Threxx11
commented
1 year ago It is also worth noting this bug occurs in 0.4.0.1 and was not introduced with 0.5.0.0.
Open Threxx11 opened 1 year ago
Threxx11
commented
1 year ago It is also worth noting this bug occurs in 0.4.0.1 and was not introduced with 0.5.0.0.
Gumball2415
commented
1 year ago this appears to be a bug regarding pattern loading, the driver variables tracking the current position of the song are correct. i might not be able to solve this if i don't figure out what's exactly going on.
Gumball2415
commented
5 months ago observation: it's almost like the frame advance doesn't happen until the end of pattern length
Gumball2415
commented
5 months ago the Dxx handling code in the NSF driver wasn't updated with the new command bytecode paradigm (implemented in #46), so it's scanning for 0CC bytecode instead of DNM bytecode
Gumball2415
commented
5 months ago https://github.com/Dn-Programming-Core-Management/Dn-FamiTracker/actions/runs/8067304836 can you confirm if this build fixes the issue?
I was unable to replicate this bug in a fresh module, and I am not sure what the context of the error is in the module where it occurs.
Either way, the bug can be observed by comparing the audio in the tracker to the attached NSF export. The patterns on the 2A03 pulse channels seem to break and do not actually skip to the desired point in the next frame. Both files can be found in the attached .zip file.
If any more information is required, it can be provided in confidence, but this song specifically is from a module I cannot share publicly so if the full module is required it will have to be shared privately.
broke.zip