YamahaSynth.com Forums

This is the place to talk about all things related to Yamaha Synthesizers!
  1. Darryl
  2. Sherlock Holmes The Voice
  3. MONTAGE
  4. Sunday, 14 April 2019
Over 30 years ago I bought a Yamaha DX7II-FD and still have it. I plan to migrate some of my favorite sounds over to the Montage; however I am coming to realize that Yamaha dropped 'UNISON' as a feature..!?

When did Yamaha start dropping 'Unison' as a feature for their Synthesizers? I could almost understand it if they did this with the Motif's & AWM based synths, but for FM synthesis, one of the most useful (If not THE most useful) features was always 'Unison'! The best sounds I have on my DX7II-FD use Unison. I have some really great voice strings that I programmed from scratch, as well as some really cool synth solo sounds that I'd like to bring over.

Why would they do this and not include Unison?
This is a "MUST", not an idea to ask for when it comes to FM synthesis!

So, looking at what I could do on my DX7II-FD vs the Montage/MODX...on the DX, I could have 2 Unison based sounds split, playing my voice strings on the left side, while doing a synth solo on the right side. This is a total of 8 voices, because Unison layers the same voice sound 4 times and then allow you to detune to taste. The only easy way I can see replicating this on the Montage, is to copy the two 2 different parts 4 x each for a total of 8 parts. And since Yamaha have limited KBD CTRL to only 8 parts, it appears that after over 30 years, they have not improved the number of FM sounds (that use Unison) that can be played at the same time. Basically, just 2 different sounds...same as on the DX7II-FD ... Why? Why even include FM-X as an engine, if you are going to take away THE most import thing that makes it sound the best???
If using 4 PARTs was the thinking, then you have just put the ability of the new Montage/MODX synths back 30+ years, which is extremely limiting!!!

They could have (& could still) given us 4 x UNISON of a PART via a "Quad Harmonizer" effect at the very least, which would have done the trick!!

For FM synthesis, this is as big a deal as if they decided to leave out 'portamento' or 'pitch bend' as features!

I put this on Yamaha IdeaScale a few months back: "Montage-MODX UNISON Layering Functionality"
https://yamahasynth.ideascale.com/a/dtd/Montage-MODX-UNISON-Layering-Functionality/214572-45978

------------------------

So, working with what has been given to us. How can I replicate UNISON for my DX7II FM based sounds without having to use up 4 PARTs? Any & all options/suggestions that you know of & can share would be greatly appreciated!

What I have tried on my own that does an OK job:
- I took an FM-X part and copied it to the next available free PART location (not ideal but better than using up 3 additional PART locations)
- I setup a Stereo Chorus insert effect on on the 1st FM PART & a Stereo Flanger insert effect on the 2nd FM PART
- I then setup the Ins B on the 1st FM PART with a Basic 'Cross Delay' effect, with a very short Delay time, Feedback=0, and Input Select to 'L' so that the sound would repeat to the Left side.
- I repeated this for the 2nd FM PART, except I set the Delay time slightly different, and the Input Select to 'R'.
Responses (10)
Accepted Answer Pending Moderation
For FM synthesis, this is as big a deal as if they decided to leave out 'portamento' or 'pitch bend' as features!


They did get rid of glissando which is related to the above examples. And even pitch bend is not fully implemented in FM-X vs. what was available in the DX7.

Note that there is also an effect which lets you have two different pitch tunings to stack along with your fundamental. It's under the "Misc" category called "Pitch Change". You can change fine tuning and do a +/- fine tuning trick with it using the two pitches available. InsA, InsB, and Var effects each can support one of these effects so you can have 4 different pitch detunings by using the effect on each PART (InsA/B) then have a more global pair of pitch shifts that can be elected (or opted out) as shared among each PART (Var). A total of 6 "unison" copies by electing to use all insertion and var effects together for pitch shifts - and only focusing on a single PART. Each part gets its own pair of insertion effects so each part get its own dedicated 4 total pitch shift "unisons" to play with.

If using cross delay or chorus or other effects - you can sort of replicate the sound of detuning. Ultimately, detuning is attempting to sound like effects which were not available in the DX7. Now you have effects - you can use these directly instead of doing tricks to sound like an effect-less system has effects.

All of that said - I wish they just built upon what was already available rather than adding features and subtracting others. You end up with lots of caveats and extra work - and sometimes a road to "nowhere" (can't get there from here) dealing with what has been subtracted.
  1. one week ago
  2. MONTAGE
  3. # 1
Accepted Answer Pending Moderation
And even pitch bend is not fully implemented in FM-X vs. what was available in the DX7.

What pitch bend functionality did they leave out?


Note that there is also an effect which lets you have two different pitch tunings to stack along with your fundamental. It's under the "Misc" category called "Pitch Change". You can change fine tuning and do a +/- fine tuning trick with it using the two pitches available. InsA, InsB, and Var effects each can support one of these effects so you can have 4 different pitch detunings by using the effect on each PART (InsA/B) then have a more global pair of pitch shifts that can be elected (or opted out) as shared among each PART (Var). A total of 6 "unison" copies by electing to use all insertion and var effects together for pitch shifts - and only focusing on a single PART. Each part gets its own pair of insertion effects so each part get its own dedicated 4 total pitch shift "unisons" to play with.

When you say that this effect "lets you have two different pitch tunings to stack along with your fundamental", I just wanted to clarify that by "stack" you mean "Layer", whereby the original fundamental sound is in the middle, and the effect generates 2 additional layers that can each be detuned differently, for a total of 3 layers for one instance of the "Pitch Change" effect(InsA)!? Then an additional 2 layers that can each be detuned, if the effect is used again(InsB), for a total of 5 layers at that point. Then another 2 detunable layers if used in (Var), for a grand total of 6 layers + the original fundamental sound = 7 layers..!?

If this is the case, why did they name that effect "Pitch Change", instead of something like "Doubler" or "Harmonizer" effect?
As I searched through all the available effects to try and replicate Unison layers, I saw "Pitch Change" and thought it was just an effect that shifted the pitch of the original fundamental sound...probably because one of the effects I have in Pro Tools is named "Pitch Shift", which basically does just that, allowing you to do the 'Chipmunk-like' effect, etc. I also have effects in PT that are called "Doubler", which does up to 4 detuned unison layers on top of the fundamental sound (or you can mute the fundamental if you wish).
Too bad they gave it that particular name "Pitch Change", but if it does what I think I understand you said it does, it is basically a doubler/harmonizer effect that does exactly what I was hoping for...!

Does "Pitch Change" also have options for spreading the 2 layers L & R to stereoize the sound more?


If using cross delay or chorus or other effects - you can sort of replicate the sound of detuning. Ultimately, detuning is attempting to sound like effects which were not available in the DX7. Now you have effects - you can use these directly instead of doing tricks to sound like an effect-less system has effects.

One thing I like about the cross delay, is that it allows for a more stereo feel that wasn't available on the DX7. If the Pitch Change effect works to layer & detune as you mentioned, the delay might still be a good effect to add to provide some stereo imaging.


All of that said - I wish they just built upon what was already available rather than adding features and subtracting others. You end up with lots of caveats and extra work - and sometimes a road to "nowhere" (can't get there from here) dealing with what has been subtracted.

Agreed!
  1. one week ago
  2. MONTAGE
  3. # 2
Accepted Answer Pending Moderation
Pitch change gives you 2 pitches that are changed from the fundamental (the original). You can have a pitch as much as +/-24 coarse steps (semi-tones) and then also fine steps for less-than-a-semitone pitch shift. Either can be positive or negative. I don't know why anything is called what it is - but I can play with the effects and hear what they do. The name becomes secondary to the function at some point. It might be nice for everyone to align on names so we can learn "everything" once and then transfer to new equipment. But the world has never worked like that with each manufacturer dancing around trademarks of others or coming up with a new-fangled way to do the same old thing.

I would suggest you give Pitch Change a try. Since "all" effects have some form of Dry/Wet balance - you can dial in how much of the effect is sounding (altered pitches) vs. the original non-shifted sound.

You can call this layer if you want. I came up with my own new-fangled lingo because layered, to me, implies the tone generator triggering multiple oscillators on the same MIDI note range. Effects do not, to me, conjure up this exact kind of "layer" - so I try not to overload the language to avoid confusion. Perhaps it (not reusing "layer" ) creates confusion - but at least you know where I'm coming from.

Also, I was hoping you may take "pitch change" and run with it. Perhaps bring up the data list and do a search for "pitch change". Ctrl+F "pitch change" then two returns brings me to page 160 which shows the parameters:

1 Pitch 1 -24 – +24 (40 – 88)
2 Fine 1 -50 – +50 (14 – 114)
3 Initial Delay 1 0.1ms – 400.0ms (0 – 127) 7
4 Feedback Level 1 -63 – +63 (1 – 127)
5 Pitch 2 -24 – +24 (40 – 88)
6 Fine 2 -50 – +50 (14 – 114)
7 Initial Delay 2 0.1ms – 400.0ms (0 – 127) 7
8 Feedback Level 2 -63 – +63 (1 – 127)
9 —
10 Dry/Wet Balance D63>W – D=W – D<W63 (1 – 127)
11 Pan 1 L63 – C – R63 (1 – 127)
12 Output Level 1 0 – 127 (0 – 127)
13 Pan 2 L63 – C – R63 (1 – 127)
14 Output Level 2 0 – 127 (0 – 127)


You'll have to excuse the formatting since it's from a PDF and I'm not going to clean it up -- but I see pan 1 and pan 2. Looks like you're covered.

A 0.1mS (100uS - as in micro second) is in the broadcast-realtime target of under 10 milliseconds (100 times faster than that). Set at 0.1mS - you shouldn't be able to tell the difference between the timing of the original note and the generated pitches (aka changed pitches). But what the delay lets you do is have some pattern of one pitch sounding first and then the other sounding second. You can make interesting movement by setting these to the higher delay values and also having some delta between pitch 1 and pitch 2 - in time.

I don't know how this relates to the original DX7 "layering" deal (the delay part).

Pitch bend didn't implement the steps. Or different interpretations of which notes to bend (pitch bend mode).

FM Converter Manual gives some insight: https://yamahasynth.com/yamaha-fm-converter-documentation

Parameters that are not converted ...

Additional Voice Parameter
...
Ø Pitch Bend Step
Ø Pitch Bend Mode Normal/Low/High/Key On
Ø Portamento Step


Although there is no direct equivalent for pitch bend step - you can fake this behavior using Mod/Control->Control Assign programming. Pitch bend mode cannot be faked - it's more of a feature found on Tyros/Genos and hasn't been part of the Motif/Montage branch. Glissando can't be accomplished either.

These are just facts - you would likely be encouraged to weigh all the things Montage has going for it that the DX7 could not do and focus on what Montage does rather than wallow too long in what it doesn't.
  1. one week ago
  2. MONTAGE
  3. # 3
Accepted Answer Pending Moderation
Ok, so I tried it out and the "Pitch Change" effect in InsA & InsB will do everything I was hoping for regarding getting the 'Unison' for my DX7 sounds, and also allows for stereo imaging of the sounds as well.
_________________

Regarding 'Pitch Bend', one thing I have noticed with the Montage's version of it is that the more bend you set, the less smooth it becomes. If I set a 2 or 3 Octave bend, I can hear that it is doing less & less fine stepping instead of smooth bending. Not noticeable on anything less than a 6 semi-tone bend (like a 2 or 3 semi-tone bend), but at -/+ 6 and lower/higher it slowly starts to become noticeable. It seems to really become noticeable at the 2 octave bend mark. Just to make sure, I tried a 1 octave bend on my DX7II and it was perfectly smooth, but a 1 octave bend on any Montage part is not nearly as smooth and sounds more like fine stepping.

Is there something I have set wrong or is this just a known thing of how the Montage/MODX's pitch bend works..?
  1. one week ago
  2. MONTAGE
  3. # 4
Accepted Answer Pending Moderation
I don't think you're doing anything wrong.
  1. one week ago
  2. MONTAGE
  3. # 5
Accepted Answer Pending Moderation
Several: Points of Order...
Pitch Bend is handled by 14-bit resolution (has been since the introduction of MIDI in 1983) — higher resolution PB will be implemented as part of the new 2.0 MIDI spec. There are 16,384 increments of PB, If you Set the PB Range to -2/+2, it’s obvious that the 16,384 increments will sound smoother over that distance, as opposed to what happens when PB Range is Set to -12/+12. That’s what you’re hearing.

There is also physics involved in what happens when sampled audio is pitch shifted...

Sampled audio (since the time Audio was first recorded and played back) in this case AWM2 will “chipmunk” as you change the pitch (also called munchkinization)... because to change the pitch you are actually speeding up playback to go up in pitch and slowing down playback to go down in pitch. The laws of physics dictate this behavior. It is the change in the normally fixed frequency components that clues our ear/brain into the weirdness of the sound quality. The farther you bend it, the weirder it sounds.

The resolution of the PB message causes the grainy sound at the extremes. It’s physics that causes the strangeness. This has not changed.

FM, both in 1983, and FM-X today, are pure synthesis and do not exhibit the same chipmunking/munchkinization of the sound quality when you bend the pitch. An entirely different set of rules apply.

Portamento/Glissando, as found on the original DX7, was gone by 1986 when the mark II hit the market. There was an entire “rethink” of the parameter. Remember MIDI was brand-spanking NEW in 1983, the year the DX7 came out. I’m sure there was much discussion about Yamaha’s interpretation of these among the MIDI community of manufacturers. (I was not there but maybe we can find out how that went down).
Pitch Bend Step as well, gone before the decade was out...

Unison Mono and Unison Poly were used back in the day, DX7II (before Effect processing of any kind was found in synthesizers) to create pseudo-Effects... at great cost to the polyphony. The original DX7 was bone dry, no effects of any kind. You were folding the engine over on itself, this could be used to create a pseudo-reverb ‘effect’- and because more Operators were involved, a phatter sound was obtained, but with half the polyphony each time you folded it. Unison Poly reduced the DX7II to just 4 notes of polyphony (giving the sound a very thick tubular type sound — not unlike a phaser). Unison Mono while only one note (Mono) it was a phat, phat note... you can do that now... and add effects (imagine). Briefly, ‘unison’ was resurrected in the groovebox product, DX200.

Once Yamaha was able to put proper cutting edge Effects into our synths, the need for trading polyphony for Effects became (thankfully) unnecessary. (It has been our experience, users hate to trade Effects for polyphony). The current collection of monstrous delay line Effects is formidable— take it from someone versed in programming “unison” Voices back in the day... I understand the ‘charm factor’, but like many things back then ... I remember also why they went away.

Some think that analog synthesizers going out of tune is apart of their charm... if you were a gigging musician back in the day, there was nothing charming about it. It’s one reason digital technology had appeal... it stayed in tune. (At the end of the day the listener is not about the charm, they just notice that “the keyboard player’s synth was out-of-tune”!!! lol

Why include the FM-X engine? Really!?!
There can be eight FM-X Parts (not just two), so you can recreate TX816 programs accurately! Try “The TX816 E Piano 1”
The often used pairing of FM-X Parts is to create a stereo sound — panning one left and the other right, use of key-on delay and effects to create the illusion.

When features change, you can sit around and lament, that’s allowed. But recognize many times it’s because either a better way was found to do things, or it’s just time to move on. And then, sometimes they can be brought back... no harm in asking. I’m sure they’ll get a kick out of how it’s the most important thing that made FM sound great!! lol
  1. one week ago
  2. MONTAGE
  3. # 6
Accepted Answer Pending Moderation
Several: Points of Order...
Pitch Bend is handled by 14-bit resolution (has been since the introduction of MIDI in 1983) — higher resolution PB will be implemented as part of the new 2.0 MIDI spec. There are 16,384 increments of PB, If you Set the PB Range to -2/+2, it’s obvious that the 16,384 increments will sound smoother over that distance, as opposed to what happens when PB Range is Set to -12/+12. That’s what you’re hearing.
There is also physics involved in what happens when sampled audio is pitch shifted...

FM, both in 1983, and FM-X today, are pure synthesis and do not exhibit the same chipmunking/munchkinization of the sound quality when you bend the pitch. An entirely different set of rules apply.

Remember MIDI was brand-spanking NEW in 1983, the year the DX7 came out. I’m sure there was much discussion about Yamaha’s interpretation of these among the MIDI community of manufacturers. Pitch Bend Step as well, gone before the decade was out...

So, it appears they have inadvertently brought back Pitch Bend Step, but not in a good way. When I pitch bend a single "FM-X" PART on the Montage that is Set to -12/+12, it sounds like very fine Stepping and not smooth pitch bending like on the DX7 set to -12/+12...!? I currently have the two Synths beside each other and just tested again. The DX7II is super-smooth even if I bend really slowly, it doesn't step or micro-step at all. But the Montage does... That being said, although it's a bit unfortunate, but it's not too bad. I don't plan to pitch bend more than -12/+12 anyway, and with reverb & delay added in, it isn't nearly as noticeable as it is when bending notes that are completely 'Dry'! ;)


Unison Mono and Unison Poly were used back in the day, DX7II (before Effect processing of any kind was found in synthesizers) to create pseudo-Effects... at great cost to the polyphony. Unison Poly reduced the DX7II to just 4 notes of polyphony (giving the sound a very thick tubular type sound — not unlike a phaser). Unison Mono while only one note (Mono) it was a phat, phat note... you can do that now... and add effects (imagine).

Once Yamaha was able to put proper cutting edge Effects into our synths, the need for trading polyphony for Effects became (thankfully) unnecessary. (It has been our experience, users hate to trade Effects for polyphony). The current collection of monstrous delay line Effects is formidable— take it from someone versed in programming “unison” Voices back in the day... I understand the ‘charm factor’, but like many things back then ... I remember also why they went away.

Why include the FM-X engine? Really!?!
When features change, you can sit around and lament, that’s allowed. But recognize many times it’s because either a better way was found to do things, or it’s just time to move on. And then, sometimes they can be brought back... no harm in asking. I’m sure they’ll get a kick out of how it’s the most important thing that made FM sound great!! lol

My best DX7II sounds were all Unison based, so it is quite important to me to be able to recreate that; however Jason told me about the "Pitch Change" effects and I tried that.
Well, not only does it provide Unison like on the DX7II, but it allows me to stereoize the sound, which is awesome! Something I couldn't really do as well on the DX7II...well, if I used two opposite detuned Unison voices and selected the 'PAN' button, it would allow for stereoization, but at an even great cost to Polyphony!!
So, now that I know about the proper effect to use and have tested it, overall the Montage with its effects is a great improvement over the DX7II when it comes to Unison!
And with the extra 2 operators, I can maybe improve the old DX7II sounds a bit..!?

I know I have had concerns regarding the Montage's FM-X ability compared to the DX7II; however these are basically a non-issue, and in the case of Unison, the Montage's effects based version is far better than the DX7's overall sound.

But looking at what really matters most, the Montage AWM2 based Piano sounds and other awesome sounds, along with all the Motion Control, ARPs, Sequencing, SCENEs, etc., and the massive amount of layered/combined Polyphony, it is the best Synth on the planet IMHO!! I was struggling several months back trying to decide between the Korg Kronos and the Montage, but someone on the Korg forums mentioned about how the Kronos Polyphony was restrictive to the lowest common denominator of polyphony...that totally gave the Montage the edge, because the Montage doesn't go to the lessor of all engines' polyphony...it adds the two engines' polyphony!! My plans to potentially utilize all 16 parts during live gigs, with much of them sequenced and have the ability to do numerous layers and splits; the Montage was really the only choice! Now that I've had more time to really hear/play the CFX and Bosendorfer pianos, they are simply awesome! I can't wait to add the Synthogy American Grand Steinway, and at some point in the future, an awesome C7 piano!
  1. one week ago
  2. MONTAGE
  3. # 7
Accepted Answer Pending Moderation
I've brought up glissando a few times because it would enhance guitar "simulation" using FM-X (assuming it was only available as an FM-X feature). If glissando rate could be tied to a controller (destination) - then interesting things could be done. Guitar is one example - there are many practical applications. Arguably more in AWM2 - but FM gliss has its place.

An FM-X guitar with gliss would make a great Metheny-esque Performance.

I did cover the lack of DX7 effects and how we could go straight to real effects in Montage/MODX without all the trickery. This is an upside for sure. That the "same thing" can be done without costing polyphony is a plus. There's tons more of that (polyphony) this generation as well.

We can play a shell game - but some do want certain features from the past that do not exist. It's not something that needs to be met with surprise, defense, or offence. Worth taking a note what items from the past still have value (or would have value) and place those thought pennies in the kitty. Sometimes what we're looking for is already there (this unison business - but better now). Worth pointing that out - Montage wins. Also worth discussing if there's something missing without either the user or manufacturer getting too bent out of shape over these facts. The user needs to work with what they have - perhaps participate in aspirational suggestions and the manufacturer should listen and clear up any misconceptions when those exist.

Pitch bend may be one of those misconceptions. More detail (which Performance, setting details) would need to be given in order to drive a nail in that one. This part of the thread is going fine. There are certainly some opportunities for the wrong conclusions to be drawn depending on those missing details.
  1. one week ago
  2. MONTAGE
  3. # 8
Accepted Answer Pending Moderation
I did cover the lack of DX7 effects and how we could go straight to real effects in Montage/MODX without all the trickery. This is an upside for sure. That the "same thing" can be done without costing polyphony is a plus. There's tons more of that (polyphony) this generation as well.

We can play a shell game - but some do want certain features from the past that do not exist. It's not something that needs to be met with surprise, defense, or offence. Worth taking a note what items from the past still have value (or would have value) and place those thought pennies in the kitty. Sometimes what we're looking for is already there (this unison business - but better now). Worth pointing that out - Montage wins. Also worth discussing if there's something missing without either the user or manufacturer getting too bent out of shape over these facts.

I agree. I think they should rename the "Pitch Change" effect to "Unison"!! ;)


Pitch bend may be one of those misconceptions. More detail (which Performance, setting details) would need to be given in order to drive a nail in that one. This part of the thread is going fine. There are certainly some opportunities for the wrong conclusions to be drawn depending on those missing details.

After BM mentioned about AWM2 parts and pitch bend vs FM-X, I retested pitch bend on the 'Montage' using a Single FM-X PART Performance set to -12/+12. I set one of my single Voice Performances on the DX7II and set the pitch bend to -12/+12. It is quite noticeable when compared side by side. The DX7 is Perfectly smooth with zero detection of any mini or micro-Stepping, no matter how slow or fast I bend the pitch. The Montage just sounds like it's micro-Stepping, but the faster I bend the less noticeable...the slower I bend the more noticeable. Also, if I set the Montage to -3/+3, it is much smoother and harder to tell the difference between the Montage and the DX7 set to the same -3/+3 pitch bend. If on the other hand I set the Montage Pitch Bend to -24/+24, it is very noticeable as mini-Stepping or fine-Stepping, and that much more noticeable the slower the bend; however even the fastest bend at that setting is still noticeable and not smooth! (Note that any and all FM-X PARTs will reproduce the same issue...)
This is definitely something that Yamaha could and should fix, as I first noticed the issue when trying different electric guitar parts with distortion for doing solos. They don't sound as close to a whammy bar dive bomb/bend as they could, because of the fine Step-like sound to the pitch bend, especially since some of the guitars are set well below -12 and closer to -24 or lower! Smoothing the Pitch Bend out would definitely improve the realism with electric guitar solos on the Montage!
I would presume that it is doable and programmable, because if I set the Portamento to a slower setting and go between notes 2 to 3 octaves apart, it is perfectly smooth on the Montage and doesn't have that fine Stepping-like sound to the bending between notes that Pitch Bend does...!

It's not a deal breaker, but it is something that they could improve on for the future.


*Here is my new YamahaSynth.ideascale.com idea that I posted. Please 'Up Vote' if you'd like to see this fixed in a future OS release:

Montage/MODX Pitch Bend micro Stepping-like Sound - Not 'Smooth'
https://yamahasynth.ideascale.com/a/dtd/Montage-MODX-Pitch-Bend-micro-Stepping-like-Sound-Not-Smooth/231136-45978
  1. one week ago
  2. MONTAGE
  3. # 9
Accepted Answer Pending Moderation
Wow! You guys frighten me ...
  1. one week ago
  2. MONTAGE
  3. # 10
  • Page :
  • 1


There are no replies made for this post yet.
However, you are not allowed to reply to this post.