Always nice to see another game decompiled like this. It's a big deal as far as laying the groundwork for possible ports to PC and other consoles is concerned, and will probably aid modders quite a bit.
If anyone needs a full list of these projects (which includes this one), there's a pretty good selection here:
When people work for an insanely difficult project for more than 2 years, they probably pick something they personally love and don't need any external request.
> That matters because a matching decompilation turns the game from a pile of MIPS assembly into a codebase we can read, build, study, and modify.
Ugh. Why pollute an interesting subject with vapid LLM slop...
I am somewhat disappointed that didn't write much of anything about how they proceeded from their previous plateau, at around ~75%, to 100%. This is basically the first and only truly agentic coding project I've seen create something interesting at all. There is clearly a lot more effort involved than prompting "decompile Snowboard Kids 2, make no mistakes", and they did write at some length about their methodology in past articles, but I would like to see a more extensive end-to-end writeup. This took 7 months invested into the agentic workflow, which sounds so miserable that it's still at a level where I'd rather do the reverse engineering myself, but if this could become a commodotized framework where people didn't have to poke the black box for months with the right incantations, we'd probably see a mass proliferation of old games decompiled where the preconditions match this one (known and available compilation toolchain, so you have a true 100% coverage test suite by requiring recompiled code match byte-for-byte).
Why not? Many folks feel an itch to play a certain nostalgic game that few others enjoyed. And they want to make it even better, especially as our expectations have grown over time.
I prefer 1080 as snowboarding games go. Though must admit some fondness for Cool Boarders and a selection of other lower quality games that few will admit to enjoying.
Decomp tools for N64 have had some breakthroughs even before AI. Now I imagine it's even better. If that facilitates folks geeking out with their favorite guilty pleasure then so be it!
I can only speak for myself but my brain was the Wild West when I was a kid. There was no canon for it to draw on in terms of how or why things were the way they were and this especially applied to creative pursuits like TV shows, movies, music, and video games. I had all sorts of insane ideas about how cool it'd be to implement certain mechanics, characters, etc. in games but this was, of course, virtually impossible at the time. Decompilation paves a reliable path to this type of experimentation - see all of the ridiculous SM64 and Goldeneye mods that are available now (with demos on YouTube).
Before LLMs made these sorts of Sisyphean coding tasks tractable for normal people, there was IRC and Discord, where people with a special interest in programming and emulation could be egged on by the people who delight in the memes. I guess another POV is, were the special interest people and the meme lords ever really friends? If you don't understand what I'm talking about, you aren't really thinking deeply enough about how and why these sorts of things actually get made. A sense of "community" no doubt.
There are idea guys that thought it was funny to decompile an obscure N64 game with little cultural and nostalgic attention, and they found themselves at the intersection of special interest doers which they could egg on into doing it?
More I am just confused for why the game was chosen. SM64, Zelda OoT for example I could easily understand the community motivations behind decompiling. This not so much, which makes the whole endeavor even cooler.
You lumping together IRC and Discord seems bizarre to me.
I'm not sure "community" was always the reason, but we might be talking about different eras. Back in the late 90s and early 00s there were the pioneering scenes for modding, emulation, fan subs, remakes, etc. and it was all highly competitive.
I don't mean to shit on anyone's legacy, but it seemed more ego driven and like there was something to prove either personally or politically. It was cultural and maybe even spiritual. Anyone working on this stuff felt powerful. Nearly a century of broadcast media and being told what to do and how feel by people from far away was ending. Disassembly felt more like deconstruction. It didn't feel like love. It was hacking. There's a reason why one can still shout "hack the planet!" into a crowd of nerds and get them to instantly light up.
I'm not even saying all this as an old fart. Things just changed so fast since then. I'm in my 30s.
If anyone needs a full list of these projects (which includes this one), there's a pretty good selection here:
https://decomp.dev/projects
Though these may have a few they missed:
https://readonlymemo.com/decompilation-projects-and-n64-reco...
https://github.com/CharlotteCross1998/awesome-game-decompila...
I see that one for Burnout Paradise is in the works, but I would love one for Burnout Revenge.
Ugh. Why pollute an interesting subject with vapid LLM slop...
I am somewhat disappointed that didn't write much of anything about how they proceeded from their previous plateau, at around ~75%, to 100%. This is basically the first and only truly agentic coding project I've seen create something interesting at all. There is clearly a lot more effort involved than prompting "decompile Snowboard Kids 2, make no mistakes", and they did write at some length about their methodology in past articles, but I would like to see a more extensive end-to-end writeup. This took 7 months invested into the agentic workflow, which sounds so miserable that it's still at a level where I'd rather do the reverse engineering myself, but if this could become a commodotized framework where people didn't have to poke the black box for months with the right incantations, we'd probably see a mass proliferation of old games decompiled where the preconditions match this one (known and available compilation toolchain, so you have a true 100% coverage test suite by requiring recompiled code match byte-for-byte).
HEY, it was a GREAT game, but GREATEST? COME ON, this ain't no goldeneye
I prefer 1080 as snowboarding games go. Though must admit some fondness for Cool Boarders and a selection of other lower quality games that few will admit to enjoying.
Decomp tools for N64 have had some breakthroughs even before AI. Now I imagine it's even better. If that facilitates folks geeking out with their favorite guilty pleasure then so be it!
More I am just confused for why the game was chosen. SM64, Zelda OoT for example I could easily understand the community motivations behind decompiling. This not so much, which makes the whole endeavor even cooler.
I'm not sure "community" was always the reason, but we might be talking about different eras. Back in the late 90s and early 00s there were the pioneering scenes for modding, emulation, fan subs, remakes, etc. and it was all highly competitive.
I don't mean to shit on anyone's legacy, but it seemed more ego driven and like there was something to prove either personally or politically. It was cultural and maybe even spiritual. Anyone working on this stuff felt powerful. Nearly a century of broadcast media and being told what to do and how feel by people from far away was ending. Disassembly felt more like deconstruction. It didn't feel like love. It was hacking. There's a reason why one can still shout "hack the planet!" into a crowd of nerds and get them to instantly light up.
I'm not even saying all this as an old fart. Things just changed so fast since then. I'm in my 30s.