101–110 of 118 posts
>n64s greatest game. HEY, it was a GREAT game, but GREATEST? COME ON, this ain't no goldeneye
As someone who has done RE for decades, I feel like I've been seeing a lot of new decompilation projects recently, but even before the rise of AI. Possibly correlated with the release of Ghidra? Either way, it's great to see and perhaps a sign of a greater trend. Controversial opinion: I think the FOSS movement was a setback and distraction from attaining software freedom as well as giving an undeserved negative repu…
Ideas aren’t scarce. Someone who reads a book, or looks at a picture, or makes use of a copy of software is not preventing other people from doing so. The idea that an idea can be restricted are given exclusive use to one particular party for any amount of time by law, is dystopic.
I've been working on porting Rock Band 3 and Dance Central 3 to PC via AI assisted decompilation. The repos are on my GitHub ( https://github.com/freeqaz/rb3 and https://github.com/freeqaz/dc3-decomp ) Like OP, I've learned at lot in this process. I have versions running in the browser now with a custom WebGPU rendering engine. Still lots of jank and vibes, but it's wild to see what models are capable of with the rig…
Earlier quoted context omitted.
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…
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 t…
The other factor to consider is OoT and SM64 are Nintendo properties. Nintendo is one of the most litigious companies in the industry. That alone may shift people away from some of those big name titles owned by them.
I'm also really looking forward to trying the Dinosaur Planet decomp, because the prototype ROM from which it was derived has a tendency to crash the various libretro cores/N64 emulators I've tried.
Earlier quoted context omitted.
[flagged]
Libraries purchase books. One for every book they loan out at a time, digitally or physically. The most recent noteworthy counter-example is archive.org breaching their "one purchase = one concurrent loan" limit during COVID, and they lost that court battle . If you're equating libraries to LLMs, then every leading-model company would have purchased ~every book, newspaper, movie, and song in existence at least once.…
Earlier quoted context omitted.
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 t…
These games are less niche than you think. I for one played way more Snowboard Kids than SM64. The other factor to consider is OoT and SM64 are Nintendo properties. Nintendo is one of the most litigious companies in the industry. That alone may shift people away from some of those big name titles owned by them.
Earlier quoted context omitted.
What is the current SOTA in terms of tools? How are these tools being used with LLMs to speed up decompilation? My current wishlist is to decomp Elite CGA version (tiny x86 binary) back into assembler and annotating all the method names, vars etc. That way I could swap out some of the inner loop using knowledge that has been uncovered in the last 40 years of optimizations.
These models are getting crazy good at examining things like core dumps and disassembly. I've been using an agent to write compiler logic, and its amazing the kind results you can get by having the agent examine the raw binary outputs. I would not be surprised to see agents excel at identifying and labeling patterns for decompilation.
Inspired by https://github.com/scosman/cursed_browser, I have a little art project going where the CPU of a virtual machine is entirely LLM-powered. But even though the ISA is well known and clearly in the LLM's training data (it answers question about it mostly fine), I can rarely get it to even decode a handful of instructions in a row correctly. It'll e.g. do 10 instructions right (even execute right!), then just lose the ability to do bit manipulation all of a sudden and fail miserably at even decoding the 11th. If I try to help it along it'll apologize profusely, do it wrong in five novel ways, before it gaslights me saying I'm in fact mistaken.
Earlier quoted context omitted.
Libraries purchase books. One for every book they loan out at a time, digitally or physically. The most recent noteworthy counter-example is archive.org breaching their "one purchase = one concurrent loan" limit during COVID, and they lost that court battle . If you're equating libraries to LLMs, then every leading-model company would have purchased ~every book, newspaper, movie, and song in existence at least once.…
Many national libraries receive copies of “every” book published in the covered country. They don’t have to buy them. https://en.wikipedia.org/wiki/National_library
>Unlike public libraries, these rarely allow citizens to borrow books.
Which would make them fairly irrelevant for this thread.