Any plans to test it for FW 1.70

bsp2 / libanalogrytm

Portable Sysex Library for the Elektron Analog Rytm Drum Computer

MIT License

55 stars 9 forks source link

Any plans to test it for FW 1.70 #4

Open alisomay opened 11 months ago

alisomay commented 11 months ago

If you'd share what to look for during testing and how you approach reverse engineering briefly I'm up for testing and contributing also ✨

bsp2 commented 11 months ago

just noticed that master_length can (well, of course) exceed 64 (and 255), so I changed the type to s_u16_t (clean-up). See pattern.h

bsp2 commented 11 months ago

in case you're wondering why not all u16 fields are s_u16_t: I simply like to keep track of what fields actually use the MSBs (and which ones could be removed since they are always 0)

alisomay commented 11 months ago

About _unknown_332c I have an idea.

It is a singled out value, it is not part of anything. (I also checked if it increases if I query patterns from other banks.)
It is the last byte in the pattern structure.
Empirically it is always 0x01. I couldn't find a way to change that.
We can represent 256 different values with it.

I feel like we've exhausted the data we save with a pattern here which leads me to speculate about

It may be a marker byte
Maybe there are some internal cases where people use this byte to mark the end of a pattern structure if they put more than one pattern together (I'm not sure)

I'd say either we leave it as _unknown_332c or name it EOP (End of pattern or similar)

What do you think?

bsp2 commented 11 months ago

as long as we don't know for sure, let's leave it at __unknown332C. Could simply be a flag (true or false). Maybe something that cannot be set to false (anymore) ? shrug

(another wild idea: could have been an external tempo sync flag, at least it's right next to the BPM setting)

alisomay commented 11 months ago

Hmm.. I can give it a go for testing. I mean that is not a data which is saved with the pattern but also represents a part of the state of the pattern. Can we decide syncing on pattern level? I thought that ought to be a global only setting.

alisomay commented 11 months ago

Nope it doesn't change with midi sync. I'm shifting my interest to the rest of the unknowns now (if there are some) but I think we can say that pattern is complete 🥳

bsp2 commented 11 months ago

tried that too (change midi sync) but it's a global setting, of course. maybe it once wasn't. who knows.

just updated sound.h (adds accent_level). another mystery solved :-)