Here's how I built the Patch.

Start with AWM2 INIT.

Part 1, Common, Part Settings,
General:-
Set Mono(phonic)
Set Key Assign Single
RevSend 0

Pitch:-
Portamento Master On Part On
Mode Full Time
Time 18
Time Mode Time1
Legato Slope 0

Part 1, Element 1
Osc Name = OBSaw

Part 1, Element 2
Osc Name = OBSaw
Coarse Tuning = -5

Both Element 1 and Element 2
Use these parameters for both Elements

Amp EG
Attack Time = 64
Attack Level = 127
Decay1 Time = 0
Decay1 Level = 127
Decay2 Time = 0
Decay2 Level = 127
Release Time = 85

Filter Type = HPF12
Cutoff/Key = 100%
Cutoff = 55
Resonance = 40

Filter EG
Att Time 6
Att Lvl 127
Dec1 Time 0
Dec1 Lvl 127
Dec2 Time 0
Dec2 Lvl 127
Release Time 61
Release Level 0
FEG Depth = +30

Ins A Effect
Cat Dist
Type Amp Simulator 2
Preset Tube Clean
(Ins Effect LPF Cutoff 2.2kHz)

@Bill....

I think you missed the important detail... it's a MONOphonic Part, i.e. a Synth Lead. There's only 1 note of Polyphony in play for each Element (which you can view as VCO1 and VCO2 in this Analogue Subtractive Monosynth emulation... the Korg MS20).

If you set a Part as Mono(phonic) it's easy to demonstrate the FEG "carrying over" from the previous note, if played Legato.

DEMO EXAMPLE SETTINGS
---------------------------------------------

Set Portamento ON

Set Filter Type to LPF24A.
Set Element 1 = Saw
Set Cutoff/Key = 100% (Full Keyboard Tracking)
Adjust Filter Cutoff Down so that Saw sounds like a Sine around 105.
Set Resonance = 40

FEG (Time/Level)

Attack = 50/127
Decay1 = 0/127
Decay2 = 70/60
Release = 80/0

FEG DEPTH = 35

If you play a note, you will hear the Filter cutoff sweep up, then down and settle on the Sustain Level (Decay2 Level = 60).

Let go of the key.

Play another Key... hear the sweep again

Play a Key and Hold it.

Play another Key and it will sound at the Sustain Level of the previous key... you do not hear a Filter sweep, I.e. the FEG does not retrigger.

You're talking about the same thing from different points of view.

The FEG is a Filter Envelope GENERATOR.

No new envelope is generated during Portamento/Legato events. I consider this a bug, but I think Yamaha considers it "by design".

Jason has MUCH more insight into this.

@Bill...

if you hadn't already figured this out...

Here is how to create a Classic ADSR (Attack, Decay, Sustain, Release) Envelope from Yamaha Style Envelope.

First:-
Set Attack Level = 127
Set Decay 1 Time = 0
Set Decay1 Level = 127
Set Release Level = 0

This makes Attack Level Dependent on EG Depth, aka "Envelope Amount" on some classic Subtractive Synths. It also removes Decay 1 from the equation.

This leaves (ADSR)...

A = Attack Time
D = Decay 2 Time
S = Decay 2 Level
R = Release Time

Note, in my Daft Punk patch, I am only setting Attack & Release (as per MS20 "AR" envelope aka EG1), the Sustain level is assumed/defaulted to Attack Level i.e. 127.

My Montage is tied up doing something else but I should be able to look at the Motif XF since AWM2 is AWM2.

In MODX the best way to force filter sweeps would be to program the shape you want into a motion sequence. Since this is monophonic triggering MSeq on each note on will give you the same function as an EG outside of maybe the timing parameters (MSeq may not allow for the same "resolution" as an EG). However, the basic operation should be what you want if the timing of MSeq is satisfactory (how fast it ramps up or down and where those peaks and valleys are placed).

Yamaha's take on portamento is a bit unique. Frustrating for some (including myself to achieve certain goals) the treatment of the samples' attacks are not like most of the other manufacturers that seem to play only starting at the start loop point rather than the Yamaha way that plays the whole sample. I know I'm getting off topic a bit -- but this means that the best monophonic/portamento samples are ones that do not have an attack in the sample.

I understand why you would want to not want to have the EG start each note (played legato) on the attack phase. I mean, most of the time that's what I want. But, as you've demonstrated - there's also a time to not do that.

See if the timing of MSeq achieves what you're after. If you're not using tempo to time anything else (like an arpeggio) then you can set the tempo to maximum and use tempo-based MSeq with unit multiply in effect (50% -- less than 100% is faster). That's the fastest you can drive MSeq from a tempo standpoint. Or you can try the internal timing with a maximum speed. I'm not sure off hand which is faster or if they both converge to the same timing limit.

Then I'd use a user curve to shape the "envelope" within the space of a single cycle. This is another area where you have control over the timing. Because you could compress down your curve to say only the first 32 input values (out of 127) and then your MSeq "EG" will take a quarter of a single cycle to finish. Of course you could also spread this out over more input values or even use multiple cycles if you want. Loop would be off.

The other option is to use an LFO for this. Filter can be a destination and I think the Part LFO mostly does the right thing if I recall correctly. Part LFO gives you user waveforms so you can draw your shape in there. A bit more crudely than MSeq - but maybe extended LFO can run faster than MSeq or otherwise achieve the goal you're after.

So, there is no "Force Envelope Re-Trigger" option on the MODX/MONTAGE.

That's cool, I just wanted to make sure I hadn't missed anything.

In truth, I think the Legato/Sustaining EG behaviour is "normal" in many classic Analogue Synths (MiniMoog is definitely an example). Some Modern synths provide triggering options per Envelope, but this is a feature, not normal/default behaviour.

My main beef is really with the sample handling in portamento which reduces the color palette of choices somewhat due to the bumpy sample restarts with each note if you use "the wrong" (for what you're after) sample. If there's a sound you like and there's not a "without attack" sample to choose from then you are best resampling the "with attack" sound and chopping off the attack then create your custom waveform. The EG behavior seems fairly conventional to me. At least it aligns with my legato portamento goals.

@Bill....

I think you might not understand how Envelopes (Envelope Generators) work.

They are basically like 1-Shot LFO's and are used to Modulate various Synth Parameters of your choice.

The difference between an LFO and an EG is you can design your own "Wave Shape" (instead of having to use a Sine, Triangle, Square etc).

The primary intended use for EGs was to "Shape" the sound of an Oscillator over time (from Key-On) so the sound more closely resembled a real instrument... for example, a Slow Violin Bow, a Blown Trumpet, a Plucked Guitar String etc. This is a big part of Sound Synthesis, and necessary in a Synthesiser... otherwise it would basically just be an Organ.

Typically, for accurate Synthesis, you need an Envelope for each of Filter Cutoff (Timbre), Amp Volume (Loudness), and Pitch (Many instruments go slightly Sharp in the Attack).

Luckily, the MODX has an EG for each, but the EGs cannot be assigned to anything else.

Traditional Subtractive Synth Envelopes have an "ADSR" design.

A = Attack (Time in Milliseconds)
D = Decay (Time in Milliseconds)
S = Sustain Level of a Held Note (dB)
R = Release (Time in Milliseconds)

There is no concept of "Level" in ADSR except for Sustain. Yamaha provide the option of setting a Level for A,D & R... but that's fairly unique to Yamaha.

Just a brief example:-

Attack is the time it takes for a Struck Note to move from 0 Level (Silence) to its Maximum Volume.

For a Guitar String, Attack = 0ms, it is at its loudest when it is Struck, and the Level Decays thereafter.

For a Violin String, Attack = 500ms. It takes some time to move the Bow across the string to bring the String up to its maximum Resonance and loudness. There may be NO Decay Time (Decay = 0ms) if the Bow continues to move across the String, therefore it moves directly to its Sustain level, which is also a Maximum level in this case.

A, D, S & R are just typical stages a musical instrument's note will transition through.

Release is the Time it takes for the sound to fade to silence after the Player has stopped playing it (i.e. Key-Off, and no better example than an undamped Piano that continues to "Ring Out").

So, if you follow my second post of example settings (Saw Wave Element with an LPF) you should hear that, if played Legato (with Portamento On), the second note does not have an Attack & Decay Portion (Filter Sweep). Instead, the second Note plays directly at the Sustain Level (Set by Decay2 Level of the FEG).

NB: "Decay 2 Level" IS exactly the same function and purpose as a Classic ADSR "Sustain" Level. Yamaha just has a different way of doing and naming things, dating back to the 70's (CS80 for example).

Regarding classic Subtractive Synth Oscillators (VCO's = Voltage Controlled Oscillators), you will find these in the SynLd Category on a Waveform Search.

Typical Single VCO Waveforms are Sine, Triangle, TriSaw, Square and Pulse (various width Rectangles).

In the MODX, Waveforms you see such as "Saw", "Square" etc are Single Cycle Samples, which are looped. There is no Attack or Decay Portion to them. They are intended for use in Mimicing the VCOs of classic Subtractive Synths (P5 = Prophet 5, MM = MiniMoog, OB = Oberheim etc).

I think it's a shame that so little time is spent by Yamaha demonstrating this "Synth Side" of the MODX Synthesiser. The Pianos, E Pianos and Organs by contrast have been done to death.

It's the "Synth Side" of the MODX where you will get the most mileage from Envelope Generators (this Includes FM-X but the purpose of EGs is more complex there, although functionality is pretty much the same).

@Bill

It may be easier to "hear" if you use pitch. Change the PEG to rise slowly then fall at the same rate for the attack/decay1 (rising to maximum for a second or two and falling back down for a second or two). This is for a monophonic Part with portamento turned on (two things required to have the EG skip past the attack phase on legato notes).

Play a single note, hold, and lift up. You should hear the pitch rise and fall due to the PEG. Press the same note and hold. I'd have you press the same key so you know there's no portamento happening here since you're going to the same key (and there are other considerations like full-time vs. fingered where this would matter more for full-time vs fingered). Don't worry if you don't know all of the portamento lingo - it's the sound we're focused on right now.

Now, hold down one note, wait for the pitch to rise and then fall back down - go ahead and hold the key for a few more seconds then, while holding the first key, press a second key. Press a close one or a far one (depending on your portamento settings it may take a while or not to slide up or down to the second note). What you'll hear is just the rise of the pitch to reach a higher note or the fall of pitch to reach the lower note. You won't hear your EG's rise-and-fall you programmed into the attack. That's because this phase is skipped.

It's not hard to hear the filter either. I recalled "Saw Lead" (on a Motif XF) and removed elements 2-4 (the elements here are panned hard left and right with two keyboard ranges -- I just kept one channel and one keyboard range). Then I went to FEG and changed the attack to rise to maximum 127 and increased the time so it would start with a choked sound and take a while to bloom to a bright sound. It's obvious hearing this playing a single note without invoking portamento or legato. Then playing legato notes it will be clear the filter does go through these paces. You will hear pitch sliding around - and that may be obscuring what you should be listening for. However, it's not difficult to hear the filter apart from the pitch which is also sliding around. Particularly when you setup the filter to slowly ramp up over 1-2 seconds.

You'll earn an appreciation for how EGs work during legato playing of a monophonic Part that has portamento turned on.

@Bill...

Firstly, your example INIT case study (C1 vs C5) is not "Normal", please see later notes.

Meanwhile, I don't profess to know the Design Mechanics of Envelopes, however their behaviour and end results are clear to see.

Whether a 2nd Note, with Portamento On, gets a "New" Envelope, or whether it inherits the "Old" Envelope of the 1st Note, seems irrelevant.

In either case the 2nd Note experiences an Envelope that starts at whichever progress point the Previous Note's Envelope was at, when the 2nd Note was pressed.

If the 2nd Note was Keyed while the 1st Note was on its Attack Slope, then the 2nd Note will experience a short Attack, then proceed to Decay, Sustain and Release as Normal i.e. the 2nd Note will start part the way up the Attack slope of the Envelope.

Normally, Attack and Decay phases are set to short durations, so normally the 2nd Note will pick up at the Sustain stage (no Attack, No Decay).

But if you set a Long Attack Time (75) and Long Decay Time (75), you will have plenty of time to play the 2nd Note on either the Attack or Decay phase of the 1st Note. And thus, witness the phenomena first hand.

A "Forced Re-Trigger" of an Envelope means an EG can be configured to Restart at the Beginning of the Envelope every time a Key is pressed, regardless of Portamento On. This would normally be an "Optional" Yes/No parameter on a Synth, but we have established this is not an option on the MODX.

The "Forced ReTrigger" concept is not difficult to visualise. If you take a look at the Part LFO Page, you will see a Parameter box "Key On Reset". This determines if, and when the LFO is restarted "from the beginning", usually 0° Phase.

Off = The LFO is free running, and always running in the background. Key-On events will not Reset/Restart the LFO.

1st On = The LFO will reset/restart at 0° when a key is pressed, but will not reset if a second key is pressed while the 1st is still held down.

Each On* = Each Key-On Resets the LFO, regardless of previous Key(s) being held.

* This is the equivalent of a Forced Envelope Re-Trigger.

NOTE: Regarding your C1 vs C5 example.

The behaviour you witnessed is due to the INIT "Sample" i.e. "CF3 Stretch Sw St".

@Jason hinted at the existence of these "Special Samples" earlier. I have seen comments in other threads regarding their strange behaviour with Envelopes, but have never dug deeper.

My high level understanding is that some Samples are effectively "Strings of Samples" designed to be played together, to provide a more realistic behaviour of the real instruments they are representing.

In these cases, there may be a Sample for "Attack", a Sample for "Decay", a Sample for "Sustain", and a Sample for "Release".

Whatever the mechanics truly are, it has been discussed in these pages, since they are known to behave erratically in the presence of all but the simplest Envelopes.

I think the premise seems to be that EGs are not required, since the Multi Sample caters for the usual ADSR phases.

Going back to your C1 vs C5 example, simply swap out the Sample for "OBSaw". These "Single Cycle Loop" samples are intended for use with Envelopes (to approximate traditional Synth functionality).

Also, keep your eye on the following Parameters when using Monophonic Portamento.

Part Settings, General -> Key Assign
Part Settings, Pitch -> Mode
Part Settings, Pitch -> Time & Time Mode
Part Settings, Pitch -> Legato Slope.

Call it what you want - I see little difference in the distinctions being drawn. I was addressing your feedback that you couldn't hear the lack of filter attack/decay in the original example given so I helped using pitch - my go-to for easily recognizing what the envelope is doing (even if ultimately I use AEG or FEG). This was my last goal more than to give a tutorial about how the EG functions. That's why I purposefully glossed over the gory details.

The answer to the original inquiry has already been acknowledged. I think most come out of this with a more firm grasp of their instrument's EGs and how they interact with mono portamento.

@Bill

I advise going back and working through the examples provided, but make sure you use a Single Cycle Loop sample, for example "OBSaw". There are many others, but it's safer to stay with this for now.

By the way, since my Montage is still not available I went ahead and built a non-looped LFO sweep that matches the "Da Funk" hook's general timing. It wasn't hard to do. The (Part level in MODX) LFO was set to each-on and I used a user curve with up/down which gives lots of flexibility to reproduce what an EG would do. This works for monophonic because we don't have to worry about the LFO restarting and impacting previously played notes (because there aren't any previous notes still sustaining in mono). I adjusted portamento fairly fast.

BTW: the LFO had lots of room to speed it up. In other words, I had to slow it down (speed) to get the timing right. I used the first maybe 5 or 6 steps of the user waveform then flat lined it for steps 7-last. If you reduce the amount of cycles then it will stretch out the time for the remaining steps.

What I'm getting at is that this suggestion works and works well. It does a what the filter is doing in the recording. The sound I used was wrong (single saw element left over from my experiments before using the Motif XF). MODX has all of these same features. Motif XF doesn't have Motion Sequence so LFO is about the only way. That said - I think Part LFO is exactly the right tool for this job.

Even legato notes (hold down first note, press a new note while holding - 2nd note ramps with portamento and plays leaving behind the 1st note) re-triggers the LFO so the filter does the right thing.

@Jason... thanks.

I never thought of that, I'll give it a go.

It's been hard unravelling what's going on in that Daft Punk riff. It sounds very much like there is portamento, but flicking through an old MS20 Manual e-scan, I couldn't fathom anything regards triggering EGs.

There's a few YouTube covers/tutorials, but the approaches seem to differ dramatically.