Anyone notice a 12 Note Event Limit on a single Track?

Apologies if this is premature before extensive testing, but I’m finding I can’t seem to have more than 12 Note Events happen on the same tick. If I add more than 12, only the highest 12 Note Events are sent.

Saw this as I was watching the destination device (DMXIS - so I can see a graphical fader). So then I loaded up a MIDI Monitor mid stream and verified. And then used the onboard Pyramid MIDI Monitor and it seems, unless I’m smoking crack, that 12 Note Events happening at the same time is a limit. Add more and only the higher valued Note Events happen (higher MIDI Note #).

Remove the upper range ones, and the lower numbered Note Events happen.

Am I missing a troubleshooting step and/or has anyone else experienced this and submitted this before I send off a msg to Squarp Admin?

Edit to Add: It’s not 12 Notes exiting the Pyramid - it’s 12 Note Events entered onto the same step. I can, for example, use the Harmonizer to send out more than 12 notes (I think I have 16 happening right now with no problem). Weird.

Hmmm. I wonder if it’s related to tempo and the time it takes to send each message.

I’ll spare you the math, but sending out 12 note-on messages takes up a sizeable portion of the time of a 1/64 note in 4/4 at 120 BPM. It’s seems plausible that it’s just a limit on how fast data bits can be sent at the standard MIDI rate of 31250 bits/second.

Try this: slow the tempo way down and see if you get more notes per step. Alternatively speed the tempo up and see if you get fewer.

So your saying it’s input not output.

do you think the limit is on track storage?
Can you enter more than 12 notes onto a step in the pyramid itself ?
Does resolution change this … eg is it only 12 at highest resolution?

Not sure I could test this without writing some code, given my lack of fingers ( only 10!)

There’s always the Zappa trick :sweat_smile:

1 Like

Thanks for your response.
12 notes on the same beat shouldnt be strange - the Pyramid can do it if the notes are in different Tracks.

I mean: kick, a closed hh, a bass, a 3 note pad chord, a lead - they can all happen right on the same beat in extremely mundane situations and thats what, 8 Note On events on the same beat?

It may have to do with the serial nature of MIDI, but it usually takes a LOT more data to overwhelm devices IMO. And I have overwhelmed pretty much every MIDI device Ive owned because Im really weird (read: insane, bent, psycho, etc) that way.

But if youre sequencing a 12 voice drum machine with one Pyramid Track and kick is your lowest note, it will get lost if you put too many other Events and it will drive you crazy trying to figure out why if you had not read this. :smiley:

MIDI is something like 3125 Bytes/sec. Numbers numbers math math math - 140bpm, blah blah blah. A Note or CC Event is 3 bytes plus i think a few other status bytes, plus the F8 clock (which i had turned off at this time).

Plus Im using the USB out and something something faster something parallel i think.

Im going to try that test when i wake up - apologies this post is premature. Im excited for a troubleshooting day.

Thanks your response.

It’s output.
Storage: you mean the 10k limit of events? I’m still on a tester project (read: very little data) and the way my system is configured the actual MIDI data is minimal. I can dupe this on a brand new project of precisely 13 Events entered, so it’s unlikely storage.

I can enter the data, but it doesnt play, evidenced at receiving unit (MIDI to DMX system with a destination of several channels on two washes).
Resolution makes no difference that Ive noted.

This is tougher to test with audio, so you might want to use a MIDI monitor, either the on board one or something like MIDI Ox. I’m pgmg lights using a direct MIDI to DMX translation, so if a light doesnt get the msg, it doesnt turn on/off which was the “wtf is going on here?” moment that started this.

I can dupe this with a brand new project.
Im on PyraOS 3.23.

No, i meant are you trying to record these to a track.

As I said, if you manually program 13 notes on the pyramid - do these output correctly?

Apologies.
Yes, i am programming these manually in Step edit and these are the notes that are outputting incorrectly/not as expected.

Although my music sounds like I used the Zappa method extensively…

I can save you some time. I tried it on my Pyramid, saw the 12 note limit you saw, and tempo made no difference. I also tried doing it across 2 tracks and could only get 24 notes. So, yeah, it looks like there’s a limit of 12 notes per step per track.

Well, yes, but having all that on the same MIDI channel seems a bit unusual. Then again, if you’re really weird (insane, bent, psycho, etc.)… :astonished::smiley:

Context: you were pointing out the speed of MIDI data possibly suggesting 12 notes simultaneously would be creating this seeming overload situatuon…so… :wink:

I have a workaround, but it’s just seems like an odd thing Ive never run into before. I’m trying to find some of the posts about people saying theyre missing notes on a “1” of a sequence, windering if it might be related, but my search-fu is currently nerf’d by fat fingers + cell phone.

Thanks for testing the tempo.

Correct. That hypothesis has been proven false. It’s not due to the time to send the notes.

Here’s weird: Shortly after i noticed this after about two days of wondering what the heck was going on, one of Tracks on the Project stopped functioning as expected so I suspected a corrupted Project File.

Except I could dupe the 12 Note thing on a new Project. So, kind of need others to help verify if the problem is system or data or…whut. heh

Thank you for your input.
Seriously appreciate it.

yeah, so i can see the same - max 12 notes on one step that is exactly the same.
however, even a 16% offset on 1/64th starts a ‘new’ set of 12 - that offset is a 1/400th note… (actually, i bet its 1/512th :wink:

interestingly, if you quantize these offset notes at 1/96th (smallest quantization unit), the all notes do go out still …

this seems in line with what you found with the harmonizer, its not a limitation on notes out, just how many notes are processed on each ‘step’ of the internal grid.

perhaps its due to the internal representation Squarp use, or perhaps a simple way to limit how much processing is done per step (as each note has to go thru the fx chain, which in turn can generate even more events)

i guess perhaps a limitation for black midi, or if you are automating control - but not really for playing.

I wonder if there are limits on other events too…

1 Like

confirmed, happens the same way even at 999.9bpm and 10bpm

if the start of one of the notes is shifted even by 1% it doesn’t get counted in the limit and is played correctly even though it’s on the same “step”

1 Like

actually just checked the specs, so recording resolution is 96ppqn, so that tallies with the 1/400th i mentioned above (so i was probably wrong about 1/512 :wink: ) - so id assume that is the ‘processing’ chunk that is used.
so yeah, a limited of 12 notes (perhaps an event limit too e.g. cc) on that resolution

it seems that CCs do get sent out correctly and are not counted in the note limit

It feels like y’all did my homework for me. LOL

Thank you for the incredible insights.

My application is definitely a control automation situation (MIDI to DMX). Luckily i have the option to pass this data through an Event Processor to resolve my immediate needs.

Wow, great info. Totes bookmarking this stuff.
Thank you!!

Also means Im not crazy.
Well, in this regard only. :smile:

1 Like