Re: [X360] Star Wars Battlefront III pre-alpha (*.rax)
Posted: Sat Mar 11, 2017 10:45 am
For the first submesh, yes, I'd say so.AceWell wrote: face indices (33858) * 2 = 67716 - 63096 (0xf678) = 4620 /2
so the last mat group would be assigned to the last 2310 faces out of the submesh total faces
(But I've made a stupid error with absolute/relative face indices, some + is required.
Submesh 6 should have 7 groups.)
well, it's not a simple '+', those crazy dev bastards might have used the same principles for building groups as the ones from FH3?
(Gosh, I really hope I don't have to work with median face indices and do a quicksorting which was a bad idea for early FH3 extractor versions.)
see this is so annoying:
f 2724/2724 2725/2725 2726/2726
f 2725/2725 2727/2727 2726/2726
f 742/742 744/744 743/743
f 741/741 742/742 743/743
f 743/743 752/752 741/741
f 750/750 741/741 752/752
f 750/750 752/752 751/751
f 749/749 750/750 751/751
f 745/745 746/746 747/747
f 748/748 747/747 746/746
f 747/747 748/748 741/741
f 750/750 747/747 741/741
f 2720/2720 2721/2721 2722/2722
f 2721/2721 2723/2723 2722/2722
f 2723/2723 2713/2713 2722/2722
Simple solutions looks good for some submesh but misses too many groups (reason see at end of post): (seems I'll have to do a "per submesh" sorting at least)
edit: I did that sorting, it looked ok for the first submesh but not for the rest.
For the unsorted thingie it's weird:
There's this maximum faceindex, # 12. maxFI 27928 (might be wrong calculated, not sure)
and there's the values from the "FACE " tables:
# 27324
# ----------
# 27519 28186 < which don't make too much sense down from here
# ----------
# 28189
# ----------
# 28284 28725
# ----------
# 28755
# ----------
# 28764
# ----------
# 29178
# ----------
# 0