I understand that control rate events may not align consistently with audio rate events, but for a given control rate event with no explicit intervening delays, can I assume that all the processing and audio side-effects of that event will occur at the same point in the audio stream? For example, suppose a bang causes a [phasor~] to reset its phase to zero, changes that [phasor~]'s frequency, and updates the multiplier of a downstream [*~]. Would all three occur at the same sample? What if the bang was sequenced by a [trigger]?

@jameslo The whole chain has to complete when a control rate event is triggered.
It all happens at the same time in the time between audio ticks but in the order set by any triggers (or the creation order if no trigger was "put".
I cannot explain it better than chapter 2.5 here...... .http://puredata.info/docs/manuals/pd/x2.htm
David.

@whale-av thanks, I wasn't sure if that's what the section on scheduling was saying. What is an audio "tick"? A single sample? A single audio block that's processed?

@jameslo one tick is 64 samples in Pd..
Actually this is a much more detailed and understandable explanation......chapter 2.0........ http://pdpatchrepo.info/hurleur/multiprocessing.pdf
David.