Probably no longer relevant, but I found a way to not have to double the buffer size.
circularBuffer2~.pd
circularBuffer2~-help.pd

Edit: Oh I see now (too late) that [list store] might be better to use than the get-prepend-set way that I'm reordering the buffer https://forum.pdpatchrepo.info/topic/13643/is-there-a-maximum-list-length-or-message-size

Just curious: say your new BPM is off by 0.001 and around 120. I think 0.001 is a lot bigger than anything you've been discussing, but I could be wrong. Anyway, doesn't that mean that over 10 minutes, your beat will be off by 1/100ths of a quarter note, and the pitch will have been off by less than 2/100ths of a cent? (I'm glad they didn't demand that kind of precision when I was in music school!)

@lead Sorry, still not quite understanding what you're after. Is the 77 BPM example bad because it has 4 decimal places instead of 3 like your good examples? Or are you asking "which BPMs when pitched down 2 semitones results in a BPM that has as few decimal places as possible?" It would be cool if you were asking https://math.stackexchange.com/questions/2438510/can-i-find-the-closest-rational-to-any-given-real-if-i-assume-that-the-denomina

@lead Yes. Here is what one of my plugins looks like on my Windows machine:

@alexandros He he he....I think David meant @seb-harmonik.ar. At least you were associated with a good idea

@high_output-5000 said:

@jameslo That sounds interesting! How do you set up the clone architecture?

Maybe like this? https://forum.pdpatchrepo.info/topic/13500/signalrate-matrix-mixer Let me know if this is not right (and please explain how it's not right) and I'll see if there's a way to make it meet your needs if you can't see it before me. Definitely check out @oid's suggestion too.

@high_output-5000 I think I've done stuff like this using nested [clone]s, e.g. the inner clones each represent a row for a particular column, and the outer clones represent the columns. The outer clones pass their column identifier to the inner clones. I'm just handwaving due to cocktail hour...let me know if you need more detail and I'll get on it!

[bang~} bangs after each DSP block, but that's also when other control processing happens. If a GUI object generates a bang, does it run before or after [bang~] in that particular control processing block?

I think it's before, and I'm guessing that [bang~] is the last control message processed before the next DSP block is processed. Check out this test:
bang~ order.pd

I always get 0 when I click on the bang at the top. If [bang~] ran first, then its bang would clock the [timer] 1 DSP block later.

Is my test and conclusion correct?

The vanilla objects can't capture the raw MIDI bytes, right? By the time they come out as outlet messages, the MIDI messages have already been parsed. For instance, many of the MIDI objects have to be filling in the running status or else the channel info would not be output.

Edit: Ooops, I'm wrong! Just read the help for [midiin]. Move along...nothing to see here....

@pajzd You can spread the file loading over several audio ticks so that audio output isn't disrupted, see https://forum.pdpatchrepo.info/topic/13759/soundfiler-issue/7