Thanks for the thorough reply! I was going to go through the steps you outlined to get the verbose stacktraces, but first tried upping the block size to 256, as you suggested, to find that it fixes both the popping and ALL of the error messages. I wish I would've thought to try this earlier to save you any trouble my not thinking of that has caused.
anyways, I've been tinkering for a bit to try to establish the connectivity with both the manta and the manta mate on power up, without using the IDE to initialize the project. I've just managed to do this by using the two-button long press (the one that moves through the loop), to just reinitialize the project, which seems to be working well, establishing connectivity.
Interestingly, when using [toggleWithLed] to make the connections (I just did a quick [sel 1] to "connect" message as well as to the "hw 0 0 0" message), I manage to reconnect to the manta, but not to the manta mate (the "hw" message doesn't seem to get the desired result in this case. Reinitializing the project does the job, but that seems strange to me, since it's just using a loadbang to the same message that the sel 1 was clicking. . .
One other thing, my first attempt at getting the connectivity was to place the patch in a loop, thinking that toggling through the loop would do the job, which it did. The only problem was that lots of crackling ensued, even when the loop had only 2 patches, both of which were set at 256 block size. Perhaps the loop itself was resetting things to the default 16 block size? The solution I have now (not using the loop but just reinitializing the patch after startup), works well, but it would be nice to be able to use the loop function if I can
Jeff got back to me and says he'll investigate for lock, IO, or other sysrecall. He said the object was made by Spencer Russell a long time ago, so Jeff will see if he can figure it out. He says there's a fair chance that there's something in the object
It's interesting that all the error messages went away, though. The manta controller does send a very high resolution of data, could it be that the rate of messages within the object was causing troubles in the low block size?
Also, do you think I should go ahead and run the debugging procedure still? I'm happy to do it, and would be interested to know what's going on. Just letting you know that I'm having a happy result with the block size fixing the main issues🙂
In addition to this letting me run my patches, it's really cool that the manta mate effectively functions as another expander for the Salt, adding another 12 cv or trigger outs through the usb connection!