michelez

I have been experiencing issues with seq~. i have several of those working with the same phasor signal feed. If I store events at the same sync point, say 0.5, I would expect to get synchronous events. Yet this is not the case. There is a significant mismatch between supposedly synchronous events.

I am wondering if there are issues with the object itself. Even decreasing the vector size does not really improve the situation much.

I would appreciate any feedback on this, there are not many topics related to that object which makes me suspicious…

I have been experiencing issues with seq~. i have several of those working with the same phasor signal feed. If I store events at the same sync point, say 0.5, I would expect to get synchronous events. Yet this is not the case. There is a significant mismatch between supposedly synchronous events.
I am wondering if there are issues with the object itself. Even decreasing the vector size does not really improve the situation much.

synchronizing-with-seq

This is one of those (extremely frequent) cases where posting a simple patch that clearly demonstrates your problem would be helpful.

On my computer I get discrepancies of a fraction of a millisecond sometimes, but not so much as to be generally problematic in the event-based Max world. Those small discrepancies do mean, though, that I wouldn't rely on the message order of multiple seq~ objects outputting ostensibly simultaneous events.

Thank you for the reply. The problem with posting the patch is that it is a large project: it only starts producing discrepancies when a large number of messages are output simultaneously. I am now running the seq~s in a another top level patch which allows multithreading and things are working a bit better.

In general I am glad to hear that, aside from small discrepancies, the object is solid (I was beginning to doubt it). I will try to extrapolate the relevant parts of the project and post it, but I think you pretty much answered my question, and no, the order of messages is not a must, I am just looking for rhythmic accuracy.

Thank you for the reply. The problem with posting the patch is that it is a large project: it only starts producing discrepancies when a large number of messages are output simultaneously.  I am now running the seq~s in a another top level patch which allows multithreading and things are working a bit better.
In general I am glad to hear that, aside from small discrepancies, the object is solid (I was beginning to doubt it). I will try to extrapolate the relevant parts of the project and post it, but I think you pretty much answered my question, and no, the order of messages is not a must, I am just looking for rhythmic accuracy.

Synchronizing with seq~