I wrote doesn't sound accurate. I was thinking in terms of EGs and how they start at zero.

 

But it doesn't sound right to me that the cutoff is added to by the envelope. I could be completely wrong as my memory is bad.

[toorglick]

Originally posted by Qwave You are a childish, aren't you ? Maybe you are not even grown up yet. Pleople posting these low level words should be banned. I wonder how you made it to become a SENIOR MEMBER by posting these rubish you just posted.

 

If you look at my registration date, you'll note that with this particular forum software I've been posting rubbish for four years. Before this version of the software I was posting rubbish for at least an additional year. Maybe rubbish is the way of the future?

 

Who knows; all I know is that I don't feel like getting into a "my knowledge and experience is greater and better than your knowledge and experience" tit-for-tat game. Yes, you're right, I confused my facts. Sorry. Like I said, lack of sleep will do that, but there's no need to start dick-waiving over it.

[Teoman]

toorglick,

If the envelopes start from zero and reach to the predefined cut off value, then what would the envelope amount be used for? If we assume what you are saying is correct, then each time a key is pressed cut off goes to 150 Hz in my previous example and the envelope amount parameter loses its meaning.

 

For the BoneyM song example, I would really like to thank all of you guys. But there is something I am missing: If I can accomplish such an evolving brigthness effect either manually opening the filter or by modulating it by a very slow LFO ramp signal, then What would happen to the sound if it is already modulated by an envelope generator or an LFO triggered by a key stroke. I mean the filter cut off will change at each key stroke following either an envelope or an LFO ramp, then I will open the filter one more time manually or by a slow LFO. I can't guess how these things will interact

 

BTW, I learnt that Moog Modular V has a resonance modulation feature in its multimode filter. So I will try to use this resonance increase with a cut off increase.

[Qwave]

Toorglick,

 

I wonder how you made it to become a SENIOR MEMBER by posting these rubish you just posted

 

I was refering to your posting with the bad word p**is inside. I wasn't refering to on topic postings of yours.

 

But it doesn't sound right to me that the cutoff is added to by the envelope.

 

Check some analog synths shematics available online. Maybe start with the Minimoog.

 

Have a nice day and try to get some more sleep if possible

[Qwave]

Originally posted by teoman I mean the filter cut off will change at each key stroke following either an envelope or an LFO ramp, then I will open the filter one more time manually or by a slow LFO. I can't guess how these things will interact

 

Its not "either...or" here. Its "and".

 

Lets say the envelope openes the filter from the set cutoff of 150 Hz to a max of 2 kHz on each note. Using a slow LFO in addition will change the max cutoff per note to 4 kHz or near zero depending on:

- is it a bipolar controlvoltage (like a LFO is most of the times) ?

- the added amount of the LFO modulation

 

Think like this:

all the times the modulations on the same parameter (here "cutoff") will be summed together. Including the manual cutoff parameter/knob value. The result of these is setting the actual cutoff value of the filter. Do this about 1000 times per second on a VA synth or infinite on a real analog synth.

 

BTW: what synth are you using ?

[Teoman]

Well I am planning to do this on my Moog Modular V. I may add the same ramp LFO modulation to the resonance parameter as well to see what happens. I see what you mean by two modulations at the same time to the cut off. What you suggest will definitely work if I adjust LFO depth high enough to exceed beyond the max level of envelope. (reaching 4 Khz, above the 2 Khz upper limit of the key triggered envelope modulation in your example). Or doing this together with turning off the triggering as toorglick suggested will also work

 

P.S: I may try this on my coming A6 if sounds good enough.

[Qwave]

 

...What you suggest will definitely work if I adjust LFO depth high enough to exceed beyond the max level of envelope...

 

 

No, you don't get me right.

even the slightes LFO added to the filter cutoff will change the filters cutoff. No need to modulate by a bigger amount then the envelope. The modulations are added. So why boost the LFO more then the envelope. Of cause increasing the amount will make it more audible. But it (the LFO modulation) will change even if added only by a tiny bit.

[mrcpro]

 

Originally posted by teoman I have a longplay from the past belonging to BoneyM (album's name was bonononos if I am not wrong). There is a monophonic synth section in one of the songs. It goes like this: it is a simple two note sequnce like: C3-A3-C3-A3-C3-A3-C3-A3-......while this sequence goes on, the sound changes from a dull character to an extremely brighter tone. While this brightening phase, I guess the equalizer band pass filter gain is increased at high frequencies.

 

 

I'm not familiar with BoneyM, but it sounds like a square wave trill to me. Modulate VCO frequency with a square wave LFO until you get the interval you want (three semitones). The notes will now not be in tune, so you will have to re-tune the VCO. Now, you can sweep the filter with an envelope and it won't keep re-triggering.

[Markeyz]

Well I didn't carefully read all of the posts above and so I might have missed this solution for a slowly opening filter over multiple notes but if the synth you are using has a legato mono mode just use that so you don't retriggering envelopes until you let go of all of the keys.

[J3RK]

 

For the BoneyM song example, I would really like to thank all of you guys. But there is something I am missing: If I can accomplish such an evolving brigthness effect either manually opening the filter or by modulating it by a very slow LFO ramp signal, then What would happen to the sound if it is already modulated by an envelope generator or an LFO triggered by a key stroke. I mean the filter cut off will change at each key stroke following either an envelope or an LFO ramp, then I will open the filter one more time manually or by a slow LFO. I can't guess how these things will interact

 

 

You may be able to accomplish this sweep using parallel filters. Put an envelope on one of the filters, and sweep the other. Then mix the signal from both until it sounds right. That's probably not the technique used on the track you're speaking of, but it might accomplish a similar result. I've done this on my ION a bit.

[J3RK]

Or, maybe you could route two filters in series, applying the envelope to the first filter, and then have that run through a slowly sweeping second filter.

[Teoman]

Originally posted by J3RK You may be able to accomplish this sweep using parallel filters. Put an envelope on one of the filters, and sweep the other. Then mix the signal from both until it sounds right. That's probably not the technique used on the track you're speaking of, but it might accomplish a similar result. I've done this on my ION a bit.

 

I think you mean that I should use one filter in normal envelope modulated mode while modulating the cut off of the parallel one using a slow LFO ramp or something like that and then combine them in the right proportion. That sounds good Thnx mate.

 

MarkeyZ, your suggestion is definitely a nice one. Using legato/mono mode while playing, it would be possible to keep the cut off at the same level and open the filter further by any method to accomplish what I want. Thanks.

 

Qwave, yeah it was my misunderstanding. LFO depth or envelope amount has nothing to do with the actual cut off setting. Either it is a small or big value, will definitely add to the cut off.

 

mrcpro, your technique sounds interesting, but I doubt I can achieve good results especially with my current expertise

 

As a last note, I still believe such a gradual filter opening (independent of the notes played) can be much more pronounced

by some other additional adjustments. It may be increasing the resonance in parallel to the cut off without entering into self oscillation or it may be amplifying the high frequency components gradually at the equalizer stage while letting more and more high frequency components let pass through the filter. It requires some automation, but I will give it a try

 

Thanks for all these valuable info guys.

[mrcpro]

Originally posted by teoman mrcpro, your technique sounds interesting, but I doubt I can achieve good results especially with my current expertise

 

I urge you to give it a try anyway...what have you got to lose? It isn't some crackpot novel idea of mine, it's the standard established way to get the sound you describe on an analog synth. These sounds...using square wave LFOs for trills and saw LFOs for effects like spaceships were all over the place in the 70's...in fact, they were so common that Oberheim put an LFO offset in Page 2 of their OB8 so you didn't have to re-tune your VCO's after getting your interval set...but it's pretty easy to re-tune the VCO's anyway. The LFO offset idea still shows up on some current synths...my XP-60 has this paramater...kinda nice if I want to freak out and go 70's...