AliceLR
commented
3 years ago Modules that don't use any of these features
| Module | Uses | Still works? |
|---|---|---|
| - unknown/anvil.mmd0 | 8 ch. mode | Yes |
| - unknown/wizardry.mmd0 | Yes | |
| Jason Honingford/trend.mmd0 | 8 ch. mode | "Yes" (order 11- tone portamento is not strong enough?) |
| Orpheus/electrosound 64.mmd0 | Yes | |
| Teijo Kinnunen/kinnunen 2.mmd0 | 8 ch. mode | Yes |
| - unknown/alien breed ii.mmd1 | Yes | |
| - unknown/fair.mmd1 | Yes | |
| Eivind Segrov/nighttrain.mmd1 | Yes | |
| Mixer/superaxel.mmd1 | 8 ch. mode | Yes |
| Toady/secret of mana.mmd1 | 8 ch. mode | Yes |
High octaves
| Module | Uses | Description | Fixed? |
|---|---|---|---|
| - unknown/death.mmd0 | Oct4-7 | Expects C#3 and C#4 to be the same note. (ord. 0) | Yes |
| soul crystal - cyberfunk.mmd0 | Oct4-7 | Expects C#3/C#5, D#3/D#5 to be the same. (ord. 0) | Yes |
| you got to let the music.mmd1 | Oct4-7; 8 ch. | Expects C-4 → C-3 in 8 ch. mode. (ord. 0) | Almost[1] |
| Barney/did.mmd1 | Oct4-7; 6 ch. | Expects *-4 → *-3 in 8 ch. mode. (ord. 11) | Yes |
| Blockhead/childplay.mmd1 | Oct8-A | Expects octave 9 to wrap to octave 0. (ord. 1) | Yes |
| Lynx 2/passo-doble.mmd1 | Oct8-A | Expects octave 9 to wrap to octave 0. (ord. 0) | Yes |
| Orpheus/hypnotised.mmd1 | Oct4-7; 8 ch. | Expects D-6 → D-3 in 8 ch. mode. (ord. 5) | Yes |
| Robert Moseley/maria.mmd1 | Oct8-A; 8 ch. | Expects *-9/A → *-0/1 in 8 ch. mode. (ord 1) | Yes |
[1]Timing or tuning bug(?) prevents various samples from synchronizing correctly but it's much closer than before. The issue here might be the timing table for 8-channel mode I copied from libxmp since both xmp and OpenMPT have this bug (that's what I get for copying someone else's homework I guess...). Notably 'maria.mmd1' also seems slightly faster than it should be in MikMod/xmp. See also: https://bugs.openmpt.org/view.php?id=1386.
Song transpose
| Module | Description | Fixed? |
|---|---|---|
| Keith Tinman/hook - ingame 2.mmd0 | Transpose -2. | Yes |
| - unknown/1000 sterne.mmd1 | Transpose +6. | Yes |
| Alexander Zutt/phasing.mmd1 | Transpose +1. | Yes |
Instrument transpose
| Module | Uses | Description | Fixed? |
|---|---|---|---|
| Daniel H. Dean/dissociate.mmd0 | Oct4-7; 8 ch. | Transposes oct4 notes down into oct3. (ord. 20-23) | Yes |
| Reverb/alain.mmd0 | Many instruments w/ different transpose. (ord. 0) | Yes | |
| - unknown/digi-t-ice.mmd1 | Inst. 5 and 6 transposed +3. (ord. 2) | Yes | |
| - unknown/gotterfunken.mmd1 | Oct4-7 | Inst. 1 transposed -2 (ord. 0) | Yes |
Both transpose types
| Module | Description | Fixed? |
|---|---|---|
| Dean Evans/getto.mmd0 | STr: +1. ITr: 3 (+3), 7 (+3), 8 (+15). (ord. 0, 2) | Yes |
| - unknown/stop chit !.mmd1 | STr: +2 (ord. 0). ITr: A(-3, ord. 2), J(-8, ord. 14) | Almost[1] |
| Dean Evans/jurassic park - ingame 1.mmd1 | STr: -1. ITr: 2 (-7, ord. 0), 6 (+10, ord. 2) | Yes |
[1]See above. This track is also 8-channel mode.
Transpose + wrapping
| Module | Uses | Description | Fixed? |
|---|---|---|---|
| - unknown/transition.mmd0 | Oct4-7; song tr. | STr: +1. Wrap: E/F-4 @ ord. 4. | Yes |
| HJE/very short!.mmd0 | Oct4-7; song tr; inst. tr. | STr: -1. ITr: 5 (-18). Wrap: in bass. | Yes |
| - unknown/three phantoms.mmd1 | Oct4-7; inst. tr. | ITr: 6 (+1), 8 (+3), 9 (+3), H (+3), J (+2). Relies on transpose+octave wrapping for inst. J to sound correct. (ord. 0, 3) | Yes |
| Alan Mackey/megaball - title.mmd1 | Oct8-A; inst. tr. | ITr: N (-6, ord. 2/3), O (+5, ord. 3). Wrap: ord. 15. | Yes[2] |
| Digital Zone/u got 2.mmd1 | Oct8-A; inst. tr. | Wrap: ord. 8. | Yes |
| JaDe/fineline.mmd1 | Oct4-7; Oct8-A; inst. tr. | ITr: 2(-5), 3(+9) (ord. 0), Wrap (ord. 1) | Yes[2] |
| JaDe/the chase.mmd1 | Oct4-7; Oct8-A; inst. tr. | Wrap: (ord. 3 & 4) | Yes |
[2]: tone portamento seems weird, similar to 'trend.mmd0'.
I can look into the two other issues here some other time (particularly the 8-channel mode timing table since it affects other players too).
Several behaviors of OctaMED affecting note translation do not currently work:
The fix for all of these is the same: the OctaMED loader should allocate instruments and implement these features with the tuning table. This is a fairly easy fix (and this will be required to implement the IFFOCT instruments down the road anyway).
Tentative patch here. It is not well-tested yet, so expect some table spam here as I do that over the next few days. https://github.com/sezero/mikmod/compare/master...AliceLR:med-transpose-octave-wrap