I'm developing so many things concurrently that you might have to wait a bit, I'm afraid.

LFO favorite features

audio rate
delay
fade in
bi/uni polar
phase control and modulation
rate mod
s&h
tempo sync (and not)
various waveforms + random


Couple quick examples of audio rate LFO from ACE...

http://draigathar.org/sounds/Ace4.mp3

http://draigathar.org/sounds/Ace5.mp3

An LFO with start delay and time and depth controlled by seperate LFO's
All of them synced to tempo.. but different lengths (prime lengths)

Playing the same note twice is so.. dull.

DCAM's modulation system is very advanced and it's very good at showing all modulation assignments in one view. This is one of the best systems to imitate, should one look to implement an easy to use, fast system with great visual feedback.

Let's banish mod matrix for good.

No cables. Just modulation range indicators on knobs. See DCAM.



Not sure what you mean there.

An annoyance with slow meandering LFOs is that in a normal implimentation they'll produce different results every time you run the track. I prefer LFOs that get their current phase position from the actual song position, but I don't know what synths do this offhand. That way you can build around the changes, and respond to them in the arrangement.

In a similar way, it'd be nice if noisey waves could be based on a seed so they're they follow the same course through a song.

Unipolar option also gets a +1 from me. Rapid squarewave octave and fifth trills are something I like to make but it's a drag if it's bipolar and you then have to correct the tuning of the oscillator because it's out a tritone (in the case of octave modulation). I'm also a fan of guitar-style vibrato.

Smoothing is not only important in S+H waveforms - any wave with discontinuities such as square/pulse and sawtooth benefit from a way of low-pass filtering them, it's so annoying when you have a really cool effect you can't use because it's making a horrible click sometimes.

Being able to phase-distort (stretch or 'PWM') any waveform is especially useful for tempo synced modulations - if it's say modulating a filter, having the filter highlighting the right harmonics at the right time can really make or break an effect, so being able to distort the wave is a very convenient way to do this. If the same technique can be applied to 2 cycles of the wave at a time, you can create shuffle/groove effects to match the sequencing.

Triangle waves are good for neutral modulation, while sine waves favour the extremities, but there's no equivalent for favouring the middle, so sometimes I wish for a wave based on an inverted circle that is acc/decelerating slowly in the middle ranges and rapidly at the extremes, to the point of bouncing off of the top and bottom of the ranges, creating symmetrical spikes. Ideally, a way to morph between Sine > Triangle > Inverted Circle lets you choose the exact flavour. DCAM Etch has something like this, though the spikey wave could use a bit more spikiness

Taking the morphing to extremes, I find the ideal "wavetable" for choosing/morphing waves like this:

Square > Rounded Square > Circle > Sine > Triangle > Inverted/Concave/Spikey > Maximumly sharp concave wave (Impulses). With the "PWM" value you can turn triangle into up/down Saw, etc.

Dcam has some very clever -and powerfull- modulations.

However, I disagree to the fact that you can see all modulations in a single view.

If I want to see the mods for one of the 8 perform knobs I have :

1/ To select it, then direct my eyes to browse all the GUI to pick and decipher all the yellow modulations one by one.
2/ Yet I only "see" the result for one perf slot, that I have to store/fix in my brain. The -eventually active- seven others slots, I have to select by turn or mouse over to see what they do : I dont have a global view.

Maybe theres one special place where you can see it all though, that I'm not aware of.

This is very diferent to a central place where all mods are shown in a "traditional" mod matrix. ( I say traditional because indeed matrixes are shown in many different ways in different synthesizers ) With these implementations, I can really see all the mods running in a simple glance, and understand my patch very easily, also change the modulations very quickly if desired/necessary.

So Dcam system might be the best ... for you. To me its only one of the possible ways to give visual feedback, as is a clearly implemented modulation matrix. To each his own.

I however won't say it would in any way prevent me to work with Dcam. It's mod system is very nice and really powerfull. Also please remember : I'm only discussing the visual feedback aspect. Not the system as a whole.

So let's bannish all the "this is the best" assertions, and replace them by "its the best for me" or "This system works very fine for me, and I prefer it". That is, if you want to let some room for discussion ......

There is a plus to being able to 'see' the nooks and crannies of a patch, but I think thats the point. A mod matrix technically displays that, but it is not easy to 'see' the patch. The grid of digits has to evolve into something.
I get that a hardcore sound designer might not really care for the frilly gui and just wants everything utilitarian sometimes, but I cant imagine it is especially inspring for your customers to have to stare in the abyss of numbers if they want to try and figure out some of your magic.

I love Massive's routing method. It doesnt allow a one page view of the whole patch, but it does show enough to at least let you have a good guess at what is going on behind the tabs that arent in view.
It doesnt bother me much in the moment, but I have wondered about ways to improve it.
So, what on the gui can happen to reflect changes to additional parameters;

-What if the 'surface' of the knob was used for visual feedback as well as the ring around it?

-What if the ring around the knob got wider radially as a parameter changed?

-What if there were radial lines that segmented the ring, and the segments got smaller/number of lines increased as the value increased?

-What if the hue/shade/saturation of the color of the ring changed?
I guess a lot of cool stuff could be done with colors changing/fading, but that might take more horsepower and more calibration...

I dunno, maybe all this has been done somewhere I havent seen.

This is just your idea, but you frequently try to impose it on everyone. Which is exactly saying your way is the best.

If someone thinks that something is best, they can think that. There is just as much room for that idea as for your idea that there is no best. Your idea is not superior.

This happens within a couple of seconds. No "deciphering" is called for. You look, see, done.

Conversely, you will also spend time going through a mod matrix as most will have several windows/tabs, and you still need to "direct your eyes to browse" all mod matrix entries and select different mod windows.



Having spend an enormous amount of time in that DCAM modulation system, I can assure anyone that no fixing of anything in ones brain is needed. You make it sound like rocket science, and instead its child's play.

There are several benefits of a system like the one used by DCAM, which scores highly over a traditional mod matrix.

1. Firstly, is extremely easy and intuitive to assign sources. You just look at the interface and go to any and every(!) knob/slider/switch that strikes your fancy. In a mod matrix, you need to wade through a spreadsheet list. Personally, I'm all for ease of use and an intuitive process, which the mod matrix isn't an example of.

2. Secondly, and perhaps most importantly, it's the way one, single destination can modulate as many sources as one would wish for - all within one modulation slot.
So to give an example, to assign amp attack envelope, filter attack envelope, filter envelope amount, to velocity, I have to use three modulation slots in a mod matrix, but only one in DCAM. This is an easy example, but with more complex modulation assignments, it is easy to use lots of mod matrix slots just for one destination - very time consuming to set up, and not very intuitive (with that spreadsheet list).




In a traditional mod matrix you still need to cycle though the windows/tabs, if the mod matrix has lots of mod slots.


I don't recall saying in this thread that DCAM mod system is the best. I said "it is one of the best ones to imitate". Quite a difference. Then do we really need to punctuate our posts by IMO, IMHO? Or isn't it taken for granted that each one of us expresses his own views???

So yeah, let's banish the archaic mod matrix. There must be better ways of assigning modulation, even better than the one used in DCAM, Circle, Massive etal.

edit: typos
Last edited by himalaya on Fri May 25, 2012 5:10 pm; edited 1 time in total

I fully agree to your secondly, wich is what I think is very clever in Dcam modulation system. It spares you matrix slots, and gathers the info, at the price that the eye has to find the related elements by looking all over the UI. 8 times ... Firstly could be discussed abit more , but yes its also fine.

However ......

You say "you look, see, done". My experience is : you select a slot ( in a matrix they are all already exposed ), you look at the GUI, all the GUI ( sure you did not miss one yellow thing ?), then you quickly associate the gui elements with what they stand for (in a matrix you just read diretctly cutoff ), then your brain stores all that ( because if you select another slot the info will not be displayed anymore, in a matrix you always see it ) and you're ready to go for the next seven slots. Ok this is done very quickly, as you've worked hours and hours with this synth, you associate quickly GUI elements with their function, and all this is done quite quickly. I agree. Still, the info presented in a matrix deosnt require all this brainstorming. So eventually you find it more intuitive, but faster is questionable, obsolete is questionable ( because like you I'm also for the fastest workflows )

So I would not call this rocket science, its just analysing what you have to do to fullfill a particular task in a logical sequence of events.

So, I still disagree with you on this point, but hope you'll be eventually pleasantly surprised in the near future.

With Massive or Dcam I spend way more time to understand, decipher, or recall a patch that in a correctly implemented mod matrix. So maybe my brain is not formatted the right way

Interesting imo is that you cant have the best of both systems. As I dont find it more easy to drag and drop from a slot to a target control, than choosing a control in a dropdown list, I mainly focus on the "understanding of a patch" and "quick visual feedback" thing.

Note : I could also argue that 8 slots, actually 8 sources, might be a bit restrictive, VS 16 mod matrix slots for example, where you can go up to 16 different sources, or that the eventual bipolar ammount knob associated to sources/dest in a matrix allows you to make negative or positive ammounts very easily, but its OK, with the multiple targets, you can still do a lot, these limitations are really not that important imo, and I've had pleasure with synths which had much less that this. Also this would be targetting Dcam as a synth, when I wanted to mainly discuss the pros and cons of systems, and the concepts they are based on ( In other words, maybe Dcam will have one day 16 slots, or 24 who knows ) Also 8 sources with infinite number of targets can actually do MORE XY modulations than 32 sources with 32 targets, giving a certain advantage to such systems

What is interesting also, is that you still have to select the source in a myriad of possible ones using a menu, wich is more complicated because you dont have the polarity that gives the ammount knob. So in a way half of the system ( sources ) is still matrix mod orientated.

So I like Dcam modulations system, and it changes a bit from the standardised thing, wich is fun and fresh, but then ..... I still prefer the mod matrix system, so until I find something that really makes it obsolete for me, theres no way it will be banished, and actually will be prefered. ( But its only a small part of a synth so no worries )

As for the imo, when you use such words as "the best" and then "let's bannish" yes, a few imo could be seen as welcomed. ..... (imo)

Then, we agree to the most important thing : There are probably better ways to handle modulations than the systems that exist atm.

My main concern was to drive attention to what I consider to be very important for sound design as well as for all users : Beeing able to decypher/understand a patch in the fastest possible way, and not only for modulations, and I'm happy to see that, other than on the methods to employ or implement to reach this goal, you also agree with me on that point.

What happens if you want to apply 3 different modulators to a single target like cutoff?

Actually, I don't. I don't think the focus should be on how well a patch can be "decyphered". This shouldn't be the main concern. What should be though, is how fast, easy and intuitively a patch can be set up in the first place, then there will be no need for anyone to decipher already made patches - people will find it easier to make their own. And how it is possible is outlined in that bullet point number 1, the one which you disagree with. To disagree with such superb ease of use...

In other words, it's the creation process which I find more important, then the analysis process. And when I create a patch, I rarely find it necessary to analyse what is connected to where. And when I do, the DCAM system has never failed me in assisting how quickly I can analyze the modulation connections.
Last edited by himalaya on Fri May 25, 2012 10:51 am; edited 1 time in total

What happens? Once you assign those three modulators, the cutoff will wiggle in three different ways. Was this a Koan?

Ok, just teasing. You are surely getting at the fact that to do what you propose, three mod slots need to be used - just like in a mod matrix. It just means that there are similarities in both systems - which nobody has denied.

It would be quite a job to represent velocity, breath, aftertouch, etc. modulators as a graphic form ( like the 'targets' are) and deviate from the 'list' form. So here the remnants of the mod matrix still remain.

I really like the visual feedback. I wish there was some visual feedback in Zebra. In Zebra there is a regular mod matrix and no visual feedback. This is mitigated by there being dedicated modulation knobs for many common parameters. For example, each filter has 2 assignable knobs for cutoff plus a dedicated keyfollow. Each Osc has 8 assignable modulation knobs. Etc

For many presets in Zebra, there is no need to use the mod matrix at all because the dedicated knobs are enough.

For Cypher especially, 8 transmod slots could be running short in some cases. In that case you have to turn to Fusor which adds complexity and offsets the advantage you are speaking about. But that is a potential complaint only in quantity, not design.

Design-wise, I think I agree with you about the Transmod method being better than a mod matrix, but the 8 slots are too few. I regularly use more than 8 modulators. Cypher should have at least 12 slots. Velocity, Keyfollow, AT, Expression, LFO1, LFO2, Env1, Env2, Env3, Unison, Random, Modwheel...

I also think that the Transmod method has its advantage more on a synth with all one page design. With synths with multiple pages, you could not look at all LFO1 targets as easily as with Cypher.