ObviousInRetrospect

Still getting basically 100% corruption in SLEEP_MODE_STANDBY with the new twi.c. Tried deleting twi.c in the middle and verified I got a bunch of linker errors (until I dropped the...

mind sharing the test code you have working? I wonder if it is related to the 32-byte chunking I am doing

In my testing it didn't resolve the problem I am seeing. It seems to work similarly to the prior version. It probably fixed at least part of the problem in...

so fwiw it is possible that doing a single write pattern (filling the modified-size-buffer) instead of sequential reads and writes of different is masking the problem. using my code with...

[edit: not sure what happened in the testing from the prior version of this post]. Looks like it is working. standby: ``` CRC valid fail: 0 pass: 1310 000102030405060708090A0B0C0D0E0F 101112131415161718191A1B1C1D1E1F...

Did more testing. For these tests I sped up the reads to read at 10hz instead of 0.2hz. This is with test pattern data for which I took out the...

so first off, THANK YOU, this is hugely better. The failure rate is down by a factor of at least 2500x. Testing in prior post and this post was conducted...

running the master at 10x the speed of the slave (10mhz/1mhz), my perception is the crash rate goes up: (this is still 16 chunk read on the master and 32...

And back to sleep_mode_standby, same slow (1mhz) slave fast (10mhz) master, no errors in 65k operations!!! that means this is totally usable in standby. code version: https://github.com/ObviousInRetrospect/DualMode/commit/5f6f3e75b4ad620d8185442521d569b4a2592c6e: ``` CRC valid...

@MX682X got it sorted, I'm just testing. Now to write an example that turns a t412 into an i2c dac while making correct use of sleep. @SpenceKonde this fix should...