I’ve been running this patch for the past few days. The MetaModule is receiving MIDI from the Oxi One plus CV from a couple of modules. The patch with those inputs uses about 64% of the Meta’s CPU at 48k/64. I can play with it for hours with no problem.
But I’ve been leaving the system running overnight, and both today and yesterday, the Meta has crashed overnight – the patch isn’t running, and when I press play it immediately indicates a CPU overage. So it seems like something is accumulating and not getting cleared when the patch overloads.
In case it’s worth mentioning, the Meta is also providing power to the Oxi over the USB cable.
Polyphonic Plaits.yml (21.8 KB)
I’ve frequently noticed the ‘OVER’ indication on the MetaModule display when I’ve left it on for a prolonged period, both when a patch is running and when not. Rebooting from the system menu resolves the issue.
Hmm… strange. I’ll run it overnight and see if there’s anything accumulating.
Is the Oxi running a sequence as well, or is it stopped? Just wondering if continuous MIDI events might be part of it…
What is your “Overrun Retries” setting?
If you try to play a different patch will it play?
I’m not sure what you mean, Dan. I’m seeing ‘OVER’ for a variety of patches running even with moderate CPU utilisation (68% or so) after leaving the MM on overnight even when the patch isn’t playing.
Yes, it’s sending in MIDI chords.
Sorry, what I meant was in regards to after you see “OVER” and the patch stops, you mentioned needing to restart. My question is if you try to play a different patch instead of restarting, does it immediately say “OVER” and stop for that patch too?
Also, same question as above: what is your Overrun Retries setting?
And, do you have MIDI connected and sending events (e.g. a sequencer that’s running overnight vs. a MIDI keyboard that’s not playing any notes when you’re not there)
@gabrielroth I left the patch running overnight here with an OXI running and didn’t get an OVERload, and patch is still running. I looked at the patch structure and it could be related to a bug I just fixed with MIDI poly voice allocation. I don’t know how it would cause an overload but it’s worth trying. v2.1.17 fixes that bug.
OK cool — I’m on the .99-poly branch right now so I’m not going to worry about it. I’ll let you know if it happens again. (But now that I can get polyphony from a single instance of Plaits I won’t have any need for that complicated patch!)
Thanks. Let me check what happens if I see OVER again. When it did happen, the patch wasn’t even playing and I had no MIDI connected. I’ve currently got retries set to 32.
Hi Dan. I’ve just had an OVER indication after leaving my rack on overnight (actually, several days). MM was loaded with four Resonator modules, each with DDLY and mixed into Plateau. Settings were 32k/512 and retries set to 32. The patch wasn’t actually playing, so no MIDI or CV input. MM wasn’t actually frozen, but UI navigation was very sluggish. Once I’d loaded another patch and started it playing, the sluggishness and OVER disappeared. I hope that helps.
Oh, ok, that makes it easy to test! Can I get the yml file?
Thanks. One more thing: was anything connected to the USB jack? And was there an SD inserted?
Oh, and what expanders are connected?
Nothing connected to USB. SD card was inserted and just the WiFi expander. Nothing connected to the jacks, either.
OK, I’ve been running this for a while, and no crashes… but, I thought of something: it might be the SD Card itself.
When I’ve experienced a bad SD card, the UI will get sluggish as the co-processor tries and retries to read the card. It might be that for some reason the SD card is acting up, and it only happens after it’s been turned on for a long time. Worth changing out the card or removing it when leaving it on overnight and see if it happens again. If it still happens, then the Wi-Fi expander would be the next thing to try.