Sync Program Changes M:C <-> M:S

[bibenu]

I thought I had this figured out, but there’s a piece of the puzzle missing.

M:C is master
M:S is picking up the M:C clock and transport changes
M:S responds to manually selected patterns on the M:C when both are stopped.

BUT … when I run a pattern chain from M:C, M:S doesn’t pick up the changes, not on time anyway the changes are one cycle late. Both machines and both patterns on each machine (A01-A02) have the same scale CHG (128). But while M:C changes at 128, M:S doesn’t change till count 256.

What am I missing … does PCH sync not work with pattern chains ?

EDIT: Just discovered that manual PCH changes, when they are both running, also happen late, by a whole cycle.

[thermionic]

Someone on Reddit posted this-

I do this with my Model:Samples and Model:Cycles as well. Samples sends prog change on channel 8, Cycles receives on channel 8.

One thing though: on my Cycles I have to setup my pattern MODE to PATTERN and CHG to LEN. Otherwise it will not sync immediately. Not sure if the Digitakt and Digitone work similar.

[bibenu]

Thank you. I was wondering if it was those values, but haven’t yet experimented with that because what I have so far is dependent on different length tracks (e.g. a len of 4 for a 4 note arpeggio and a half-scaled track with an 8 bar melody). Hope I don’t have to change too much to make PCH sync work.

[thermionic]

Could possibly be an issue of the track lengths, can you create simple projects with those parameters and set up a few patterns to verify whether the changes follow as you’d expect?

[bibenu]

Yes, I’ll need to experiment something along those lines. I think I may start by taking the song I have and making copies, then warping them bit by bit to see what magically suddenly makes PCH sync correctly … but there are so many variables, it’ll take time. If I can come up with some concrete conclusions (e.g. it’s definitely only about scale or definitely only about how the master is set) I’ll post here.

[thermionic]

I figure take it back to basics before you warp your existing song and get lost in those variables.

Some simple 4/4 patterns with a few unique trigs in each so that you can tell you’re following the right patterns in the chain.

[bibenu]

Alright, starting with the simplest possible thing (MC sending to MS):

Two patterns A01 and A02, 16 step pattern both machines scale set up as Mod PTN, Len 16, Chg 16. Also Chain A01-A01-A02-A02

Everything works

Change receiver (MS) pattern A01 to Mod TRK … immediately things are broken. Change A01 to A02 is one cycle (16 steps) late on the receiver. Though change A02 -> A01 (with A02 on the receiver still set to pattern) the changes are in sync.

Assuming that’s the only thing that went wrong (which is a big assumption) we have one big rule already:

DON’T use Mode TRK on the scale setup of the RECEIVER.

Glad I followed your advice and started simple @thermionic

[thermionic]

Hah! Excellent. I get precious about my patterns and often need to create disposable ones to trash when learning new tricks (or debugging.)

[bibenu]

Further Experiments

Experiment 2

Lengthening and shortening CHG on the receiver makes no difference

Experiment 3

Shortening length on receiver messes things up unless receiver len is a submultiple of sender len (e.g. sender 16, receiver 8)