EDIT: As long as a bunch of notes are detached using the Sxx command, EVERY note effect (including arpeggio, vibrato, pitch slide etc. excluding back to frame, halt etc.) will be cancelled.
EDIT2: I uploaded a Mega Man 1 boss battle demonstrative cover to test the ability of the new Sxx effect. Sxx makes life easier but in return needs a channel volume for each note and an instrument-made triangle drum.
DPCM editor bugs:
-When you select a part of sample from right to left and press delete, it deletes the last part of sample, not which you selected.
-If the sample is too short(after you deleted most of it) and you press play, Fami freezes
Bugs:
-Custom VRCVII instruments in loaded modules "freeze" (refuse to play) upon trying to modify them. The only way to solve this is to open another module WITHOUT saving, and then reopening it.
When custom instruments are modified, the four bars (attack, decay, sustain, release) on both sides are reset to 0, so to solve this the composer must remember the settings on those eight bars (four on modulator, four on carrier).
Hope this helps because I'm not sure whether the bars always reset.
After exporting my VRC7 cover of +33 by Ryuichi Sakamoto (just to see what garbled trash it would produce), and I've noticed (and tested out) that the only thing that exports correctly is channel 1 of the VRC7 track.
Everything else except that very channel is corrupted somehow. Here are some observations:
- VRC7 channels that aren't channel 1 plays G notes regardless of octave when G, A, and B notes are supposed to play and doesn't play C, D, E, and F notes when they are supposed to.
- 2A03 channels behave erratically based on what's inputted in the VRC7 channels. I tested this out by modifying TechEmporium's test1 file to not trigger VRC7 notes. Glitches still occur, however (2A03 triangle only plays and the low E note in that channel becomes a detuned high E when VRC7 notes aren't programmed).
Other bugs I've encountered that are related to FamiTracker:
- Undoing a note on the VRC7 channel with recently added rows causes those added rows to be filled with --- 00 0 --- on every channel.
Have you made the 3.5 final yet? It's been months since I've been here, but I AM looking forward to the final. My first try on 3.5 beta was Dreams Come True (Sonic the Hedgehog reference ^__^). This is shaping up to being a quite awesome super-tracker!
I love the implementation of the KONAMI VRC7 and the FDS. Coolness! I am itching to see the finished product. ^___________^
out of 5 so far...
_______________________
You now process Dracula's Rib. Good luck processing it...
Thanks for the response all. I've fixed a couple of bugs now (including VRC7 export) and will upload an update later.
ElHuesudoII: The FDS cannot reach high notes because of hardware limitations, and the note isn't clipped so it wraps around. One way to reach higher is to repeat the wave, each time will double the octave.
Completed the Super Mario 64 "Slider" to test the note release and the 1xx/2xx pitch slides on VRC7 channels.
It worked well, with the only problem that I don't feel like I'm tracking in 8-bit.
That's how I feel when I listen to anything that uses these chips in an NSF file. It's not a true NES song.. there are some very cool songs that push the limits of the original hardware though, without adding chips.. Hooray for limitations!
) 
