Welcome to the fifth and final article in the Manny Fernandez FM-Xplorations series! A longtime FM programmer who contributed sounds to the DX7II, SY77, SY99, FS1R, DX200 and reface DX, Manny has lots of tips and tricks to share in this highly informative series. So let's join him now . . . Manny’s FM-Xplorations Article 5 More Deep Diving and Useful Tips, Tricks and Features
In our last article we discussed setting up multiple Motion Sequences for complex and unique modulations of the FM-X engine, recalling them via Scenes, as well as interplay of those with the Arpeggiator. This article will take those concepts more in depth, as I breakdown some Performances to highlight various, tips, tricks and additional features of FM-X.
MFMX VeloSyncThwip: TIP: use the the Velocity Limit capability of Motion Sequences. First gradually play notes at higher velocities and listen for when hard sync timbre comes in and ends with the “thwip” – a kind of noise like ‘sucking’ sound. This is created by using the Motion Sequencer to control the Carrier Ratio in both integer Ratio steps using Pulse A with the stepped Hold shape for the ‘hard sync’ part of the timbre, and then the continuous ‘Standard’ sloped shape for Pulse B that creates the ‘sucking noise’ effect:
Then the velocity range for the Motion Sequence Lanes are set to only play for velocities above 106:
To see this in detail, please watch “Tutorial Video 4, Part 1” below:
MFMX Krazy 8 HarmSwp: Trick – Fixed Frequency mode Carrier with non-sine Spectral Shapes. Go into Part Edit mode and go to the Form/Freq page for Operator 8. Mute all Operators 1 through 6 and turn off the Motion Sequence:
Again, make sure the Motion Sequence is turned off using the Motion Seq on/off button. When you play a note you will hear an accelerating/decelerating LFO style tremolo effect. This is created using a low Fixed Frequency value (3.4965 Hz) for the Carrier Operator 8, combined with the All 2 Spectral Shape. Adjust the Skirt and hear how the brightness and intensity of the effect changes. Also change to the other waveform and check out how they sound different and how their Skirt and/or Resonance changes the effect. This is set up in Scene 4, along with Motion Sequence control of Operator 8 frequency to continue to speed up and slow down the tremolo effect. Tip: Use the Motion Sequence Shape control! It’s a little too awkward for written explanation, so make sure you watch “Tutorial Video 4 part 2” below:
MFMX Aether PadCombo: Trick – Combine multiple destinations to the same controller, using different Curve Types to have them active in different positions of the controller. Here Mod Wheel controls both the FM harmonic pitch modulation and the filter cutoff modulation. The pitch modulation has the Bell Curve to have it active in the middle of the Mod Wheel range. The Curve for the filter cutoff destination has the Threshold shape so it only becomes active in the upper half of the Mod Wheel range and the FM pitch modulation effect fades back out. So you can choose which of the two modulations you hear based on the position of the Mod Wheel. This is covered in detail with “Tutorial Video 4, Part 3” below:
I want to touch on the ‘monophonic’ nature on the Motion Sequencer. It is similar to the paraphonic behaviors of analog or hybrid synths that share single components of the audio signal path, such as a single VCF (Filter) or VCA (Amplifier) that simultaneous process multiple note (polyphonic) playing. Two classic examples of paraphonic architectures are the ARP Odyssey and Korg Mono/Poly.
The difference in MONTAGE is that instead of a signal processing component or module the Motion Sequencer is a control source, like an LFO or Envelope. So as such in FM-X it allows for control of many Operator parameters like Frequency, Level, Spectral wave, Skirt, etc., -- the core parameters that control the timbral behavior of our sound. Thus in a ‘round about’ way the Motion Sequencer can be used as a complex paraphonic waveshaper. And that’s the paradigm that I often utilize it, as it really enhances the sonic capabilities within the FM-X engine especially that there are parameters in the FM-X synthesis engine available as destinations that we haven’t been able to modulate before in such a highly controlled or customizable manner in FM synthesizers before.
OK, so now we’ll deconstruct another Performance and give an example how Motion Sequences expand the sonic palate of FM-X in new ways compared to legacy FM synthesis. Go to the Live Set and select the Performance “MFMX FMSyncLeadMSEnv”:
One thing missing in the timbral palate of legacy FM synthesis is a good emulation of an analog-style envelope modulated oscillator hard sync. In the past we’ve faked it with overmodulation, raised Carrier frequency Ratios and judicious use of non-sine waveforms. But while the brightness and character were somewhat similar to hard sync, the harmonic structure and movement were completely wrong. The frustrating thing for long time FM programmers is we could see a huge step towards a solution since the beginning – modulate the frequency of the Carrier in harmonic related intervals. We could only continuously modulate the Carrier frequency with an LFO or a Pitch EG. Sweeping a Carrier operator’s Frequency Ratio Coarse parameter in real time is easily done in edit mode of even the original DX7, but to be able to get that same behavior as a modulation option has never been available until now with FM-X.
Play some notes and you’ll hear the default Scene 1, a monophonic lead with the fairly typical static FM sync sound, best heard below C4 (due to the need in keeping aliasing under control.) Press Assignable Switch 2 to turn off the effects. Select Scene 2 and play some notes, and you’ll hear a pretty good traditional fast attack hard sync sweep in the sound, especially in the G1 to C4 range. Select the remaining Scenes 3 through 8 and listen to the different “envelopes” controlling the sync effect. So let’s see what’s going on and look at the Assignable Knobs. The important parameter is Knob 2 assigned to Carrier Ratio. You’ve seen me use this assignment in a lot of the prior examples, because operator Ratio is a key parameter controlling FM timbral structures. However, before I’ve either used the Knob to manually select a particular frequency ratio, or used the Motion Sequencer to slowly step through the Ratios while the Arpeggiator plays for timbral emphasis and changes of individual notes within the arpeggio pattern.
This time I’m using the Motion Sequencer as an envelope – at a much faster rate and without any looping – to control a key FM parameter for emulating hard sync, the Carrier Operator coarse frequency. As part of diving in to that, let’s make a quick side trip to another programming tip and look at why I’m using three Parts for this Performance. Each of the three Parts are different from each other and highlight the concept of ‘component voicing’ where each Part creates a portion of the final sound. The Parts can be drastically different, say in using 4 FM-X Parts all using Algorithm 1 each supplying eight sine waves, so you have a total of 64 partials to do additive synthesis. Or using each additional Part to supply more components of the final sound once you’ve maxed out the capabilities of the first Part, etc. (an old TX816 trick). In this example, Part 3 creates the underlying synth lead timbre. Part 1 enhances the lower end of the hard sync effect. Part 2 is doing the main portion of the hard sync, so we’ll dive in there. But first, take a moment to mute or solo each of the Parts to hear their contribution to the final sound, and double check the effects are off (remember Assignable Switch 2 toggles effect off and on).
Now let’s continue to the programming of the hard sync effect. Select Scene 3, and mute Parts 1 and 3 leaving only Part 2 active. I’m using Algorithm 69, which is two 4 Operator stacks, using both the 1-4 and 5-8 stacks contributing to the sync effect:
As I mentioned previously, the foundation for the hard sync effect is a fast modulation of the Frequency Ratio in harmonic series steps for our Carrier Operators 4 and 8. Go into Edit mode for Part 2 and go to the Edit – Part 2 – Common page and let’s look at the Control Assign for Knob 2:
You can see we’re controlling the Frequency and Detune of the Carrier Operators 4 and 8, utilizing custom User Curves for both Frequency and Detune to step the Frequency modulations through the harmonic series from 1.00 to 8.00. Notice we’re also modulating the Frequency and Detune of two of the Modulators – Operator 2, the third operator in the 1-4 stack, and Operator 5, the top operator in the 5-8 stack. This is to enhance the sync effect and give a slightly different flavor from each stack. To hear the differences between the stacks, selectively mute Operators 4 and 8 to hear each stack in isolation.
You may be thinking – ‘Hey, I thought you said you were controlling these parameters with the Motion Sequencer, what’s the Knob got to do with it?’ Good question, for which the answer is the next tip that I’ll walk you through. Exit to the Performance Home page and select the Motion Control – Super Knob subpage:
You can see I turned on the Super Knob link for Assignable Knob 2, and the Super Knob MS (Motion Sequence) has been turned on as well. So now the Motion Sequencer for the SuperKnob is controlling our Carrier Frequency Ratio, creating the sync effect. This highlights a conceptual decision in building our final sound: Do we: 1. Use the individual Part level Motion Sequences where we have 4 Lanes available to be assigned as control sources as covered in the “MFMX MultiMS Example” Performance from the prior article, OR 2. Use the Performance level Motion Sequencer via the SuperKnob link as in this performance?
The answer is – if you have multiple Parts that you want to use Motion Sequences to modulate a single or group of destinations in the same way, then assign a Knob to be their control source. That way you only have to create a Motion Sequence once at the Performance level then use the SuperKnob link to send it to all the desired destinations within all the Parts. Otherwise you have to duplicate your programming of the same Motion Sequence into every Part and sequence slot which can take a lot of effort. But more importantly, remember there’s a maximum of 8 total Motion Sequence Lanes than can be active at any one time and play in a Performance, so you should avoid allocating them to play redundant sequences at the Part level.
So in this instance I did not want to create and duplicate the same 8 Motion Sequences in two different Parts using 2 Lanes of the 8 available, thus I chose not to use the Part Motion Sequences. The fun added bonus of using SuperKnob Motion Sequences linked to the Assignable Knobs is watching all the panel LED’s move around ! So, another programming tip – consider using SuperKnob Link to Assignable Knobs to save Motion Sequences resources. Please watch “Tutorial Video 4 part 4” below to see a lot of this in more detail:
Let’s go back to the Live Set and look at some Performances to look at some some more interesting behaviors, tips and tricks.
MFMX GalaxySlowJam, MFMX Internexus and MFMX Internexus 2: These Performances have the ARP/MS selection set as the default for the Assignable Knobs, so they’re immediately ready for you to play around with available articulation and timing parameters of the Arpeggiator and Motion Sequences. Play around with Swing, Unit, and Gate time for changing characteristics of the polyrhythmic interplay of the Parts. Play around with the Amplitude, Shape and Smooth to control characteristics of the Motion Sequence modulation. Please watch “Tutorial Video 4 part 5” below to see the example using the Knobs to change of the various Arpeggiator and Motion Sequence parameters for “MFMX Galaxy Slow Jam”:
This allows you to alter the shape and contour of them in real time. Depending on the original Pulse shape and initial settings, this can create both subtle or wild variations in your Motion Sequence modulations. Here’s some values to try out for those Performances to see it in action -- Remember to listen to all the Scenes!:
To again review what’s occurring with the Motion Sequencer Shape Knob, review “Tutorial Video 4 part 2” for “MFMX Krazy 8 HarmSwp”, which shows in deep detail how the Shape Knob will control the Prm 1 and Prm 2 values of the Motion Sequence’s Pulse A and Pulse B when their Control flag is set to “ON” (just before the 9 minute mark of the video).
Whew, we’ve covered quite a bit, and there’s even a lot more capabilities we haven’t had time to explore in depth, such as the powerful multimode filter and the versatile Effects. I’ll end with one last programming trick, using Effects modulation to enhance the FM-X Synthesis modulations. Go to the Live Set and select the performance “MFMX DidgidooVoxLdFX”:
This is a sound is particular favorite of mine, a very unique and Zawinul-esque monophonic lead sound. First, play and hold the E below middle C (E2) for a good 10 seconds or more and listen to the rhythmic shifting vocal formant effect inspired by the didgeridoo. The majority of the timbral shifting is created within the FM-X engine using multiple complementary Motion Sequences controlling Operator Level, coarse Frequency, and Resonance. Then that is enhanced using the Motion Sequences to shift the Vowel Formant of the Talking Modulator Effect:
This is covered in detail in the “Tutorial Video 4 part 6” below:
What I’ve described here and in the video is just from Part 1 of the 5 Part Performance. Each of the other Parts have variations of the modulations and different formats from fixed frequency Operators and Spectral waves to add to the final overall sound. The result of this combined approach of complementary synthesis and effects modulations is a much enhanced and interesting timbral modulation compared to using either approach on their own. For fun, Assignable Switch 1 will change to poly mode to play chords, and you can use Assignable Knob 1 to soften the attack for a more pad like envelope.
Alrighty then, that’s a wrap! I hope this article series, videos and examples have fueled your curiosity to jump into begin making you own Performances, and help lessen the mystery of diving into programming FM-X in the MONTAGE. Please check out all the accompanying videos here to see these concepts explained in more detail and make sure you check out all of the Performances in the Live Set.
Download the library file for all the Performances referenced in the article series here: MannyFMX_4
Comments or questions? Join the conversation about this session on the Forum here.
And if you missed an earlier session - check them out:
Manny Fernandez has been involved with sound programming and synthesizer development for over 30 years. Initially self taught on an ARP Odyssey and Sequential Pro-One, he also studied academically on Buchla modular systems in the early 80's. With that solid background in analog synthesis, he then dove into digital systems with release of the original DX7. Along with his aftermarket programming for Sound Source Unlimited, he is well known for his factory FM programming work on Yamaha's DX7II, SY77, SY99, FS1R and DX200 as well as the VL1 and VL70 physical modeling synthesizers.
With the release of RefaceDx and Montage, Manny has contributed many FM synthesis programming tutorials on YamahaSynth.com. In addition, he has been a frequent guest on the Behind The Synth podcast series, sharing his insights of the development process of many of Yamaha’s significant digital synthesizers.