[Request] Wipeout HD/Fury

[josema0890]

Hi guys, first of all sorry if I revived an old post, today I was trying to extract the tracks from WHD Fury (moa therma) following your steps, but I ended with a pile of "garbage" (I think i ended up with "this" because i didn't found the face index, or mb because is a big object with multiple meshes) so i ended up taking a little help and upload the track file asking for your good guidance.

And ty for your guide on how to do it with the feisar ship, so far i got it looking nice with quixel suite and i want to make a wipeout with UE4 for free to enjoy it on our pcs

Here is the rcsmodel file: https://www.dropbox.com/s/t3bgnunlrgyse ... model?dl=0
And here is the pic of the Livery

Sorry for bumping the thread, but i think I have to make multiple h2O extracts and play with the fvf value in every case, cause i got some models looking pretty nice,plus I managed to extract the detonator ship, but is missing some faces on the bottom side (mb i need to use other value instead of 14 on the FVF?), so far so good is looking georgeous, now is time to work with the zone ship, and sorry for bumping the thread, but i need help with the track models

[shakotay2]

Possible values for the FVFsize were 14, 18, 22 (and, untested: 10, 26), iirc.

I've never checked such a large model like track.rcsmodel.

Vaild obj files are 1 5 39 74 68 69 70 71 72 80 125 142 143 151 152 160 161 for example.
First line of track_160.obj reads:
# H2O parameter file: X:\..\tracks\track.rcsmodel_1141.h2o
which you could load seperately into Hex2obj_0,24c_WO.exe (contained in the Make_H2O zip for Wipeout only):

track_1141_H2O.JPG

Spoiled submesh:
For track_246.obj the FVF size must be corrected from 18 to 22 (in track_rcsmodel_1219.H2O).

(This is clear to be seen if you check vertices at 0x7af5f0 in track.rcsmodel, The block ends at 0x7AF7FF.
Size 0x800-0x5F0= 528 dec. = 24 vertices x 22)

(If I have some spare time I maybe will check why it's wrongly set to 18.)

[zaykho]

First, nice guys for the good work, keep it up !

Now, would be cool if we can get a Noesis script or something a bit easier to export (I mean a User-friendly one).

I'm starting to export wipeout things too, sadly, I don't know if it will be better to wait a little if some change gonna being made on the H20 wipeout export ?

And finally here a part of the ship loaded in webgl, you can clik the delivery button to switch the texture applied to the ship. (ps; I downsampled the texture to speed-up the loading process)

http://gravitrek.com/forum/junkyard/wipeoutx/test.html



For people interested by doing a remake/fangame of this Wipeout, you would be interested by this too:

https://docs.google.com/spreadsheets/d/ ... li=1#gid=0

[shakotay2]

First, nice guys for the good work, keep it up !

Now, would be cool if we can get a Noesis script or something a bit easier to export (I mean a User-friendly one).

Coding "userfriendly" devours a lot of lifetime - so sry, no time for such.
(Even if we had 500 instead of 16 interested people/downloaders here.)

I'm starting to export wipeout things too, sadly, I don't know if it will be better to wait a little if some change gonna being made on the H20 wipeout export ?

I'd vote for waiting a little...:

track_rcsmodel.JPG

[shakotay2]

There's 676 obj renamed to objx. Read the former posts how to treat them.

Copy Make_H2O-Wipeout.exe to your unzipped Make_H2O-WipeOut folder (get zip from post as of Sun Nov 22, 2015 1:05 pm, read How-to-use_Wipeout.txt)

(Hex2obj_0.24d_WO.exe is a special version for Wipeout so don't use it for other games/models!)
Copy Hex2obj_0.24d_WO.exe to your hex2obj folder (view link in my sig to get the hex2obj basic zip).

Make-WipeOut.zip

[zaykho]

Coding "userfriendly" devours a lot of lifetime - so sry, no time for such.
(Even if we had 500 instead of 16 interested people/downloaders here.)

No problem, I can understand it.

I'd vote for waiting a little...:

Ok, well I will make test and see what is broken or not before doing full-rip. Thanks again for your work !

Thanks for the new Make-WipeOut, I will start testing map now.

[josema0890]

Well zaykho, to extract the track (not the buildings or ships), u can use the original maya files (most of the maya files that the devs used in the game, ships, tracks, billboards, etc) if you have an original copy of ps2 wipeout pulse, and extracting only the race track meshes (cause they are the same on all the versions (ps2,ps3 or psp) and then focus on the high-poly ps3 models from the ps3 version (building, race track bumps, all the environment and textures).

The path folder is this one: X:/PADZ2 (open it with winrar or 7zip, and woala)

Moatherma track PS2 version inside maya:
https://i.gyazo.com/cf93b769455295f314c ... 2f0cb6.png (link cause img res is too big and f*ck up the theme)

Moatherma track PS2/PSP version with PS3 textures, testing out the mag shader on UE4:
https://i.gyazo.com/d8849799fd11e70560c ... d02e74.mp4

And thanks to shakotay2 and o0DemonBoy0o for working first on extracting the models from ps3 version, i got again the motivation to start again and finish it (that, and because i learned more things in ue4, now is time to focus on the physics coding)

EDIT: Yesterday night, I was thinking about the .rsc model and the PS2/PSP workflow of Studio Liverpool, so it came to my head if we are in front of a maya binary (or an .max/.fbx) file coded to be readable to ps3, so I was thinking if is a way to decrypt the file or maybe is harder to do than the H2O method.

Here is an image so you can see what i'm talking about: http://i.imgur.com/TccGxnD.png

[zaykho]

Sorry if I didn't answered sooner, I was really busy.

So yes, thx for that info, I was already aware of the studio folder leak in the Wipeout pulse ps2 edition, long time ago a guy was talking about this in a blog.

I wasn't aware that the road of the track are the same in every editions, so thank you for that tips.

Still, we will need to work on most buildings, since yes, they are more high poly than the old versions. :p

EDIT: Yesterday night, I was thinking about the .rsc model and the PS2/PSP workflow of Studio Liverpool, so it came to my head if we are in front of a maya binary (or an .max/.fbx) file coded to be readable to ps3, so I was thinking if is a way to decrypt the file or maybe is harder to do than the H2O method.

Here is an image so you can see what i'm talking about: http://i.imgur.com/TccGxnD.png

I don't know, but knowing that shakotay2 is pretty busy too, it would be better to stick to the H2O method.

[josema0890]

Well, i finally achieved to have an anti gravity system (with ag-sys ship ) so far i tested it with the weird box that is under the ship and u can se that the collisions are fine, (i just selected all the track mesh from the maya file, and use complex collision on ue4, there are the results, tomorrow it's time to focus on line trace and have a working system that rotates and tilts the ships just like the track, and then made a "magnetic lock" and polish everything. Epic and Blueprints language are awesome.



T̶o̶m̶o̶r̶r̶o̶w̶ ̶i̶ ̶w̶i̶l̶l̶ ̶p̶o̶s̶t̶ ̶a̶ ̶g̶i̶f̶v̶ ̶o̶r̶ ̶a̶ ̶w̶e̶b̶m̶,̶ ̶i̶t̶'̶s̶ ̶t̶o̶o̶ ̶l̶a̶t̶e̶ ̶a̶n̶d̶ ̶i̶'̶m̶ ̶t̶o̶o̶ ̶t̶i̶r̶e̶d̶ ̶:̶(̶

Good night, or good morning

woop, forget the tired thing, i just let it uploading while i was making some chocolate milk before sleep (don't mind the sparks, i was testing network replication with part. systems and i needed a place holder)

https://www.youtube.com/watch?v=JLTw-iL ... e=youtu.be

[zaykho]

Really nice !!! good progress !! Would be cool to have a website or place to see the progress on it.

As for me, I want to do the same but with HTML5 and Webgl technology. I'm too busy right now. : )

Keep the good work !

[josema0890]

I finally found the Fury pads (weapons/boost) inside the tech da ra rscmodel after a long search hiding models, and then UV-mapped them with blender (not the best, but they are nice) for using it with ue4, here is a screenshot and a download link if someone wants to use them, improve them (i didn't do a perfect job uv-mapping them) or whatever, and you know, sharing is knowledge.

https://drive.google.com/file/d/0B9dy0w ... sp=sharing

Again, ty guys for this awesome post

[josema0890]

It's been a long time.

Hi again

Due to life circumstances, I ended up losing all the progress I had in the previous project because my hard drive broke , and for personal reasons and because of my job I did not had time to restart the project, but now with the quarantine and having a lot of time, I have "resurfaced from the ashes " and get better progress.

Here is a little video, there is a lot to do, but is much better than the progress i had earlier.

https://www.youtube.com/watch?v=Bi5i-Rcr06A

I tried to fiddle with the guide of hex2obj to extract wipeout 2048 ships and tracks but i'm lost because i don't know where to start with the face index.

Here is a link to two models (Ag-sys ship and Sol track) and here are some screenshots i took while following the hex2obj tutorial, also i tried to use the makeH2O_Wipeout from the HD/Fury version but there is no luck because the model workflow have changed

https://drive.google.com/file/d/1YUgC2e ... sp=sharing

Following the guide, on the ag-sys the scrambled alphabet starts on 0x2110, but i don't know where to start, let me explain it with a picture.

For example, on this pic, should i star from the first . after the gxt format or from the last dot after the space on the 0x20B0 row?



And i supose this is the end of the first face index



Now i'm stuck and watched a lot of tutorials from youtube but i end up on the same place and i don't know how to continue nor how to get the vertices or normals from a model

[shakotay2]

well, I just dived into the agsys_2048.rcsmodel and didn't find it too hard:
.

agsys_2048-rcsmodel.png

(Maybe there's parts/blocks that are harder to understand - didn't have the time to check.)

[josema0890]

Hi. good night, or good morning, i don't know if i'm too dumb or what , but i only got to display two models (the one you posted on the previous post and the other one mirrored) when i try to get more meshes all of them are messed up or they are empty like the one on the screenshot.

I followed this tutorial you made earlier that helped me got the two "wing" model from the ship.rcsmodel.

viewtopic.php?t=14695#p120937

Those are the steps i'm making, but getting an empty obj



First of all, find the faces indices, get to the first dot starting by 00 00, then 01 00, then find the end of the scrambled alphabet with three dots.

When i have the ending and the starting positions, I subtract the ending position +1 (c189 + 1) minus the starting position (c022) to get the face indices count, on this case, c18a - c022 = 168 / 2 = b4 or 180 dec, write the value on the step 1 count field and press go1.

After that, i get the max face index, on this case 128, and I write down that value on the step 3 count field.

Then, press go2 to get the Vertex block and then press mesh to view the mesh, but is appearing as empty.

Here is a succesfully shown submodel (the other one you posted earlier but mirrored)



Thanks a lot, at least now i know how to show something on the program.

Don't know if the empty model is for the version of Hex2Obj i'm using but uploadmb is down and i can't get the latest version so i'm using Hex2obj_0.24d_WO

Gn and ty

[shakotay2]

Hi. good night, or good morning, i don't know if i'm too dumb or what , but i only got to display two models (the one you posted on the previous post and the other one mirrored) when i try to get more meshes all of them are messed up or they are empty like the one on the screenshot.

Hi, good morning,
sometimes it's a little bit tricky to find the start of vertices.
From the tutorial you pointed to, concerning detecting floats it says:

v: xx xx xx nn with nn= 42..44 (these are rough values only, can also be 3F, 40)

[agsys_2048.rcsmodel]
The starting address of vertices you chose (0xc18a) is in midst of the face indices block, so deadly wrong.

Correct one is 0xe092:
.

start_of_vertices.png

When i have the ending and the starting positions, I subtract the ending position +1 (c189 + 1) minus the starting position (c022) to get the face indices count, on this case, c18a - c022 = 168 / 2 = b4 or 180 dec, write the value on the step 1 count field and press go1.

Everything fine up to now, but, as mentioned above, you're still in the face indices block at 0xc18a. It ends at 0xe091.

Another trick is to find out the address where the face indices of the next sub mesh start.

but uploadmb is down

thanks for reporting - I hope this is temporarely only. (Otherwise I need to find another hoster... )

so i'm using Hex2obj_0.24d_WO

well, from 4 years ago? (Can't recall the differences to the actual version atm, need to dig for the related source code)

stay tuned!