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. Monday, 10 June 2019
One thing that could be improved on with FM-X on the Montage/MODX synths is the variety of Algorithms available, especially with the 2 additional operators and how they are set as carrier operators in most of the algorithms. It would be great if Yamaha would add/provide more algorithms whereby they are stacked as carrier & modulator.

Having said that, over the weekend I took a 'voice strings' sound that I had programmed from scratch on my DX7II over 30 years ago, and decided to see if I could enhance it. I found another sound that has a 'ghost-like' sound to it and broke it down, stripping out all operators and LFO settings until it had just a basic 'Awww' sound. Turns out it was 2 Carrier operators that provide this. After adding them to my existing 'voice strings' sound, which had those carrier operators volume set to '0', adjusting the volume of the 2 new operators along with several other common and operator settings, and with the "Pitch Change" effect giving them a more Unison voice quality, it definitely improved my old 'voice strings' sound!

Having enhanced a sound that I thought was great to begin with, I can imagine the possibilities of enhancing other existing DX7 sounds if there were more stacked carrier/modulator operator algorithm options. If it is possible, this sounds like a good one for IdeaScale! ;)
Responses (14)
Accepted Answer Pending Moderation
Does alg. 67 match what you're asking for?
  1. more than a month ago
  2. MONTAGE
  3. # 1
Accepted Answer Pending Moderation
Does alg. 67 match what you're asking for?

Yes, alg. 67, as well as 'most' of the algorithms in the 66 - 85 range match as well.

It's mainly the ones in the alg. 1 - 65 range that have operators 1 & 2 or 7 & 8 as the additional carriers, where it would be great if they added 65 new algorithms in a future update, that are basically the same as the ones in the 1 - 65 range, but instead of having both 1 & 2 or 7 & 8 as carriers, I propose they make 2 the carrier & 1 stacked above it as the modulator (same for the alg. in that range with operators 7 & 8 ... make 8 the carrier & 7 stacked above it as the modulator).

The sound I had created on my DX7 many moons ago uses alg. 26! Luckily the enhancement I wanted to make was to have two carriers, so that worked out perfect for me, as I just made use of operators 1 & 2.
In the future at some point, I may want to take a sound/voice/PART that is built on alg. 26 (which was a very commonly used alg. on the DX7 and many of my best sounds were constructed using it), but I may find & want to add a stacked part from another sound/voice. That's when it would be great to have a new version of alg. 26 that has 1 & 2 stacked...
  1. more than a month ago
  2. MONTAGE
  3. # 2
Accepted Answer Pending Moderation
I would upload a picture of some examples, but I keep getting the error message "The file Alg2019d.JPG, which you are attempting to upload is in an unsupported format.", and the .jpg I am trying to upload is quite small and definitely not near the 2MB limit...!?

Alg2019d.JPG
Attachments (1)
  1. more than a month ago
  2. MONTAGE
  3. # 3
Accepted Answer Pending Moderation
Your 25 edit is already 85. But the picture makes it clear what you're after. Specific algorithms which are currently wider to take the two lone carriers and add algs. that do not currently exist making one carrier a modulator of the other to create a "2 op" stack.

This isn't likely to happen soon (perhaps never). What you can do is add a second part to gain 2op stacks if you need more. You have up to 8 DX7's at your disposal (ignoring the icing of 2 extra ops per). So assuming you have a free PART in slots 1-8, you can mix in another FM-X PART.
  1. more than a month ago
  2. MONTAGE
  3. # 4
Accepted Answer Pending Moderation
Your 25 edit is already 85. But the picture makes it clear what you're after. Specific algorithms which are currently wider to take the two lone carriers and add algs. that do not currently exist making one carrier a modulator of the other to create a "2 op" stack.

This isn't likely to happen soon (perhaps never). What you can do is add a second part to gain 2op stacks if you need more. You have up to 8 DX7's at your disposal (ignoring the icing of 2 extra ops per). So assuming you have a free PART in slots 1-8, you can mix in another FM-X PART.

I'm just surprised to see that the .jpg actually uploaded, because it wasn't there when I tried uploading it yesterday.
I'd be curious to hear from a programming engineer if adding new algorithms is a lot of work or not!?

My basic ask and the reason behind it is that for the majority of algorithms in FM-X, those having the "two lone carriers" are not as useful and will likely not get as much use by most people as a "2 op" stack, so adding these new algorithms (or replacing the 'two lone carrier' ones) would be much more useful for most people.

Especially ones like alg. 26, as per picture above...
  1. more than a month ago
  2. MONTAGE
  3. # 5
Accepted Answer Pending Moderation
I think the request is fairly well articulated now with the picture. There is no replacing algorithms. Cannot do this and pull the rug out from those who use those algorithms along with any presets that may use the algorithms you suggested to alternatively replace. Has to be add, not replace. If this is possible or not is probably not something you'll learn - nor the complexity for the change. Usual path forward is you either magically see your suggestion in later firmware or not. Given you can add another PART for more complex combinations of operator geometries (and mix/match of them) - it's unlikely something like this which can conserve PARTs is going to make the cut. Other suggestions around optimizations to allow for squeezing more out of less have been met with much resistance - which is the broad category this suggestion fits into. But who knows - maybe it's low enough hanging fruit that something will happen.

Please do add a PART with 2-operator stacks if you need more of those and see how that works for you in the interim.
  1. more than a month ago
  2. MONTAGE
  3. # 6
Accepted Answer Pending Moderation
There is no replacing algorithms. Cannot do this and pull the rug out from those who use those algorithms along with any presets that may use the algorithms you suggested to alternatively replace. Has to be add, not replace. If this is possible or not is probably not something you'll learn - nor the complexity for the change. Usual path forward is you either magically see your suggestion in later firmware or not. Given you can add another PART for more complex combinations of operator geometries (and mix/match of them) - it's unlikely something like this which can conserve PARTs is going to make the cut.

I was definitely thinking adding new algorithms and not replacing existing. It's a bit unfortunate that they chose to add the 2 additional operators both as carriers instead of a carrier/modulator "2 op" stack on many of the new FM-X algorithms. I would think most FM-X users/programmers would have preferred & get most use out of the "2 op" stack; however what do I know...I just made use of the two carriers to enhance a sound I had programmed from scratch 30+ years ago that uses alg. 26. :p

Still, if there is enough people suggest it, it might not be too much of a programming change to add new algorithms in the future, so I have posted my Yamahasynth.Ideascale.com idea URL below for anyone who reads this thread. If anyone likes this idea, please 'Up Vote' it ;)

"Add new FM-X Algorithms with the 2 Additional Operators Stacked:"
https://yamahasynth.ideascale.com/a/dtd/Add-new-FM-X-Algorithms-with-the-2-Additional-Operators-Stacked/233604-45978
  1. more than a month ago
  2. MONTAGE
  3. # 7
Accepted Answer Pending Moderation
Just FYI. As pointed out in several of the FM-X tutorials here on the site, a single Operator can be the source of a complex wave shape. They can be used, as Dr. Manny says, ‘to fix a hole’, to fill in high harmonics, or to create inharmonic noises. There are many uses for the lone Carrier... particularly when using it to create a complex Waveform.

Since the FM-X engine in the MONTAGE/MODX goes even beyond the 48 Operator TX816, you can use 64 Operators to build an instrument sound. It is logical that the lone operator Carriers are made available as the addition to the 6-Op Algorithms. I highly recommend spending time with a single Carrier Operator, you will be amazed at just how flexible and powerful it is (all alone)

If you go through Manny’s “FM-Xpert” series of articles you know and have heard the usefulness of the lone Carrier. When ‘component modeling’ there is no substitute. Have you gone through Dr. Fernandez’ tutorials?
  1. more than a month ago
  2. MONTAGE
  3. # 8
Accepted Answer Pending Moderation
Just to illustrate what you can do with two PARTs:
https://i.ibb.co/gvxzQM5/Montage-Alg-Suggestion.png

Apply similar techniques for your other two desired algs. I could have picked any other alg. with a 2op stack. I chose one that has some other things available. Brings in an extra 3 2-op stacks and a single lone one. Then an optional parallel modulator on one of the 2-op stacks.

In the picture, I have part 1 ops without a dot ("." ) and part 2 ops have the dot.
  1. more than a month ago
  2. MONTAGE
  3. # 9
Accepted Answer Pending Moderation
Just FYI. As pointed out in several of the FM-X tutorials here on the site, a single Operator can be the source of a complex wave shape. They can be used, as Dr. Manny says, ‘to fix a hole’, to fill in high harmonics, or to create inharmonic noises. There are many uses for the lone Carrier... particularly when using it to create a complex Waveform.

I highly recommend spending time with a single Carrier Operator, you will be amazed at just how flexible and powerful it is (all alone)

I have no doubt that a lone carrier is very powerful and useful, as I just made use of two of them to enhance a sound that I thought was already great from my DX7II. I found another sound that might complement the main sound, and after stripping it down to the sound I was looking for, it ended up by chance to be the two carrier ops that I needed, so in this instance it worked out perfectly; however seeing that there are 2 additional ops, most of my FM programming has not been building sounds from scratch, but rather by taking existing sounds and modifying/tweaking them or taking two or three voices/sounds, stripping them down to use just the operators that I want, and find an algorithm that will allow me to merge those 2 or 3 different partial algorithms together.

Having two additional operators provides more flexibility and options for doing this. In some cases, I could definitely see using a particular sound/voice/PART, such as one that uses algorithm #26, and wanting to add a 2 op stack from another sound/voice. My ask is basically that Yamaha, in a future OS update, add some additional algorithms to the FM-X engine, the same as all the ones with the 2 additional carrier ops, and stack those 2 additional operators, but leave/keep all the 2 carrier algorithms (just add new ones), providing a bit more variety and possibilities, for programmers that like to merge operator formations from 2 or more sounds and could make good use of a 2 op stack as well as the existing 2 carrier versions. Hopefully this would not be too difficult or take too much time for Yamaha's FM-X engineers to implement..!? But regardless, it never hurts to put it out there and ask!

Maybe adding to and enhancing the already great FM-X engine, would bring more interest & spark additional renewed enthusiasm to those who are not too familiar with FM-X programming, thus setting the Montage/MODX apart from the competition a bit more..!?
  1. more than a month ago
  2. MONTAGE
  3. # 10
Accepted Answer Pending Moderation
Just to illustrate what you can do with two PARTs:

Apply similar techniques for your other two desired algs. I could have picked any other alg. with a 2op stack. I chose one that has some other things available. Brings in an extra 3 2-op stacks and a single lone one. Then an optional parallel modulator on one of the 2-op stacks.

That is definitely the good news about having more than 2 PARTs (unlike on the DX7II) to play with, as I can easily do as you suggest and add the 2 operator stack, etc. in another PART; however I am an operator snob and want to be able to do it all within 1 algorithm of 1 PART :p

But yeah I see what you're saying!
It never hurts to ask though, and I'm sure you are correct in saying that it's likely low on Yamaha's priority list; however if it's a somewhat easy engineering/programming change to add some new algorithms to the mix & there is enough interest stirred up on forums/ideascale/etc., then maybe they will see some value in implementing new ones in a future OS update!? It could be a good marketing thing also, plus providing more options for end user programming and enhancing an already great FM-X engine may likely stir even more interest in the FM-X engine, thus ultimately more sales!
  1. more than a month ago
  2. MONTAGE
  3. # 11
Accepted Answer Pending Moderation
I've only really seen requested improvements that fit the broad category of "I can't get this done, because the system doesn't allow any way to do XYZ" get implemented (with some exception such as MIDI receive channel assignments). Requests that fit this category - where you can "get there from here" although "burn" more resources are not well received and I have not seen these get implemented.

The way FM is "taught" with Montage/MODX is to use sliders which will be changing ratios and use ears. Not necessarily to get too microscopic on algorithms. Manny does get to this level - but generally the approach is not so "scientific". As such, increasing the amount of choices for something people generally do not relate to (on any level) is unlikely to have any impact on sales.

I'd personally rather have 4 (... or some amount of) user algorithms where I can assign any operator's destination to any other operator (make a modulator) OR the final output (make a carrier). Arguably that's more work since it requires new GUI screens and all. And I don't know how these are stored/implemented so the current amount may be a hard limit already. You're not going to really know - in all likelihood.

Since I'm a user - and don't have to implement this - I'd be great with more options and more flexibility. So I don't think the idea is bad or that it shouldn't be done. I just know there's a reality here and am trying to help convey it with a "softer landing" than you may get from an official response.
  1. more than a month ago
  2. MONTAGE
  3. # 12
Accepted Answer Pending Moderation
Daryl,

I totally hear where you're coming from... a build you own algorithm would be cool, but IMHO it's one of those things you think you want but once you have it, it ends up being less useful than you anticipated. The capability is there in the Yamaha's SY77/99 and in NI's FM8 softsynth, and a new 4Op FlowMotion from Waves.

Personally, yes I get that "if only that Operator was over here I could do X or Y easily" situation (really, I know,.. I did a lot of the FS1r presets!), but then take it as a challenge to get the same result in a different way. Ends up being a good way to keep advancing one's programming skills.

Specifically to you're request, IMHO that capability ins't anything that can be changed in firmware -- all Yamahas prior to Montage have the FM implementations coded on custom chips. I can't speak to Montage as I've never taken one apart but I highly doubt it's a fully sofware implementation as given the polyphony, it would be less expensive, higher performance and more robust to implement on chips than compared to needing a faster & more power hungry CPU and RAM space trying to run everything...

Regarding a "science" vs "ears" approach in the general sense the scientific 'math' approach turns most people off. So my personal experience in teaching FM over the years, the "ears" approach connects with most people as fundamentally it seems the most coomon approach synthesizers is the "the play around with X, Y & Z until it sounds cool". Dr Chowning himself has expressed that one's ears are more important than the math.

Manny
  1. 3 weeks ago
  2. MONTAGE
  3. # 13
Accepted Answer Pending Moderation
Daryl,
I totally hear where you're coming from... a build you own algorithm would be cool, but IMHO it's one of those things you think you want but once you have it, it ends up being less useful than you anticipated.

Hey Manny, yeah it's not near the top of my list of enhancements, but thought it wouldn't hurt to ask and put it out there to see if something comes of it (a long shot at best). I had a use for it recently, but it turned out that the correct algorithm was already there.
And like Jason said in an earlier post, I could easily just add what I want via a second PART in the Performance.


Personally, yes I get that "if only that Operator was over here I could do X or Y easily" situation (really, I know,.. I did a lot of the FS1r presets!), but then take it as a challenge to get the same result in a different way. Ends up being a good way to keep advancing one's programming skills.

True. Recently I had another challenge with a 'Unison' sound from the DX7II that was converted to four PARTs on the Montage, each detuned from the other, creating close to the same unison effect for that particular sound, and trying to get it down from 4 PARTs to just 1 PART (like how it's using 1 voice on the DX7). Jason suggested that the 'Pitch Changer' effect used in both Ins A & B might work.
I played with that effect in the A & B inserts, and it definitely made using just 1 PART close enough to the original DX7 Unison sound for my liking. Using this effect, I can also stereoize the sound; however what I found was that the more stereoized it becomes, the more it loses it's Unison-like sound. If I want both the Unison sound & stereo, I've found that using the 'Pitch Changer' effect as I described, but not stereoize the PART via that effect, and then just use 2 identical PARTs panned to opposite sides works perfectly.


IMHO that capability ins't anything that can be changed in firmware -- all Yamahas prior to Montage have the FM implementations coded on custom chips.

I never considered that, but you could definitely correct and this may be a definite NO. Not that big a deal if that is the case.
I'd far rather they increase the number of KBD CTRL PARTs that can be played from 8 to 12 or 16. Or that they allow KBD CTRL to be an option to define in different SCENEs, so that when switching it would be a bit more seamless than Mute/Unmute (or even worse...using volume at 0 instead of mute, whereby you can't hear the sound but still lose polyphony).
But my #1 thing I'd love to see as a feature/enhancement in the next OS update, would be for Yamaha to give us the C7 Piano! (or at least provide it as a free library like the Bosendorfer one)


Regarding a "science" vs "ears" approach in the general sense the scientific 'math' approach turns most people off. So my personal experience in teaching FM over the years, the "ears" approach connects with most people as fundamentally it seems the most coomon approach synthesizers is the "the play around with X, Y & Z until it sounds cool". Dr Chowning himself has expressed that one's ears are more important than the math.
Manny

I've kinda used both over the years. I've never built a sound from scratch from the Carriers up, as I've always used existing voices, disabled the operators until I get it stripped down to the sound (or part of a sound) that I'm looking to build, and then use that in my new sound. I then do the same with another sound, stripping it down to only the operators I need. Then I find an algorithm that fits, which allows me to merge the two (or three) sections together to create a new voice/sound (basically layering/merging 2 or 3 sections or sets of operators from different 2 or 3 different voices). However, I am a tweaker, so I always end up playing with mostly all the operators in the new sound that I create and use my ears to get it to where it sounds best.
  1. 3 weeks ago
  2. MONTAGE
  3. # 14
  • Page :
  • 1


There are no replies made for this post yet.
However, you are not allowed to reply to this post.
2018 © Yamaha Corporation of America and Yamaha Corporation. All rights reserved.