New FX: Shift event by fractions of a tick

[knights who say neve]

I mentioned this on Muffwiggler but I figured I should probably post the idea here:

A FX for shifting an event by fractions of a tick. 

I don't know what the internal resolution of the NerdSeq is but it's higher than 24PPQN. So I'd like to be able to use some of that resolution.

Right now if I want (for example) swung 16th note triplets to sound close to correct, I have to run the NerdSeq from a 4x external clock and input everything in 32nd notes or smaller. So this feature idea would really help with these kinds of cases, and for generally dialing in the right rhythmic feel for a part.

The syntax could be: SHFT <amount>

[knights who say neve]

OK, after further experimentation I was able to get decent 16th note triplets at normal clock rates by using groove settings like 5 3 4 and then an inverted trigger setting on the middle note to fine-tune the timing. Fortunately, the drum modules I'm using trigger cleanly on the rising edge and don't mind having the signal held high between triggers. Also fortunately, the timing of the NerdSeq is very tight and consistent! 

This solution won't work for all cases though. A SHFT command would still be very useful I think.

[XORadmin]

With shift you mean delay all events from this step by X ticks?

I am currently checking out if i can get 48ppqn internally without performance issues, then you could have double the resolution.

[XORadmin]

OK, after further experimentation I was able to get decent 16th note triplets at normal clock rates by using groove settings like 5 3 4 and then an inverted trigger setting on the middle note to fine-tune the timing. Fortunately, the drum modules I'm using trigger cleanly on the rising edge and don't mind having the signal held high between triggers. Also fortunately, the timing of the NerdSeq is very tight and consistent! 

This solution won't work for all cases though. A SHFT command would still be very useful I think.

Thats the spirit!  Get the thing done even if it is not implemented. Probably to get the whole thing complete and working for all cases, add also a kill command which will reset also your inverted trigger at some point


I will investigate the possibilities.

[knights who say neve]

With shift you mean delay all events from this step by X ticks?

I am currently checking out if i can get 48ppqn internally without performance issues, then you could have double the resolution.

I meant delaying by a fraction or percentage of a single tick. Even 4 divisions would be great, that would give the same resolution as the Roger Linn MPCs :-)

Double the resolution would be a good thing, but in that case though, would a 16th note still be 6 ticks in the groove column, or would it be 12? It would need to be 12 for the increased resolution to be useful without adding a SHFT command. That's fine with me but it might throw other users off. Maybe 48PPQN, if you do it, should be a user option and not the default.

The previously discussed idea of using letters beyond f in the groove column will help with this - extending to the letter 'o' will allow eighth notes in the groove column at 48ppqn. Pretty cool.

OK, after further experimentation I was able to get decent 16th note triplets at normal clock rates by using groove settings like 5 3 4 and then an inverted trigger setting on the middle note to fine-tune the timing. Fortunately, the drum modules I'm using trigger cleanly on the rising edge and don't mind having the signal held high between triggers. Also fortunately, the timing of the NerdSeq is very tight and consistent! 

This solution won't work for all cases though. A SHFT command would still be very useful I think.

Thats the spirit!  Get the thing done even if it is not implemented. Probably to get the whole thing complete and working for all cases, add also a kill command which will reset also your inverted trigger at some point


I will investigate the possibilities.

Awesome!

Regarding the kill command - I don't understand what the number does. Is it specifying when the kill command should start, or how long it will last?

[XORadmin]

The Kill Command sets a time when a gate will be reset again. This is not in ticks but in times, from really short milliseconds ( nice for acid basslines) up to i don't even know right now...seconds?

For the 48ppqn, of course then the groove is set to 12 automatically...and of course the user have the option to choose. I just got to see how i will implement it. For a new project it is easy...anyway, i integrated the groove being set up to 32 already, increased the max bpm to 500.

And of course, going 48 ppqn means that the groove have to change automatically. Same input clock, higher internal resolution. Just keep in mind that 1/16 clock is awfull if one want to do stuff with a higher clock...i multiply internally and the more multiplying the less high precision. In the end it is a approximation and best effort recognition of the real clock.

Just parts of this will be in the next release. But then parts that will get you a bit further already

[knights who say neve]

With 48ppqn and 500 max bpm I can double the effective resolution to 96ppqn if I really need it without using an external clock, just so long as I keep my actual tempos under 250 bpm. This is getting better and better.

I'll play with that kill command - from the manual it looks like inverted triggers are also delayed gates, and the kill command can be used to set how long the gate will stay on before being turned off. Time to get out the oscilloscope...