I prefer documentation that gives me absolutes. Having asked in other departments without getting answers I move to the stimulus/result model where the behavior is just documented as memories that last until I forget or choose to write something down.

It's worth mentioning that the 2nd LFO on an FM engine is not new. Whatever deficiencies in documentation have existed for quite some time predating Montage or MODX.

From the FS1R manual (circa 1998):

LFO2 (Low Frequency Oscillator 2)
...
05: FilterModDpt

❑ Filter Cutoff Modulation Depth
❑ Settings: 0 … 99

Sets the maximum amount of filter cuttoff [sic] modulation that can be applied to the current voice. A “0” setting produces
no modulation while a setting of “99” produces maximum modulation. Filter cutoff modulation produces wah-wah
type effects.

It seems for backwards compatibility or just to "plop in" the past - it's likely that the ambiguous "Filter" modulation is really, as previously suggested, cutoff. And then it's possible that, for compatibility, the behavior of the LFO has been made to mimic the past LFO2 implementation which has a different kind of reach than FM-X parameters alone give you.

At least I believe the FS1R documentation/discussions gives another path to run down looking for literature/discussions that might provide more insight.

Incredible.

Says so much. Wah-wah, indeed.

Thank you, very much!

Since the technology is highly leveraged I typically search through old gear to find "better" documentation. Which is often just a bit differently stated where the combination of sources helps. Lots of these parameters, features, concepts, implementations go way back and sometimes looking at gear closer to the introduction of a feature provides less editing. Overall documentation has become more accessible and, sometimes, better. However, there are cases where cryptic notes from that seem closer to the engineer's description serves better. Not that the FS1R docs are that - but there are times when you see this in looking at older docs.

So, many times, I'll go back to old samplers for AWM or other various previous products to pull out how previous products documented like-features.

The seemingly arbitrary limitations of this LFO now make sense. They're arbitrary limitations from a much earlier period 😉

Which should have made it easier for them to see what's wanted/needed/expected now.

Which might have been what led to the ideas that created the Motion Sequencer as a support act for these archaic LFOs.

Cannot understand why there's no Filter Control (nor any other Part Controls) in LFO 1 of FM-X, though. That's cutting off your nose to spite your face.

Extended LFO is new in the timing arena for LFO2. So that was, at least, a late-bloom addition. However, out the gate the design was highly leveraged. I'm not sure why full FS1R compatibility wasn't done considering how much is borrowed from the FS1R -- but, at any rate, having a conversion from older FX instruments automated was an early design goal and therefore certain commonality had to be in place to accommodate this goal.

There's plenty to critique about what could have been -- and I've done my fair share. It's all fair - but hopefully there are no glaring holes in understanding what is presently offered and how to use it.

A Wah-Wah is just a Resonant Band Pass Filter where the Centre Frequency (of the Band Pass Filter) is moved by the Wah-Wah's Rocker Pedal.

To say LFO of a Filter sounds like a Wah Wah is not really true.

It depends on the type of Filter and what's being modulated.

I've got 10 minutes spare, and thought I'd share some knowledge stemming from my Guitar playing experience. I am not trying to insult anyone's intelligence. It is apparent in the guitar world that there is a lot of confusion and plenty of misconceptions about "Modulation".

Modulating literally means "Changing over time".

Calling a "Chorus Effect" pedal a "Modulation Effect" can be a bit misleading. "Modulation" is a shorthand nickname for any Effect that modulates either all, or some part, of the source sound over time.

Modulation Effects can include:-

Reverb
Delay
Chorus
Flanger
Phaser
Tremolo
Wah
Auto Wah
Auto Filter (e.g. Mutron)
Synth LFO (part of the "Modulation Matrix)

In the Guitar world these types of effect pedals are sometimes collectively referred to as "Time Based Effects". This distinguishes them from "EQ Based Effects" (since they primarily change the Timbre) such as Compression, Overdrive, Distortion, Fuzz, and of course EQ itself.

Some Effects fall into both categories (Time Based and EQ Based) such as Auto Wah, Auto Filter etc.

In the Guitar world, Time vs EQ Based determines where in the Signal Chain the Effect should be placed (traditionally). EQ first, Time last.

Two things that Reverb, Delay, Chorus, Flanger and Phaser all have in common:-

1) They split the input signal into at least two paths... Dry (unmodulated) and Wet (modulated).

2) They modulate (change over time) a DELAY TIME on the Wet signal. You read correctly, they do not modulate Pitch, which most people believe is the case with Chorus and Flanger. Pitch shift is only a side-effect of changing the Delay Time (Doppler Effect).

In these cases there is a Wet and Dry component, which introduces some notion of a Wet/Dry MIX. Traditionally, you will find a MIX control on Reverb and Delay. This allows the relative volume of the delayed wet signal to be lowered or raised for the desired "natural" effect (Reverberation, Back Wall Echo).

In Chorus and Flanger, traditionally there is no Wet/Dry Mix control. However it is still a factor, and is usually fixed electronically (static) at 50/50 I.e. Wet = Dry volume. This is because the desired Chorus or Flanging effect requires a 50/50 mix (for technical reasons I won't explain here).

However some modern Chorus and Flanger Pedals do have a MIX control. There are various implementations, but usually the input signal is split into 3, providing a Dry "Channel" and a "Wet/Dry" Channel. The User's Mix control governs the relative volume between the two Channels. The "Wet/Dry mix" in the Wet/Dry Channel is still 50/50. Again there are technical reasons that necessitate this (it's mostly to do with unwanted Comb Filtering if you want to do some research).

The last category here is, in effect, "Mono Modulation". The Modulation is performed entirely on the Dry signal. There is no separate "Wet" signal path and therefore no Wet/Dry mix.

Tremolo is an example (modulated amplitude), but Wah, Auto Wah and Auto Filter are also "Mono Modulated" (Modulated EQ in this case). Because the only "Dry" signal is modulated, it becomes a "Wet" signal. There is no Dry (unmodulated) signal at the output of the effect, only a Wet signal.

If you can imagine 2 Guitarists in a Studio, each with his own "track" on the mixing desk. One guitarist is playing "Dry" (Track 1), the other Guitarist is using Tremolo on his Amp (Track 2). The Studio Producer can implement a Dry vs Wet (Tremolo) mix with the Faders on the Mixing Desk. This actually happens quite a lot, and is used to great effect (Listen to The Smiths - How Soon Is Now).

However, this does not make Tremolo a "Wet/Dry" effect. Traditionally, it isn't.

The same applies for movement of the Filter Cutoff in a Synth. In fact Mono Filter Modulation is a foundation stone of Traditional Subtractive Synthesis. Start with a Broad Harmonic Spectrum (VCO), use a Filter to "Shape it" (VCF).

The LFO (Low Frequency Oscillator) is a "basic method" for "automating" modulation. This is also true in all* the "Guitar Effects" mentioned above, likewise in Synths.

*Modulation is just "something" Changing over time. It does not have to be an LFO making those changes. In a Wah pedal, the changes are made by your moving Foot. In an Auto Wah, that movement is handed over to an LFO or an Envelope Follower.

Somewhat unique to Synths is that the LFO can be Assigned to whichever parameter you choose. By contrast in the Guitar Effect Pedal world, LFO's are embedded, your Chorus has its own LFO, your Flanger has its own LFO. Although, in Guitar world, many pedals now come equipped with a CV (expression) Jack allowing for external Control, including external LFO Hardware.

Why would you want to know all this?

The power of the MODX (and other modern Digital Synths) has so many options that it allows you to create your own FX and Modulation. This is not hard to do, if you understand the basic functionality of "Traditional" FX.

Remember, before we had "Effects Pedals", most Effects we take for granted had to be "built" manually inside Recording Studios. For example, the name "Flanger" is derived from the practice of slowing down a Tape Reel by pressing your finger on the "Flange" (outer rim) of the Reel.

For a bit of fun, here is a very basic example of how to create your "own Chorus" effect.

Create a 1 Part Performance. 2 Identical elements in that Part with equal volume. Assign an LFO to modulate the Pitch of just one of these Elements.

But, if you want to get a bit more detailed. Forget the Pitch Shift. Assign a Control Delay (Insert Effect) to just one Element. Set the Delay Time to 20ms, 100% Wet and then Assign Delay Time (InsA Dly Time) as the LFO destination. (Tip: experiment with Delay Transition/Slew Rate, and Feedback, keep LFO Depth Low = 1).

You are now designing your own "Unique" Chorus. Whether you'd want to, or appreciate the result is up to you.

Back to the OP. I do not hear any "secondary" parameters being changed by the LFO. Also, I believe that Depth only affects the amplitude of Parameter (Cutoff Value) variation. I cannot see or hear any "Dry/Wet" component that is moving, unless you count other unmodulated sound sources as a Dry signal (For example, another Part, or an FX introducing another source/Split path).

The crux here is that "Depth" control (example on a Chorus) is often confused as being a Wet/Dry mix. It makes me wince when I see YouTube Reviewers/Demoers describing a Flanger Depth Control as a "Mix/Volume" control.

The Depth is governing the Amplitude of the LFO, that is all it does. High Depth settings usually result in a much more drastic effect, so it is understandable that people would confuse this as some form of mix/level control.

The definition of a Dry Signal is one that is not subjected to any form of Modulation.

The definition of a Wet Signal is a Dry signal that is subject to Modulation. Quantifying how "Wet" that signal sounds is dependent on how severe the Modulation is, that is, the DEPTH of the Modulation.

Wonderful post. Thank you.

Just a little aside, on this particularly vexing issue...

[quotePost id=116742]
The crux here is that "Depth" control (example on a Chorus) is often confused as being a Wet/Dry mix. It makes me wince when I see YouTube Reviewers/Demoers describing the Depth Control as a "Mix/Volume" control.

The Depth is governing the Amplitude of the LFO, that is all it does. High Depth settings usually result in a much more drastic effect, so it is understandable that people would confuse this as some form of mix/level cintrol.

[/quotePost]

Jason isn't entirely wrong when discussing a wet/dry sensation from this particular feature. There is something else going on, other than merely modulation of the filter's cutoff, it sometimes seems. It might be as simple as a slight wet/dry mix also happening when changing the depth, could be just a hidden slight adjustment that's from the origins of DX FM such that the later adding of this feature required some fudging to make it more musical, but it also might be that the order of the mathematical operations that determine resultant cutoff (from all the different possible ways it can be modded in the MODX) are also somewhat oddly structured such that there is a compounding and/or normalising bit of maths happening, too.

My experiments favour the latter part, but that's also out of naivety, I don't know anything about the origins of this feature, and do quite a lot of dynamics work on filters and their cutoffs in particular, not just from this source of modulation, so am more likely to sense the operator precedence/ordering results.

Also, this is quite particular to the the FM-X 2nd LFO's Filter Modulation feature. It is not at all pertinent to AWM2, wherein the feature is named Cutoff and does just that modulation.

It's telling that this is named Filter Modulation: the programmers might be telling us something quite distinctly by not being specific about what it does - kind of like saying "if this was merely Cutoff Modulation, we would have called it that."

I've just been messing with my MODX...

Regards FM-X and "2nd LFO", and some "maybe's".

For the uninitiated let's get some facts on the table:-

1) Unlike AWM2, Filters can only be applied to the whole Part (Common Level), not to individual Operators.

2) Part LFO (LFO #1) can only be applied to Insert FX, so if you are not using Insert FX it is redundant. In any case it does not affect the Synth Filter, unless you are using an Ins Effect as the Filter (Misc -> VCM Mini Filter).

3) FM-X "2nd LFO" can only be applied to any or all off:-

a) Pitch Modulation - Individual Operator, Part or Both.
b) Amp Modulation - Individual Operator, Part or Both.
c) Filter Modulation - Part Only.

4) In the case of the LFO acting on the Filter, whatever aspect of the Filter the "Cutoff Parameter" is controlling, will be modulated by the LFO#2. This is an important distinction to visualise, because depending on the Filter, the Cutoff may or may not be cutting off the Frequencies you expect. For example, in a Band Pass Filter (BPF) the "Cutoff" is actually controlling the Centre Frequency (Fc) of the Band. The "Q" or Band-width would actually dictate the Higher and Lower cutoff frequencies (Fq).... but you are not provided any control of this.

5) Movement of the Filter Cutoff is ONLY audible if the Cutoff Frequency is affecting (changing) the original Harmonic Spectrum of the Input Sound Source. A very clear example of this, is if the Input Source is a Sine Wave. A Sine Wave has no Harmonic Spectrum. It has 1 Fundamental Frequency (the piano key "Note"), and that is it. If you adjust the Cutoff, in a different Frequency Range than the Fundamental, you will hear no change. Using an LPF example, if the Sine Wave Fundamental is above the Cutoff Frequency, you won't even hear the Sine Wave sound source... it is cut off.

In FM-X, the resulting Harmonic Spectrum is entirely dependent on the interaction of the Operators "Frequency Modulating" each other. This is a seemingly obvious statement. But this is fundamentally different to a Subtractive Synthesis "Synth". On an FM Synth, the "moving filter" sounds are NOT created by a Filter, rather they are created by moving (modulating) Operator Levels and Frequencies. So in FM Synthesis there is no practical need for a Filter. However, the addition of a Filter is a nice Bonus to have, providing additional Sound Sculpting possibilities. Essentially, on an FM Synth, a Filter is just an "Effect" like Chorus, Flanger etc. It is not integral to FM Synthesis.

By contrast, a Filter is integral to Subtractive Synthesis. It is at the very core of the design.... it is the "Subtractor".

In FM-X, depending on the Algorithm, and how that algorithm is actually programmed, You can have between 1 and 8 Sound Sources. In this case the Source is whatever is defined as a Carrier, with a Level > 0. The output of the Carriers is Summed (traditionally Mixed, they have no direct modulation effect on each other). No different than you would mix a Bass, with a Guitar, with Drums.

Now in FM-X, the Filter ONLY acts on the Sum of the Carriers. Let us use an LPF to illustrate. If any one of those Carriers has a Harmonic Spectrum that is positioned entirely below the Cutoff Frequency, then any small adjustment to the Cutoff Frequency WILL NOT change the Sound of that Carrier.

By contrast, if one of those Carriers has a broad and high-frequency content (a Broadband Spectrum), then any small changes to the Cutoff Frequency WILL affect its sound.

Take an easy Example.

Carrier 1 = Sine Wave. It is Pitch Shifted down to Provide a Sub Bass.
Carrier 8 = Heavily Modulated Sine Wave, Synthesising a "Trumpet"

If we apply a Low Pass Filter to this FM Part, and then LFO Modulate the Filter Cutoff with a Depth of 20%, we will hear the "Trumpet" Changing, while the Sub Bass remains unaffected. This creates the "illusion" of a Wet/Dry mix, where the LFO Depth is controlling the Wet/Dry mix. But it isn't. It is just affecting a "zone" within the total (summed) Harmonic Spectrum.

This is an extreme example to exaggerate the point. The reality is all about Aural Perception and also Perspective. You will notice, and be drawn to sounds that are being "severely" affected by the LFO, and may conclude that other sounds in the mix are unaffected by the LFO. In truth, they may be affected less severely, more subtly, but that does not constitute a "Dry Signal" in the mix. A "Dry" signal will remain unaffected by any increases in Modulation, regardless of the Modulation Source (LFO etc). In this case, if you were to increase the DEPTH of the LFO, then ALL sound sources (Carriers) would be affected. Proof of point, LPF Cutoff 50%, set LFO Filter Depth to Max (99) and you will get a Tremolo Effect... All Sounds Off, All Sounds On... rinse & repeat. You won't get a lingering "Dry" signal still sounding during the "Off" portions. Unless, there is a second "unmodulated" Part (Part 2), or, Delay/Reverb is providing an Echo of the source over the gaps in the modulated audio.

Antony, is this an accurate parsing of your screed?

"Despite this feature being named after a whole system (Filter), I believe it is actually only modulating a single feature of that system, namely the Cutoff of the Filter."

[quotePost id=116748]Antony, can I parse your screed as:

"Despite this feature being named after a whole system (Filter), I believe it is actually only modulating a single feature of that system, namely the Cutoff of the Filter."

?[/quotePost]

If I understand you correctly, then I believe the 2nd LFO is only Changing the Value of the Cutoff Parameter.

It's Semantics, but depending on the Filter, the Cutoff Parameter isn't necessarily changing the Cutoff Frequency. The example I gave is the "Cutoff" governing the Centre Frequency on a MODX Band Pass Filter. But regardless of semantics, the LFO is still controlling "the number" that appears under the Title "Cutoff".

Is the LFO controlling anything else?

No, I don't believe so. However that is not saying that auditory side-effects will not be introduced. Whether or not "side effects" are introduced depends on the complexity of the Input Sound, the complexity of the Filter, and also the complexity of the "System" as a whole.

To a large extent, such side-effects are unpredictable.

A very basic example of a "side effect" is Frequency Beats. Two differing frequency sources (let's say 2 Harmonics) that differ in frequency by only a fraction (Hertz) will create a "volume beating" at low frequency (of beats per second), which can easily be confused as an LFO controlling Signal Level (a tremolo type of effect).

When you change the characteristics of an Audio Filter, you are also changing the dominance (energy levels) of encumbent Harmonics, such that two "beating" frequencies that had previously been "non-dominant" (before you moved the cutoff) are now "dominant" causing the "beating" to be clearly heard.

This is especially true in FM, because "un-natural" harmonic spectrums can manifest in FM, before you even get to a Filter, or Effect downstream.

I experienced this first hand when using FM, wishing to remove an unwanted LFO, but I found the LFO was not active. The audible effect I heard was due to detuned operators AND a Ratio that was not exactly 1.0 (it was 1.003 or something).

I was following Manny's FM Tutorials when he highlighted and provided "Spectographs" of "Over Modulation" which I recalled from University lectures being a problem in Telecommunication Transmission Systems. It's a bizarre link to be mentioning here, but it is exactly the kind of thing that can introduce "new" sounds as a Spectrum is altered, by a Filter for example.

So I do not doubt that you and Jason can hear "other stuff going on". I believe this to be due to the eccentricities of FM rather than additional Parameter changes made by the LFO.

.

[quotePost id=116749]
So I do not doubt that you and Jason can hear "other stuff going on". I believe this to be due to the eccentricities of FM rather than additional Parameter changes made by the LFO.

.[/quotePost]

Unlike Jason, I'm not hearing this. I sort of can... but I'm primarily seeing this.

Sometimes, when modding a lot of stuff to do with the FM filter, I get to situations where other modulations (not coming from the 2nd LFO) are seemingly blocked by it and/or frozen from direct input. Reducing the influence of the 2nd LFO (depth), a few times, has released this blockage, and/or re-enabled the Motion Sequences that I've set up to mod something of the Filter to get back to doing their thing, and re-enabled Assign Knobs I might have wired up to this or that...

...some other times, it crashes - this is a freeze of the entire system, but often with some stuck audio getting streamed out. Only a restart unlocks this. I've left this going one time for over an hour to see if it would resolve.

It's this behaviour, combined with me listening a bit closer in isolation, that caused me to begin to think that more is going on than a mere cutoff parameter modulation.

I primarily wonder why modulating the cutoff value directly (which applies to the sum of all carriers) does not sound the same as an LFO2 modulating filter. Even automating the "directly" modulated cutoff parameter value using motion sequence to sweep in a triangle does not sound the same while the modulator (MS Lane for "direct" or LFO2) are nearly identical in range and timing. Other comments about LFOs say that the modulator's output value (that is, the oscillator's influence on the destination it is affecting) identify that these destinations have ceilings and floors at 127 and 0 respectively. So, therefore the modulator isn't able to "push" the cutoff parameter (if that's all LFO2 is doing) more than motion sequence.

I'm not really experimenting with the general case here - I always setup a filter without a notch. Something middle-of-the-road. Just as a vehicle. And that same filter doesn't seem to share all of the characteristics comparing LFO2 and "direct" modulation.

There's a range where there are similarities. A fairly narrow band of cutoff parameter changes. But even if I get the two fairly close by adjusting the ranges of each arbitrarily to say - maybe there's a "k" scale factor - there's still an element of dry/wet apparently here. Only because I don't hear the same original non-cutoff-changed sound in the case where the cutoff value is directly modulated by something else such as MSeq, mod wheel, etc.

I'm fairly OK with not knowing the secrets. This is an area where I take the sound and live with that. It's not a place where I personally paint by numbers. I know the impact of LFO2 alone is different from the impact of modulating the cutoff parameter itself. And I have a sense of how these two differ.

Now there are other parts of the system where "the math", so to speak, would help me program my keyboard better. I would work faster by using a ruler to draw straight lines rather than depending on my freehand. To a large extent I've moved on since asking similar questions for other areas, received a lot of pushback for even asking (which, FYI - I respect the question and don't judge the "why" at all here). Bumping into enough walls, it's healthy for me to go around them rather than persisting.

That sounds more pessimistic than it is. I get along pretty well with ambiguity even when I wish it wasn't there.

@Jason...

If I've learned to expect anything from the MODX, it is not to expect anything logical.

I treat each function in isolation, and test to see what it actually does, and how it actually behaves. I can see the logic in creating a "3rd" LFO with Motion Sequence, but I am not shocked that you can't get behaviours to match (with the 2nd LFO).

Based on what I know about LFO's in Guitar Effects, the LFO (IC chip) usually overrides any "Manual" control, by virtue of its connection and position later in the Signal Chain.

You don't see "Manual" controls very often... but they are starting to make an appearance, usually with a CV input for Expression Pedal Control.

Anyway, that's a different story. The LFO (traditionally in the Analogue world) biases the Voltage controlling another IC Chip. The LFO varies this Voltage (Alternating Voltage - Sine Wave, Triangle Wave or other) and by doing so causes the Audio Signal to "vary in sympathy".

If you happen to have a Flanger lying around you can see the overriding effect an LFO has over a Manual "Static" or "Direct" Voltage.

The Flanger's Manual Control sets a Static Delay Time between the Dry and the Wet (aka Lag). With the LFO Depth Off or Very Low, the Manual will vary the Lag quite drastically. If you turn LFO Depth up, adjusting the Manual control has increasingly less audible effect. In fact with the LFO Depth at Max, the Manual is effectively bypassed.

It doesn't matter that you understand the electronics, other than this is how LFO's were/are traditionally implemented. I would guess that the MODX LFO "models" this same behaviour

By contrast, Motion Sequencing would be more akin to placing a CV Expression Pedal on the Manual Control.

The observed behaviour is similar, but not identical due to different implementation.

That's my best guess.

I haven't tested the MS Lane. So far I prefer to keep things as simple as I can. If I can get what I need from just EG and LFO I will. Probably because of a lot of the music I listen to, the Synths in those days only had ADSR and LFOs.

(EDIT: actual waveforms output by LFOs were rarely "pure" e.g. Sine, Triangle etc). At best they were a corrupt "approximation". But, in some cases that "corruption" became a unique part of some "legendary" Effects, which just cannot be reproduced. The technology is defunct, long since out of productions, and blueprints in landfill. If you want to spend $5k on a "still working" vintage, be my guest 🙂 ).

As I've alluded to, I'm not really searching for what it is - but am presenting data to show what it isn't. It doesn't seem to be, in its entirety, what others have presented as the end-all. I think it's pretty easy to isolate, as I already do, the cutoff parameter to modulate it outside of the LFO. At first I used just the mod wheel and manually controlled it.

With this one, I'm not in search of what it actually does since I already understand acoustically what it does and how it differs from the MS modulation of cutoff. It's sort of a "warm" (LFO2) and "cold" (direct mod) cutoff sweep. That's how I think about it and have those at the ready if I wish to add that "effect" of respective temperatures.