<- 1  -   of 75 ->
^^
vv
List results:
Search options:
Use \ before commas in usernames
Edit history:
Antidote: 2014-09-10 01:46:33 am
A bit of progress on AGSC, still working on figuring it out, but we're getting there!
Code:
//--------------------------------------
//--- 010 Editor v5.0 Binary Template
//
// File:
// Author:
// Revision:
// Purpose:
//--------------------------------------

BigEndian();

struct Entry
{
    uint EntrySize;
    char EntryData[EntrySize];
};

struct SoundInfo
{
    SetBackColor(cLtGreen);
    uint unk1;
    uchar unk[unk1-4];
};

union SoundTable
{
    Entry entry;
    struct
    {
        local int __dataStart = FTell();
        uint DataLength;
        while(FTell() - __dataStart < DataLength)
        {
            struct
            {
                while(ReadInt(FTell()) != -1)
                {
                    SoundInfo soundInfo;
                }
                uint terminator;
            }soundGroup;
        }
    }soundTable;
};

struct
{
    string ModuleDir;
    string ModuleName;
    SoundTable  soundTable;
    Entry  loopPoints; // Not entirely certain
    Entry  adpcmData;
    Entry  offsets;
}file;
nice

SetBackColor(cLtGreen); ???
Quote from Paraxade:
Unrelatedly, I went poking through a couple of the demo's files and found what seems to be another unused morph ball.


Can you please hack this into the game 😍
Edit history:
Antidote: 2014-09-10 05:04:35 pm
DJ, it's to highlight use bytes, that way I can tell how much progress I'm making.

Edit:
I'm currently hunting down PAS4 (Otherwise known as CPASAnimState, and CPASDatabase), and Skippyjr was right, it's a rats nest of virtual functions, and pointer tables. It might take a while to figure out where the actual data handling is performed.
Speaking of that, Skippy, if you happen to be lurking around, have you managed to make any progress on that?
yeah somebody else needs to do that so I don't have to touch it, it sounds fucking painful
I just know I'm having absolutely no luck figuring it out by analyzing the data itself, lol. There's all these random numbers of varying type, and the PAS4 section is rarely ever the same size between different ANCS files, and I have absolutely no idea what's supposed to indicate what data goes where. I wouldn't even care if there was some value somewhere that would just let us skip over it, but nope! Thanks Retro :P
Edit history:
DJGrenola: 2014-09-10 07:58:44 pm
how could we do it

perhaps searching RAM dumps for chunks of them might work like I tried for ANCS AGSC ... they're only small files so you'd have to keep the chunks really short but the PAS4 section might be copied into RAM separately
Edit history:
DJGrenola: 2014-09-10 08:05:56 pm
then at least you'd have a means of establishing how long it is

or wait, you know where the PAS4 bit starts, right? so yeah you could take the first 8 bytes or something of the PAS4 part and check RAM dumps for that sequence, then walk from there until the bytes stop matching and you know how long that particular PAS4 section is
Well, the thing that always immediately follows PAS4 is an array of PART file IDs, so I know how to tell how long it is. The problem is it's not always the same length. There's also some value at the beginning that seems to be some sort of count value, and it directly corresponds to the size of the section, but even between multiple PAS4 sections with the same count value, it's not the same length.
just put a big hardcoded list in your software of all the PAS4 lengths for all known ANCS files >_>
Edit history:
Aruki: 2014-09-10 08:29:08 pm
sounds reasonable!! And I guess if anyone ever wants to make a custom ANCS, fuck you
well it sounds like finding all the lengths could be automated, and it sucks but it would be temporarily better than no support at all
oh haha, I thought you were joking. It's not really so important to have support for it right this second that that's really necessary IMO; I'd rather just have no support at all until it's figured out properly. At this stage the only real disadvantage is that stuff looks weird in the editor, which is annoying but not an incredibly important problem. It's gonna need to be figured out eventually, though, especially if we want to crack animations.
Edit history:
Aruki: 2014-09-10 11:04:52 pm
Made a video to show off the Dark/Light Suits. I even threw in Zero Suit Samus at the end, who... didn't quite turn out like I hoped.

haha

is that her model from prime 3?
I suggest not beating Echoes with over 75% again, since you've managed to block out the memories.
No, it's from the ending of Echoes.
Quote from MilesSMB:
I suggest not beating Echoes with over 75% again, since you've managed to block out the memories.


haha

yes, I remember now
Not impossible
just highly unlikely
Samus needs a chiropractor I think.
i was trying to decide if the arms or the hair was funnier, until her hand got stuck in the air.
Edit history:
Aruki: 2014-09-11 12:32:04 am
I don't even know how she got screwed up that bad lmao, I didn't think I did -that- badly converting the skin
so, what's difficult about converting suitless samus vs. the suits?
Edit history:
Aruki: 2014-09-11 12:40:00 am
Well, I was replacing an actual suit model with a no-suit model, so there's a lot more differences between the skeletons than there are when just converting between the suits. For example, ZSS has a bunch of bones on her right hand - a wrist and three bones on each finger. The original model, on the other hand, doesn't actually have a right hand and just has one bone called R_wrist_orient for orienting the arm cannon. I tried to rig all the right hand bones to R_wrist_orient, but apparently either that wasn't good enough or I screwed something up. :P

Another obvious example of a difference between the two is the shoulders. The suit's shoulder are higher up than ZSS's, so her shoulders got messed up pretty bad when rigged to the suit's shoulder bones.
and i guess editing the skeleton would be too difficult.