Being hacker and translator, is it insane?

Started by Dezmancer, November 28, 2021, 10:04:20 PM

Previous topic - Next topic


I really want to contribute to some translation projects one day, but I'm kind of torn between which role I want to do-translating (which means learning more Japanese than I currently) or hacking (which means learning a *lot* more programming than I currently know)

In general, it seems more often than not that having a hacker who can figure out how to extract and re-insert the text back into the game is the thread on which a lot of projects live or die, while there's more people who know Japanese every day. Not to make light of effort involved in translation at all, but you're always working with the once you've got the text, where as every game is a bespoke design that's going to need a custom solution. So from where i'm standing now, focusing more on Japanese seems like the 'easier' thing to do, but then I'd be limited to doing whatever project had a hacker if I wasn't willing to translate a game and just hope one eventually came along.

But honestly, there's a part of me that really wants to do both. Hacking and translation. I usually don't see those roles done by the same person on a successful project, and I imagine for good reason. Is it just nuts to take on? Should I pick one and stick with it no matter how difficult a choice it feels like? Really appreciate any insight on this from folks in the community.


I would say the opposite. Certainly there are projects for which hackers up and vanish, or maybe ran into issues inserting where ripping was easy (if not making things harder in the insertion stage but sub optimal ripping).
Though this is tempered by hackers knowing the work involved so maybe not making the leap until a translator is onboard, translators maybe not being able to readily assess the abilities of the hacker (or staying power) certainly does happen but it is less rare. Plenty of hacking work happens in private as well until it is all done -- open collaborative stuff I remember seeing threads around here maybe mid 2000s as a kind of show piece to bring in some more people but it would not be until the DS was current (most of which failed) that things got tried in earnest.

Equally knowing both, or at least a bit of both can be a massive help on both aspects. is the usual link for life as a commercial game translator and many of the horror stories there are technical, or from technical people maybe not appreciating language (not that there is a good technical approach to the massive variation in languages).
In my case I can treat Japanese as the proverbial moon runes and probably get it done but knowing the types of characters involved, general approaches to fonts, general approaches to punctuation, what furigana is and other things that might reliably be on that list allowed me to dodge several things that I would not have known about or thought to look at, and do some of the more mathematical approaches/think like the original dev approaches I quite like to do over more brute force or textbook.

You may also be thinking too much in the Japanese-English mindset -- how many games translated into non English, especially by hackers, use at least the English script as a base, possibly because they also learned English to if not native then more than good enough by playing games.


Is "insane" really the right word here..?

Quote from: Dezmancer on November 28, 2021, 10:04:20 PMhacking (which means learning a *lot* more programming than I currently know)
In the end, ROM hacking involves a lot of specialized skills that you would be unlikely to pick up over the course of an ordinary education in "programming". But nonetheless, if you've never done any programming before, you might want to try getting your feet wet one way or another. Some people do not take to it at all.

I expect a knowledge of Japanese would be more versatile. (I'd like to think Retroarch's "auto translate" might one day make it much easier to do a practical translation – albeit one that would never work on original hardware – without any hacking knowledge at all, but perhaps that is wishful thinking.)
This signature is an illusion and is a trap devisut by Satan. Go ahead dauntlessly! Make rapid progres!


Thanks for both your responses, apologies in my lateness on this thread, been a bus few days.

"Insane" probably wasn't the right word, I guess I was just wondering if it made sense to focus on both at the same time. I do one day hope to work on porting and translating retro games in a commercial capacit. For porting, think projects like the recent port of Gley Lancer to modern console ( and the upcoming Night Slashers remake ( I'd like to get experience in both, though I probably would lean more towards the translation skillset.

Based on response and some more general research, it sounds like learning how to hack/text dump games isn't a going to have a lot of crossover with the kind knowledge you'd need to port a fourth or fifth generation console game to Steam and the like (happy to be wrong about this, though), so in that vein it might be best to focus on translation. I have dabbled enough in languages like C, C++, and Javascript to at least understand fundamentals, so I'm not a total newbie, and a part of me does want to try it out anyways just to see if I can do it. Any recommendation good "trainer" games in that are easy to dump and reinsert text for in both English and Japanese would be appreciated to get familiar with the basic ideas, if nothing else!


I vary on whether the truly simplistic stuff does any good. I might not throw someone in at the deep end but giving someone a game that is basically a text file to translate, or more space than you will ever need and basic encoding, pointers and font you could stand to alter/expand is of dubious merit.

If you want to take a selection of ROMs and fire a relative search tool against it (I like but there are others), maybe explore what encodings you are allowed (can be a brute force thing, can be a code thing, can be from looking at name/high score input, can be guided by finding the font in a tile editor) and tweak a pointer or three then that will do you well.

Porting games varies as well.

PC, maybe also Amiga, aside then anything 16 bit or older will likely be mostly assembly. To this end you are probably going to be in the straight remake (how much the recent GTA debacle is going to have soured people on things there I don't know. It is also expensive) and emulation overlay territory (it is cheap and effective if done right, not everybody has to be Steam Final Fantasy). Some hacking skills might not hurt -- being able to fish or current text out of RAM, whether to display it more clearly or as an identifier for your overlay (IF text = this then display that from the database sort of thing), being able to make cheats enough to get a handle on variables as it actually shakes out (or simply make trophments if such things are deemed necessary), and at the highest end being able to recreate the feel of a game (a complex interaction of inputs, enemy behaviours and more besides, and something even the unskilled players will pick up on, even if they don't necessarily have the vocabulary or even being in the "You didn't notice, but your brain did" regime. You can do reasonably well with matching acceleration and jump height curves for different levels of button fondling and playtesting/video but he what can do code does well, and might well also do well when you realise that game feel has come on leaps and bounds in the last few decades).

PS1 through early-mid PS2 is going to be C with a bit of assembly, newer will be C++ and also start to mix in serious amounts of engine (Unreal engine today is very much its own thing), might get quirks like the xbox 360 technically doing a version of .net/C# and there likely being a few other languages in the mix as well (lua and possibly python depending upon where you look). Porting might also be a "you had to be there, man" type scenario as one knowing all the libraries from having gone through it* will give you a massive leg up over even a skilled coder unfamiliar with it all, and unlike K and R C or COBOL there are few real avenues to read much about their quirks to replace a greybeard in a company with their back against the wall. This also says nothing of the lost source problems (do a search for sega lost source). When we will see commercial decompilation in action (it is very good now -- I assume you have seen the Mario 64 and Zelda efforts, several more N64 things likely to drop before too terribly long, Diablo was more of a lucky break but still got it done) I don't know.

As far as translation in a professional capacity I would go back to the loekalisation link from earlier as that is probably the best resource I have ever found for how that shakes out. Short version though is it is a professional skill all of its own, though some Japanese and coding knowledge to interface between things (translation and engineering usually get along OK, certainly no art and/or business consultancy and engineering type slugging match, but a bit of knowledge of the general approach of either makes life so much nicer), maybe read 20 year old (possibly in both senses of the term) Japanese programmer comments... good stuff.

*not really the same thing but fascinating and I would be serious money on it paralleling all manner of things
GDC in general has a load of good stuff (see if you can find the post mortem series, Diablo being a pretty good one there), Ars technica has a series called war stories that is also good, and actually I might suggest some testers ( ), has some good stuff too on n64 era Rare. As I am linking favourites then might as well also make an appearance.


It really depends on the type of game that you want to translate.

For example, a game that is pun-riddled, makes many pop-culture references and has many dialogues while at the same the text and graphics use an unconventional and severely complex compression routine would be the wisest to ask for help.

Conversely, a game that is simple, with very few dialogues and uses uncompressed text would be an easier to manage project.


Insane? No. Very difficult? You bet. It also usually takes forever to finish most projects, unless you are really dedicated. Most of the really good stuff left is either difficult to hack or text-heavy, which is why most groups have at least 2-3 people. Gotta remember that it's usually Asian languages that are involved, there aren't "that" many country-exclusive games in Europe to translate (there are some, of course, they just seem less popular targets). So, I'd say it's up to you if you wish to either focus on one thing or try your hand at both. I'd argue knowing Japanese is perhaps more versatile, since there's no real 'pattern' for hacking different systems, but if you want to hack a specific game - that may be the only way, seeing that hackers also usually just work on what they like.


Quote from: cccmar on December 03, 2021, 01:43:49 AMMost of the really good stuff left is either difficult to hack or text-heavy
I think one of the answers to this is not to try to translate the "really good stuff" (whatever that may be), but to translate the really niche stuff instead that only a few people will be into.

Not every translation has to be a localization of some massive RPG; it's also time well spent to translate a shmup, a simple platformer with only a few lines of text, a sports game, a kids' game, etc. And once you've done that and actually finished the job (instead of adding to the piles of overambitious unfinished projects out there), you'll have more confidence for the next, larger, more ambitious task. :)


Yeah, there are loads of VNs few groups tackle due to the sheer amount of text, for example, so I'd argue the 'big' game descriptor mostly covers RPGs. I think it's mainly due to the translation competition in the 90s that mostly such 'big' projects are tackled by most of groups predominantly. In fact, some systems don't have so many massive games at all, so finishing off Master System, N64 or GameCube library shouldn't be "that" difficult in the future, if there's willingness to do so. :)


Can't recall, did any version of Tokimeki Memorial ever end up translated? (the proper game, I'm feeling like there's at least some odd sidegame that's probably been done)
That was probably the most well-known game for which lots of people/groups had the desire to start (especially the SFC version) but none had the capability to pull it off.
"My watch says 30 chickens" Google, 2018


Quote from: KingMike on December 08, 2021, 09:08:52 PM
Can't recall, did any version of Tokimeki Memorial ever end up translated? (the proper game, I'm feeling like there's at least some odd sidegame that's probably been done)
That was probably the most well-known game for which lots of people/groups had the desire to start (especially the SFC version) but none had the capability to pull it off.

Tom (Retrotranslator) and DDS are working on it currently. They've both gone mostly silent online thanks to the cancerous bullshit that happened over the Goemon translations, but Tom recently tweeted out that he is very much alive and still actively working on all projects he'd previously announced. These two (plus Pascal Vahanian) were the ones who finally finished translating all of the SNES Goemon games after decades of them being started and abandoned over and over again. They seem to have a knack for finishing previously "cursed" projects that always end up abandoned and doing them very well, so I expect they'll have it out sooner or later. They don't have a specific release window/time frame of how long it usually takes them on a game, so there's no telling when it'll get done, but I have confidence that it will.


@Dezmancer if your interested in doing both, as someone who kind of does both, i'm willing to teach you if your willing to be recruited.

My discord is Dawnbomb#3408


Both are great skills to have in general in life, there's no harm in adding them to your personal materia slots in your brain. You should always be adding to your brain slots, in fact.