Quick update, in less than a minute lol, but I wanted to keep what I wrote above for reference...

[quotePost id=114858]Hey @Anthony, thanks again for these great examples and clarifications... One thing I did not get though:

In Mono mode, if the MODX detects you are playing Legato, it knows not to retrigger the Attack and Decay of the Element Envelope Generators. In effect it "leaves the EG's open" at the Sustain (Decay2) Level.

Do you mean for a normal part?
If I take any of your examples, which btw 1 to 4 sound so smooth during transitions, and put the attack to 0, I will hear a plucky sound, right? But based on what you are saying, if I play legato, I should not hear this pluck again, assuming the Decay2 is not plucky, but I do hear this pluck regardless! So bit confused here.

All that said... if you start delving into FM-X you will find some truly whacky shapes and uses of EG's... most definitely not "normal".

FM-X or any FM is something I never tried before, now I own a MODX, I have to.

[/quotePost]

Skipping to Decay2 during Legato seems to only happen if Portamento is On, i.e. Part SW is On, even if time is zero.

Update #3 - Sorry questions keep coming as soon as I post. Question this time:

I have the feeling that the "Mono/Legato Checkbox" I and other users have been asking for like in Korg or Yamaha PSR is nothing but the Portamento Part SW in the MODX, with Time Zero (or more if Portamento is the desired result) as long as the waveforms for legato is edited without the attack (or let's say properly) and the EGs are proper for the sound. Is that accurate?

I know @Anthony you mentioned that at the top of the this post, is that what you were trying to say? I guess the only "issue" that Jason brought up is that AWM2 play the legato element of a sampled waveform from the start, so some artifacts may lead to undesired sounds.

The AWM2 Engine, Normal Part is playing pitch adjusted "audio recordings" aka "Samples" each time you press a Piano Key.

Imagine a recording (sample) of a "Hand Clap". The recording would be a few 100ms of sound, followed by silence. A Hand Clap is by nature a percussive sound. The recorded sound has in its entirety:-

- An "Attack" portion... i.e. a rapid rise from No Sound to the loudest part of the sound.
- A Decay portion, the clap has happened, and finished, the sound level will now start to fall back down to silence.

In the AWM2 Engine, if you applied typical ADSR Envelopes to a sample of a Hand Clap, they wouldn't change the sound much. The Attack and Decay are already defined in the recording... the Attack and Decay is "Baked In".

So, regardless of Portamento, Legato and ADSR EG settings... each time you played a new key, you would still hear the distinct Attack and Decay of a Hand Clap.

If on the other hand you recorded the sound of "Sustained" French Horn (after the note was first "blown", and before the player stops blowing) you would have a fairly consistent volume and timbre sample (no Attack, no Decay, no Release). The Timbre would in fact be fairly similar to a Saw Wave.

If that "sustained horn" sample was loaded into an AWM2 Element, you could shape an Attack, Decay, Sustain and Release... and maybe make it sound instead, like a Cello.

Such is the principle of Subtractive Synthesis.

Regarding correct use of ADSR.

You say you followed the examples I gave you.

In none of those examples was Attack Time or Attack Level set to 0 (Zero).

In all cases, the Attack values were used to define a sharp increase in Volume from 0 to a "Higher Level" e.g. 110.

For some reason, you wish to set Attack Level and Time to 0, and use the Decay1 and Decay2 to define the Attack slope.

To be blunt, you are doing this wrong.

The Attack parameters should be used to define the upward slope.

I advise you to use the ADSR parameters as they were intended (See Parameter Reference Manual) and learn how they influence the sound. I would do this before taking creative liberties (as per FM-X).

Call up Preset Performance "Early Lead".

This uses FEG only to define its sound (check FEG for each element). Decay2, and Release are effectively zeroed out.

Note: Initial Filter Cut-off = 144

Play a single note and hold it down. Listen how the FEG slowly opens the filter (Attack) and then slowly closes it again (Decay1).

Note: Decay2 time = 0, and Decay2 Level = Sustain Level = 0 = Initial Filter Cut-off = 144.

Play and hold a Note and then play a 2nd Note an Octave above.

If you play the 2nd Note before the Filter closes, it will be subject to the Filter Decay of the First Note. It will not retrigger the FEG.

If you play the 2nd Note after the Filter closes on the first note, then it will play at Sustain level of 1st Note (Cut-off 144). If you hold the 2nd Note and play a 3rd, again the FEG will not retrigger.

Change Portamento Time (pitch shift glide) to 0. You will get the same effect without portamento.

So what is going on?

The 1st note triggers the FEG and it runs through the Attack, Decay and eventually settles on the Sustain (Decay2 Level), but because all subsequent notes are played LEGATO, they DO NOT retrigger the FEG, they are "introduced" onto the Sustain portion of the First Note. I.e. The Attack and Decay portions are removed for each subsequent Legato Note.

This is why you should use Attack and Decay1 correctly.

If you Zero Attack and Decay1, and instead use Decay2 to define the Attack slope.... you will get "unsupported" results. As in, it won't work as intended.

Before you go any further, I recommend studying PEG, FEG and AEG, how they are used in various Factory Presets, and how they behave in Mono (in Poly mode, EGs will retrigger and so does not apply).

Strart looking at Presets in Performance Category Search with:-

Bank/Favourite = Preset
Attribute = AWM2
Main = Syn Lead
Sub = Analog

@Bassam. I was just reading your posts again, trying to understand your reasoning for setting ADSR's as you did.

To clarify, you "eliminated" the Attack parameters by setting both = 0. I assume you did this to "eliminate" Attack for legato. It doesn't work like this, read on.

You used Decay1 and Decay2 parameters to define an "attack" slope. The Decay2 Level value (127?) therefore defined the "peak" of the Attack. In many instruments, the peak of the Attack is usually loudest, but only briefly, and then immediately starts to Decay (I.e. get quieter) to some lower "sustain" level. The Peak of the Attack is sometimes referred to as a "Transient Spike" or just "Transient". In this case Transient means "short-lived", "in-passing", "momentary" etc.

The way you have built your AEG makes the "Transient", permanent, for as long as you hold the key. This is because Decay2 Level normally defines the "Sustain" Level. This is the natural volume of the instrument when it is not being plucked, hammered, blown, bowed, struck or otherwise forced to make a sound.

Lastly, you have used the Release Parameters to define the "Decay" curve. Release does not work/behave the same as the Decay1 & Decay2 parameters. If for no other reason, Release just sounds different. Its normal role is to add some "natural resonance" to a sound. For example a Piano will keep sounding (resonating) even after you take your finger off the key.

All that said, it looks like you zeroed out the Attack parameters to force Legato onto "subsequent" notes.

But, there is only 1 AEG per Element, regardless of whether it is the first Note you play, the 2nd Legato Note or the very last booming Crescendo Note. They all follow the same AEG (and PEG, FEG).

But if the Note is played Legato, the MODX knows to follow the EG's (ADSR curves). The ADSR's are its set of instructions.

In Legato, it's instruction is Skip Attack, Skip Decay1, Skip Decay2 time, go straight to Decay2 Level... I.e. Sustain Level. At least for the sake of this discussion.

In reality, the EG (ADSR) of the preceding note is not "Closed"... it is left Open or Active, such that subsequent notes are played on the existing, active ADSR... which will mostly be on the Sustain level.

You can experiment by setting short, medium and long Decay2 Time values. It is easier than trying to explain it.

Back to AWM2. If the sample being played is a recording that INCLUDES the audio of the Attack (example - a Guitar string being plucked), then no amount of ADSR manipulation will ever truly remove that "Attack Audio".

There are lots of cases of "Recorded Attack Audio" Samples on the MODX. These are usually accompanied with their own "_of" offset samples specifically for Legato (as per first few posts in this thread).

The reason this is done? For the sake of realism. A recorded Saxophone Attack will always sound more authentic than anything you could create with an ADSR.

You may notice in such "real sample" cases the FEG and PEGs are zeroed out, and the AEG is "Boxed" I.e. short Attack, Max levels across the board, and some moderate Release time (not 0). These "boxes" create an "Open Window" through which the real sample can be heard, without infringement.

Finally.... For Legato - Yes, the Portamento switch must be ON, and MONO must be ON, IF you are using ADSR to manipulate and shape the Sample.

Otherwise... real Violins, Cellos, Trumpets, Flutes, Oboes etc... use the XA Control and the Offset samples.

The ADSR method harks back to the very early days of Analogue Synthesisers. The sound source was a droning oscillator (Saw Wave, Square Wave, Triangle wave etc).

The Filter, FEG and AEG were used to shape the sound of the oscillator into something that resembled a Trumpet, Violin, Harp, Bass, Flute etc. This was the school of "Subtractive Synthesis".

Sampling Synthesis came later, and were largely enabled by Digital Technology. As Digital Technology improved so did the quality of the Samples and the Synths that "played" them.

All this stuff, mono vs poly, legato vs portamento, ADSRs, retriggering etc was a complete mystery to me until only a few months ago.

I was asking only recently "How do I stop Legato?". This is how I found out all the above regards ADSRs, Sustain Levels, Re-triggering etc.

It's not in the Manuals. It is assumed you already understand how old Analogue Synths and Subtractive Synthesisers worked. The Manuals are very good for the "What?", but not so much for the "Why?". A moderate level of Synthesiser knowledge is assumed. I had none when I bought the MODX. ????

Might I suggest spending some time with basic "Analog Synth Lead" Presets (Syn Lead, Analog) if you truly want to understand all this.

Hey @Anthony,

Thanks again for the great clarification. I just want to clarify one thing though!

Regarding correct use of ADSR.

You say you followed the examples I gave you.

In none of those examples was Attack Time or Attack Level set to 0 (Zero).

In all cases, the Attack values were used to define a sharp increase in Volume from 0 to a "Higher Level" e.g. 110.

For some reason, you wish to set Attack Level and Time to 0, and use the Decay1 and Decay2 to define the Attack slope.

To be blunt, you are doing this wrong.

The Attack parameters should be used to define the upward slope.

I only set the attack to zero for testing purposes!! Something I learned in design in general (sound (hobby), games (hobby), engineering (work), etc.), always test the extremes and check how they behave...
In this case I was testing if "legato" would skip the plucky attack directly to a smooth decay...
I thought I know what I was doin/testing, until I read this:

If you Zero Attack and Decay1, and instead use Decay2 to define the Attack slope.... you will get "unsupported" results. As in, it won't work as intended.

So thank you!

I do understand how ADSR works for Filters, Amp, and Pitch, and any parameters (in Korg Trinity MOSS, you can basically use EGs for any parameter), but I do agree with you that I need more practice, and more exploration especially in presets to see how typical ADSRs are for each sound. And again, I am a beginner lol, so yes I'm sure there are some interactions between EGs or something specific about each tone generator that I'm not aware of.

Might I suggest spending some time with basic "Analog Synth Lead" Presets (Syn Lead, Analog) if you truly want to understand all this.

Thanks again for the suggestion, and indeed I will!! I learned about substractive synthesis using Syntorial (not sure if you know it), but I found it a great tool for practice and ear training during sound design.

@Bassam.... cool!!

Sorry the text read like a "hard lecture"... it wasn't intended to be 🙂

At least I got the knowledge across and that's great, that's what I was hoping for.

All this ADSR stuff was/is still fresh in my mind, because I was having similar difficulty understanding the behaviour.

The consensus and probably best advice is to just to keep messing with it, until your ears get tuned in to what is going on.

PEGs are probably the easiest to "hear the changes". AEGs and FEGs are a bit more subtle.

Further down the road I highly recommend reading Manny's and Bad Mister's FM-X tutorials. There you will see how ADSRs can be used in "unconventional" ways to create new sounds... anything goes!! It is pretty mind-blowing 🙂

Hi all. Here is an idea that I submitted on ideascale regarding to this. I think it would greatly improve things with mono AWM2 sounds.
Montage more portamento options