Initial D Extreme Stage [PS3]

[skys215]

I guess this is the main body of AE86L.
AE86L_Body_genuine_00.efo this is just one of the extracted file for AE86L.
https://www.dropbox.com/s/c2o60uko4873e ... 0.efo?dl=0

And there's another file called texture.efo, which I believe it combines all parts into a car.
https://www.dropbox.com/s/fdkklq9le0nsz ... e.efo?dl=0

Both file was under the AE86L folder.

Unfortunately the files with YS header have compression, I already asked for help with from Aluigy but he could not understand the method of file compression.

Thanks anyway.
The next thing I want to "crack" is the path of the course. If that is possible, then I can make a course from real life(though it can't be so accurate as reality).
Are you interested?

[zoloat]

I guess this is the main body of AE86L.
AE86L_Body_genuine_00.efo this is just one of the extracted file for AE86L.
https://www.dropbox.com/s/c2o60uko4873e ... 0.efo?dl=0

And there's another file called texture.efo, which I believe it combines all parts into a car.
https://www.dropbox.com/s/fdkklq9le0nsz ... e.efo?dl=0

Both file was under the AE86L folder.

Unfortunately the files with YS header have compression, I already asked for help with from Aluigy but he could not understand the method of file compression.

Thanks anyway.
The next thing I want to "crack" is the path of the course. If that is possible, then I can make a course from real life(though it can't be so accurate as reality).
Are you interested?

I'm also interested in courses. But I'm a tech noob, I know nothing about coding. But if you interested, I want to help as anyway as I can. Such as emulator tests, finance...(I'm not rich, though)

[skys215]

Hi guys, I have not been here for a long time, there was a problem with my hardware iMac ))
I tried to write a script on Noesis, and I have some success, but the script contains a lot of mistakes ...
This script works with InitialD Extreme Stage/ InitialD Arcade Stage 6AA / InitialD Arcade Stage 7AAX and InitialD Arcade Stage 8 Infinity with *.efo containers that contain meshes and textures.

At the moment the script does not work correctly, it reads not the hex offset in the files but it looks for the content labels
I could not fully understand the structure. * .efo, who knows how to handle hex, I will be glad to any help and advice in search of the necessary displacements and decoding of the structure.

Currently, the script reads only files containing single models (not group models)

several primitive models for tests here:
https://www.dropbox.com/s/geejzobp9m6yw ... 6.zip?dl=0

example structure cube.efo
I was not able to find the offset at the beginning of the file to these blocks, I can not understand where they can be in * .efo (((

I can only open the cube.efo with this script.
The other files in the package which has YABX header is still not openable.

[gmodoza]

Hey if anyone can get me takumi's 3d model and textures (of the character) id be more than happy to pay for it

[blinky5788]

Hi guys,

I too am after images, not the models but the actual files to convert id6/7/8 to English.

I have done some research on the .xaf file type used in the game (based on ACROBATA_PATCH.xaf (ID8))
(Linux) (i know it's not a linux game nor am i trying to make it one.)

maybe someone can find a use here,

and the strings in the header are :
https://pastebin.com/fUY6Lq4L (too many to list here)

EDIT, found some more interesting things,

there is a Id5 dump on the net (linux game) contains a file called xafMaker.exe, there are dll files with it as well, but each game has newer versions and the exe does not work with those, anyway, extracting the XAF, gives, .abd .abr. and a huge amount of .efo

the dlls also point to some sort of SDK called "Tea SDK" and that's where the yabukita object is referenced from, and all the other references, there is heaps. each file is signed by who compiled it and has a email address, i won't post it here and i won't contact him as i doubt very much he will give me the sdk or any info on it.

here is some infor on the xaf with the program executed:
from initial D5
ACROBATA.xaf
--- xaf volume information [ACROBATA.xaf] ---
title : InitialD ArcadeStage 5
comment : (c) SEGA CORPORATION, 2008.
format type : xaf0
major version: 1
minor version: 3
sector size : 2048 bytes
total volumes: 1
total records: 1159
total dirs : 23
total files : 1136
header size : 112640 bytes ( 55 sectors)
body size : 389384192 bytes ( 190129 sectors)
total size : 389496832 bytes ( 190184 sectors)
https://pastebin.com/P8YmQY5M

[Band]

So, anyone had any success in unpacking/packing textures since the last message? To see ID8 in english, mmm...

[Neitancrost]

I had success on extracting textures and models manually from Arcade Stage 4, with those I've made some Playermodels for Gmod, like this one: https://steamcommunity.com/sharedfiles/ ... 1694762111
To extract the models the way I did, you will need:
-QuickBMS and a .xaf script to decompress .xaf files (you can find both here: https://aluigi.altervista.org/quickbms.htm)
-X-Ripper program to extract textures (http://www.zeus-software.com/downloads/xripper)
-3D Model Researcher to get the mesh (I recommend this tutorial to understand the program: http://lazov.ru/mr/tutorial.php)

.If you extract the xaf file from Arcade Stage 4, it you give you a bunch of files.The useful files are: CHARACTERNAME.efo and CHARACTERNAME_texture.efo(this is the file to use xripper)
.In order to get the mesh, you will need to put these values on Model Researcher: [Vertices(Padding-24)][Faces(Type-short;Format-Tri Strip)]
.What I found is that the vertices content is divided in three blocks: 12 bytes of mesh/12 bytes of a "3D UVMap"/12 bytes of a sphere , and it repeats in the entire file

Hope that helps.

[Undividual]

Neitancrost, thank you for sharing this information with us, however I am a noob when it comes to extracting game models, especially with the hex editor. I was able to get all of the software to work and open the file for the model I wanted, but I can't figure out where the data for the vertices and faces begins. I've viewed a couple of tutorials on this and most say to look for the large buffer or area without any text, and there does seem to be buffers here and there but none large enough for me to tell whether it is separate data or not. The model I am specifically trying to get is FC3S_FBumper_amemya_03.efo found in CARS.xaf.

[Neitancrost]

So, I tried to extract some car models and found that their vertices are sometimes separate by a "desciption", wich is a problem for Model Researcher, so you can have some problems with cars. But I'll give some tips to extract it.(I'll exemplify with FC3S_FBumper_amemya_02)
First, you will need the vertices, that can be found on the "dense" part of the file.
To find what to put in Offset, copy the "coordinate" of the byte that seems to be the byte you want and paste as DEC, after it, check in the text tab if it is a mess, if so, just add 1 to the number you put in offset; and the value to put on Count, just put a value that end at the start of a new "description"
Then, to find the faces, go to the part that have a sequence of _ _ 00 _ _ 00 _ _, and follow the same steps made with vertices.
You are going to have a lot of faces problems, like messages of face erros because some vertices are "missing", and some other faces are going to be clearly wrong, wich can be fixed in a 3d program, like blender. Good luck with it.

hex2.png

[Neitancrost]

Hi everyone, I am trying to extract the content of map files, but they all have YS header. What I found was that IDAS8 has some compressed (YS) files that are the same as IDAS4 uncompressed (YABX) files. So with this is it possible to create a script to extract all YS files?

[guki]

hey there,

so i am currently looking at the arcade version of initial d stage zero because of the top secret supra v12. as far as i can tell the efo files are still the same structure compared to the files already posted here from previous games

http://www.mediafire.com/file/9jvpdxpf0 ... C.rar/file

i am still learning the hex editing / hex2obj - try and error - so excuse me if my way is confusing or nonsense

shakotay2 wrote to search for "1a 27 04 00" - what i can see is that this differs from file to file. for example on the jza80_cage_inid_01.efo it is "19 27 04 00"
i am looking for the combination of "27 06" followed by "27 04" adding 15 bytes offset leads me to the face indices hex "DF5"

https://imgur.com/a/NnzVSwk

neitancrost wrote the face indices ends with an description but it seems to end a few bytes before that - for the cage i was testing hex "111C" as an end

ignore this if it is totaly wrong but i tried: 111C minus DF5 gives me dec 807 divided by 2 ending up with dec 403 as a face count? which gives me a verex count of 185

i simply copied step 2 from shatokay2 example fvfsize 36 and uv pos 28

just after the description of cage_geom0 adding again 15 bytes, the vertices seems to start at hex "121E"
https://imgur.com/a/sdKQeb0

the result seems to be correct - i could only guess that the uv is correct here - since i dont think that there is an interior view or any details
https://imgur.com/a/dMnzsJL

sadly this "method" does only work on files with one object only, tested on cages of a few cars, calipers and so on.

i still strugle on the other parts like the supra body - something went wrong here
https://imgur.com/a/IndJlIj

would be nice if someone could help me out here

cheers!

[shakotay2]

I'd use another vertex start address:
.

JZA80C_Body_genuine_00-efo.png

The face indices count seems to be 18425 but gives some superfluous faces.
(Maybe look for start/stop of submeshes.)

better:

0x382D 14700
Vb1
28 99
0xC92A 8348
021000
0x0 255

[guki]

sooo i'm still trying to get the top secret supra

JZA80C_Body_genuine_00
0x382D 14752
28 99
0xC92A 8348
JZA80C_Muffler_genuine_00
0xDDD 205
28 99
0x10cc 146
JZA80C_NSeat_genuine_00
0xE11 123
38 28
0x1002 44
JZA80C_TailL_genuine_00
0x71A2 885
36 28
0x12D2 560
JZA80C_WHEEL_genuine_18_gold_FL
0x1236 200
36 28
0x2887 150
JZA80C_Window_genuine_00
0x7146 389
28 99
0xFBA 259
JZA80C_Wiper_genuine_00
0xBC4 796
28 99
0x1307 549

https://imgur.com/a/154In05


for the body i went back from the 18425 indices to ~14752 ... this should be the complete body - found one broken face next to the door - also there is no UV available for 28 - 99

single meshes like the muffler, cage, are pretty easy this way - but i still struggle to locate the submesh start

vertices starts could be at
0xDBC - MODEL_BD_mask6_BD_mask6Shape_Shape0
0xFC9 - MODEL_BD_mask6_BD_mask6Shape_Geom0
0xC92A - MODEL_BD_body1_BD_bodyShape1_Geom0
0x494F1 - MODEL_BD_body1_BD_bodyShape1_Shape0
0x4973E - MODEL_BN_bonpin_bolt_BN_bonpin_boltShape_Shape0
0x4996C - MODEL_BN_bonpin_bolt_BN_bonpin_boltShape_Geom0
0x49A36 - MODEL_MR_body_MR_bodyShape_Shape0
0x49C52 - MODEL_MR_body_MR_bodyShape_Geom0
0x49D22 - MODEL_MR_mirror_MR_mirrorShape_Shape0
0x49F1E - MODEL_MR_mirror_MR_mirrorShape_Geom0
0x49FF0 - MODEL_RF_antenna_RF_antennaShape_Shape1
0x4A211 - MODEL_RF_antenna_RF_antennaShape_Geom1
0x4A2DE - MODEL_BD_body1_BD_bodyShape1_Shape1
0x4A514 - MODEL_BD_body1_BD_bodyShape1_Geom1
0x4A5E1 - MODEL_MR_light_MR_lightShape_Shape0
0x4A900 - MODEL_MR_light_MR_lightShape_Geom0
0x4A991 - MODEL_BD_parting_BD_partingShape_Geom0
0x515E2 - MODEL_BD_parting_BD_partingShape_Shape0
0x519C7 - MODEL_BD_parting_BD_partingShape_Geom0
0x51A8E - MODEL_FB_net_FB_netShape_Shape0
0x51E3A - MODEL_FB_net_FB_netShape_Geom0

many of them seem to contain only a few vertices? badges?
looks like the body also contains a LOD mesh - maybe that is the reason for the broken mesh going on 18425?

[shakotay2]

- but i still struggle to locate the submesh start

Search for "_Geom0" as a guideline (for example 10 findings, 3 hits for vertex blocks some bytes behind)
Counts are also easy to track (left part of picture).
The problem is to understand the rules for separating the submeshes (right part):
.

efo - countsAndSM.png

(both parts of the pic have no relation; just didn't want to create a second post for the right part)

[guki]

thanks again for helping me out here

i think i found the correct starting points of the submeshes now
starting with 0x382D for the bodymesh, the superfluous faces are part of the backfacing mesh(pointcloud), followed by the hoodpin detail, a tiny broken bit, the mirrors without any issues, again something broken and the mirror object, ending on 0xC821 covering all of your 18425 / 18426 faces

with an offset i found the underbody starting at 0xFC9 and the doorseam at 0x4A991(pointcloud).

https://imgur.com/a/ym7xvyW

Bodymesh - mesh looks good - uv doesnt
0x382D 14752
28 99
0xC92A 8348
backfacing mesh? - Pointcloud looks right? FVF Size?
0xAB6B 2437
28 99
0xC92A 855
Hoodpin Badges - correct again
0xBE75 112
28 99
0xC92A 8432
Exhaust Badge? - cant tell what this tiny bit is
0xBF53 8
28 99
0xC92A 8741
Mirror Body mesh - looks right again here
0xBF61 960
28 99
0xC92A 8816
Exhaust Diffusor? - again something broken?
0xC6DF 58
28 99
0xC92A 8826
Mirror Chrome Wiper? - Mirror looks good again - there is also a tiny part in the area of the rear screen wiper - cant tell where the broken face comes from
0xC751 104
28 99
0xC92A 8877
Underbody mesh - pointcloud - cant tell where the face indices start
???
36 28
0xFC9 855~
Doorseam mesh - same here
???
36 28
0x4A991 750~

pretty close to a proper result