Hacker News Reader

Comment by ggambetta | Hacker News Reader

While I admire the sheer skill necessary to reverse engineer a game to that degree (and starting from making a hardware dump!), I can't think but helping that the approach I propose at http://gabrielgambetta.com/remakes.html would take far less work than recreating the game from scratch.

View on HN

Well, then they can't do widescreen, 60hz, etc. All of which they want to do and arguably have to do to sell it to console gamers nowadays. Aside from that, perhaps they want larger sprites, better physics, a different game saving mechanism, at some point you're holding yourself back by emulating.

And then there's the fact that this blog post just convinced me to buy this game.

I love wonderboy, I think it's extremely underrated and is as classic, and timeless, and as mechanically refined as Super Mario Bros. and Zelda were on NES. Master System just had so much less market penetration that Wonder Boy is now nowhere near the size of those other franchises, not because it lacks the merit, but because it lacked the publicity.

I had thought this was simply a "hand the IP to some new guys so we can milk the cash cow" when I saw the news that it was being made. The fact that these guys give this much of a damn has convinced me to pay full retail, I want them to be able to keep doing this kind of work after this game is done.

/my 2 cents

That's similar to the idea used in a lot of emulators, especially some of the Nintendo 64 ones, except that they programmed those to look for specific patterns of opcodes that they could replace with higher-level C/C++ functions. Some of the emulators might only support 1 or 2 games, but get phenomenal speed emulating them.

Somewhat separate, but related, I remember one of the video plugins that would accept external images as replacements for in-game textures.

Anyhow, I agree that for at least some games, replacing parts of the code makes getting the game running with some degree of improvement easier than completely re-implementing it. On the other hand, I like the idea of reverse-engineering the original binary, and along with an engine re-implementation, providing documentation of how the game originally worked. I think of it as a kind of software archaeology.

I think your approach is clever, and can result in a best-of-both-worlds situation where the original logic runs with improved assets. But I can also realistically see your solution as a stepping-stone to reverse-engineering the game logic itself to the point where it too can be replaced.

That _is_ what they're doing. They're not recreating it from scratch.

Thanks for the clarification. I didn't get that impression on a first read, but ended up having a chat with the OP and I'm now even more impressed by what they're doing!

Reply on news.ycombinator.com