Image: Nintendo

Homebrew site decomp.dev has announced that it is "sunsetting", effective immediately.

The resource and progress report database, which was first established in 2024, submitted a lengthy statement explaining its actions and speaking heavily on the topic of A.I.-based decomp projects, which have blown up tremendously in the past 12 months.

"Historically, matching decompilation has been a brutally difficult, time-consuming task, on the scale of years for any moderately-sized game with a number of active contributors. As we developed better tooling and shared knowledge around the intricacies of decompilation, the compilers, and the games themselves, progress began to accelerate."

"AI's use in decompilation has been experimented with for years, with mixed results... However, those results still required a significant amount of human cleanup and fixes for obvious issues or cheating (code that technically compiles to the same machine code, but could not be reasonably passed as what the original developers wrote), and the cost per task was prohibitively expensive relative to the results."

While at first glance, the sunsetting of decomp.dev seems to be directly because of the superseded nature of A.I.-based decompilation, the statement remains relatively neutral on the use of A.I. tools themselves.

Instead, greater focus is placed on what A.I.'s expediting of the decompilation process has done to the community aspect of the hobby, with what decomp.dev calls "bad actors" swooping in to snatch up and bastardise the work of others - while leaving no meaningful contribution to the scene itself.

"It is now clear that, with the right guidance, human input, and code reviews, AI can expedite the decompilation process in a significant way. What is also now clear is that the community aspect of decompilation has been broken. Our projects have always been open source to encourage contributions and to benefit the community as a whole. Bad actors have been abusing this in increasingly troublesome ways."

decomp.dev_sunsetting_shuts_down_AI
Image: Sega

Examples given include the scooping up and slapdash "completion" of others' work; an unmanageable "race" to port as many games as quickly as possible, lined with an unnecessary spirit of competition and a lack of quality control; and a blurring of the language, methods, and mentality behind decompilation - with projects essentially being auto-made, dropped, and the modder moving on, with no research shared, no lessons learned, and no history documented.

Paraphrasing, the issue seems to be that the introduction of A.I. to the decompilation seen has solved some difficult questions, but has also created a "conveyor belt" economy, where modders are rolling out port after port or decomp after decomp, with no communication, no sense of technical evolution, no uncovered history of a game's original design, no dumping of rare or interesting resources, no quality control, and most pointedly, no sense of community.

Or, to put it another way, Slop.

"decomp.dev was originally designed to gamify decompilation: watching the progress go up was a fun way to stay motivated when every fraction of a percent took real human effort. Today, that same idea has warped decompilation into a hollow numbers game, exacerbated by folks who just want to burn tokens to make number go up or "port game to PC", all else be damned.

While the increased attention has brought in genuine folks who want to learn and contribute meaningfully, I believe that decomp.dev serving as a public index and leaderboard is now causing more harm than good."

decomp.dev appears to be shuttered as of this statement. Sister sites Decomp.me and ghidra.decomp.dev are still operational. decomp.dev notes that any replacements that might spring up should not use any of the site's original code.