Leaderboard

So I know it's been a long time, but I have gotten around to figuring out how to extract most of the other models in the WDB. I had to overhaul the entire old extractor, but it is working well after a long time of hacking away at it. The project is still on github, so here is the link to the source project: https://github.com/LiamBrandt/WDBRipper But most people will probably just want the exe, in which case here is the link to the direct download: https://github.com/LiamBrandt/WDBRipper/releases/download/v0.2.0/wdb-ripper-v.0.2.0-binary.zip I found the hospital (medcenter) and the race track, and a bunch of the models that were previously missing. The tool extracts 451 out of the 592 models, whereas the old one only did around 300. I have a feeling that with some more effort the rest of the models could be figured out, but most of them seem unimportant, and a bunch don't even have model data in them, just weird numbers that I don't know what they do.

LEGO Island scream simulator 2016.

Well this just saved me a buttload of time. I thought I would have to recreate the police station, jail, and a bunch of other things. You've rescued me from a terrible fate. Thank you.

I threw this together in around... oh god, 4-5 hours, and now it's 8 AM It'll let you look around the LEGO Island racetrack now that we have the models for it, and enable/disable the individual models to get better views (especially needed since the main island actually intersects with the track in a few places). You can also change the sky color since that's pretty crucial to it looking right (most of it doesn't have any ceilings; it's actually a black sky, dark blue in the dome), and the graphics are fairly faithful to the original game, and I even added a goddam FOV slider, all because I got carried away AND IT'S NOW 8 AM This only has the main landscape models since that's what's important, and they've been vertex snapped together so their positioning should be pretty much dead on accurate http://i.imgur.com/puCZ0V8.png http://i.imgur.com/dGbDqGC.png http://i.imgur.com/cNDFB7S.png Also it's kinda fun to run in 640x480 for EXTREME 1997 MODE http://i.imgur.com/NRQ6l5V.png ... Man, that looks like a screenshot from the actual game, haha, BUT IT ISN'T~ I've uploaded to both Oresome and Mediafire, doesn't matter which site you download from. Windows: http://oresome.rockraidersunited.com/download/416 http://www.mediafire.com/file/w7oe03khwb9ymxj/Racetrack_Explorer_(Windows).zip Windows 64-bit: http://oresome.rockraidersunited.com/download/417 http://www.mediafire.com/file/5ww995cmpmo9kca/Racetrack_Explorer_(Windows_64-bit).zip Mac: http://oresome.rockraidersunited.com/download/418 http://www.mediafire.com/file/qj4hblxxpnat6b5/Racetrack_Explorer_(Mac).zip Linux: http://oresome.rockraidersunited.com/download/419 http://www.mediafire.com/file/s3wcc7znpn8od35/Racetrack_Explorer_(Linux).zip

I just took a look at the SKELETON.bin and I think I know how it could be extracted. There might be other models with the same oddity as this one that didn't extract, If I manage to get it I will update the extractor on GitHub. Essentially the problem is that there is that there is a header specifying all of the components in the model (in this case torso, head, helmet, arms and legs...). Each component header is preceded by a number, I assumed it was always zero, but that number is sometimes non-zero, as is the case with SKELETON.bin. This number for some reason specifies the addition of more component headers, which is just weird. I really don't fully understand a lot of the choices that the developers made when coming up with this format.

Gotta fix the normals but if you wanna see what the racetrack looks like assembled, with and without the main island: Yeeeaaaah... Don't think about it too hard

i made it a little easier to find

I have made a little progress. The files that do not extract are similar to the files that do extract, only more complicated. I would say about only half of the models can actually be interpreted and converted to .obj at this point. If anyone wants to take a shot at it or wants to know more about the format, just ask. I have also made a github repo for the wdb ripper, so you can find all of the source code and current documentation there: https://github.com/LiamBrandt/WDBRipper I am sorry it has been so long, I have been busy, and the format is complicated, but really if anyone wants I will post the details of the extraction so far, and explain what I have done. I don't know how much of the stuff I wrote makes sense, the file that is most descriptive of the format is format.txt in the main folder. Inside of the format.txt, there are #SECTION[x] headings, each representing a section of the file that is repeated x number of times.