Layers of matrices with round robin linked

[nar]

Hi there,

I've just started using the tx16wx with reaper to try and build a multisampled drum kit, using a matrix for each instrument. The problem is that I have close and overhead and room mics for each drum/cymbal and I can't figure out how to sync the round robin so that the correct samples play together.

For example I have a matrix 4 across and 2 up for close mics, and another one of the same for overhead mics. Is there a way to ensure that when close sample A1 plays, so will OH sample A1?

Of course, they do do this if they happen to start on the same one, but if I click on a matrix square this messes it up because it counts as a triggering.

Thanks

[elcallio]

You are definitely using RR in a fashion I had not considered, though it has a certain logic to it I guess.

First off, the trigger counter on the matrix is only incremented on an actual note-on (though clicking in the mapper editor keyboard counts as one). But its per-matrix, so if the same matrix is mapped to several groups/programs, each note, on possibly separate channels will indeed increment the counter. So keeping two matrix RR in sync is gonna start failing quickly indeed.

One option for you is to not use RR, but either an external controller (allowing manual control of the sound), or map a LFO set to sync:none as the X-source instead. Downside here is of course still that it would not necessarily mean a new sound every note-on.
...

IDEAS FOR THE FUTURE:
Your conundrum has made me think of a couple of potential enhancements, and as usual I'd like some feedback from the community on these before I commit anything:

1.) Group/matrix trigger counters should reset on either/or of All notes off/All sounds off. Reset would then be to "least common denominator" of active mapping.
2.) Ability to use group trigger counter in matrix RR. I.e. all matrices in a group can share same counter, along with group switching.
3.) Ability to use a per-midi-note counter instead. I.e. each MIDI note keeps a trigger/hit count. (Btw, note counters should definitely reset on All Notes off)
4.) Move matrix trigger count to its mapping. I.e. make it unique per split.

So, what do you people think of the ideas? Which are good, which are pointless?

Cheers
/C

[IXIX]

I think all four ideas are good. Different methods would be probably be useful in different situations.

[dexterbella]

Personally I like option 2 best.

[Burzukh_Studios]

Agreed. Option 2 is the most straight forward implementation in terms of usage.

[Eleventh]

Could a possible implementation of number 2 be an option to make the matrices work in three dimensions? Z-dimension would be used for different sets of samples.

For example in the case of a snare matrix, x would be round robin, y would be velocity, and z would be different mics (close, bottom, oh, etc.), and all the layers would share the same cell structure, which would fix the round robin problem. You could also use random instead of RR, and it would still trigger the appropriate sample from each of the z-layers.

Then you could just map the matrix to overlapping groups in the keyboard mapper, each triggering one of the mic layers and sending it to it's appropriate output.

I think this would be a logical implementation. What do you think?

[Burzukh_Studios]

Thats's a nice way of putting it @ Eleventh.

The problem is with close , bottom and overhead etc samples control.

If they are seperate matrixes = seperate groups, as is the case now, they can be multi-outed to seperate audio channels / outs = tracks. In this case,the mixing/ volume / processing between these elements can easily be controlled through DAW faders/fx as seperate tracks.

If it were a single 3D matrix , this would not be possible, UNLESS multi-outing is implemeted in the wav matrix itself, which may create conflict with the mapper's group outs.

[elcallio]

Actually, you can use group switching to create a z-vector.

[Eleventh]

If they are seperate matrixes = seperate groups, as is the case now, they can be multi-outed to seperate audio channels / outs = tracks.

If it were a single 3D matrix , this would not be possible, UNLESS multi-outing is implemeted in the wav matrix itself, which may create conflict with the mapper's group outs.

The idea was that the different z-layers of the same matrix could be mapped to different groups, but they'd still be part of the same x-y structure, so as to trigger the correct velocity and round robin for each of the mics.

Actually, you can use group switching to create a z-vector.

Yes, but does it help in this case, where you need to sync the round robin & velocity across several matrices to trigger the corresponding samples from different microphones? I couldn't find anything useful in the manual.

[nar]

Yes the z layers idea was what I was thinking of in the first place, but not having multi outs would be a pain. These other suggestions are getting a bit confusing so I think I need to have a bit of practice with the manual open!

Could there not be a drop down menu of RR counter groups, much like choke groups? This is probably much more complicated than I think it is though.

PS this is a great vst though!
Also, is there somewhere to list/post user generated content? The kvr bank has only one thing on it, maybe a sticky thread could encourage people to share some things?

[Eleventh]

Yes the z layers idea was what I was thinking of in the first place, but not having multi outs would be a pain.

The idea was that the z-layers could be mapped to different groups and hence to different outputs.

Could there not be a drop down menu of RR counter groups, much like choke groups?

Then again, a couple of user-defined RR counters would do the trick much more simply. I think Battery 3 does things that way, IIRC.

Also, is there somewhere to list/post user generated content? The kvr bank has only one thing on it, maybe a sticky thread could encourage people to share some things?

I'd like to know about this too. I recently multisampled my electric bass, and would like to share the samples with other tx16wx pro users, when I'm done setting the loop points to all of the samples (a hell of a chore).