TIGSource Forums

I'm looking to localize my game into several languages but have never gone through a localization process before so was hoping to get some advice from any others as to whether its really worth while.

Basically how much does it cost to translate, say, 10 phrases of about 10 words?

Does being localized have much of an impact on popularity and ease of us for the local market? Or are they all just used to playing games in English and don't really mind either way?

If all you need translated is 10 phrases 10 words each, then you could have it done free. Just make a post on a popular gaming forum (or game dev forum) if you need the translators to be game-savvy, or on a language forums (like http://forum.wordreference.com/). If you are secretive about the actual text just say what languages you need and what kind of vocabulary is needed and PM the text to the volunteers. If not - just put all the text in your post, the upside here is that if a translation is wrong in some way, someone else might reply with a correct one.

Whether you go that way, or use a paid translator, provide as much context as possible for the translation. If it's a line of dialogue, specify what happened before, what tone is the message supposed to be (joking? angry? ironic?), anything that might help. If it's a part of the GUI, consider the fact that words may and will vary in length - include game screens so that the translator can use a longer equivalent of the word or use it in a phrase (or: use a shorter one, abbreviate properly).

Also, some (many?) languages that use the Latin alphabet, extend it with diacritics and new letters, so that's one more potential problem - especially if you are using custom fonts. On the similar note, if object/subject/numerals in your phrase will be supplied on runtime (e.g. "X hit Y for Z damage"), know that words in your target language could be affected by declension (http://en.wikipedia.org/wiki/Declension) and/or conjugation (http://en.wikipedia.org/wiki/Grammatical_conjugation) (and possibly other magic like that).

As to how much of an impact localizing your product will make, that depends on a lot of factors. If you are aiming for "casual" demographic, then translating your game might be a good idea; seasoned gamers should have a good enough grasp of gaming English to play your game, unless you include some really obscure words. While Flash games are generally accepted as one-language medium, people might expect localizations in stand-alones. Etc.

All those things considered, I would say "f@$k that s#!t". English is the lingua franca in today's world, and if someone won't bother learning it, I'm not going to do the work for them.

All the translation for Eufloria was done by fans and they got a free copy and manual credit in exchange. It was great for us. I bet we sold a lot because of this, especially to Russia.

My only successful game was (still is) translated by volunteers and fans too. There are more than 20 language paks now.

So people like to have localized versions, but one can start with english only, I think.

Thanks for that info guys, I wasn't really sure where to start.

biomechanic: i'll check out the word reference forum, thanks! that looks great

and to alex and hajo, the idea of a free copy in exchange for translation is a great idea.

i do agree that english is the lingua franca, but know that i'm sometimes annoyed that i can't get a jrpg in english so thought i'd try and find a nice way to make my game more accessible to a broader audience.

you've all helped heaps, thanks!

Make sure that your game is succesfull before trying to localize it, otherwise it doesn't make sense to translate a game in lots of languages when nobody wants to play it.

I say, stick with english first, and leave the option for localization in the future. See if the game works in english first.

Just so you know, Contact me if you need a Spanish Localization, I'm Mexican. And I wouldn't mind doing such a short quantity for free!

About that, what you all think about the problem of implementation?
I mean, if you're not sure of the success of the game, you might write your code (if you code...) in a way making localization harder. So, should you "always" prepare your code for future localization if you're planning to be successful (I mean when you're trying to do something that is not a raw sketch of the idea of the game be a polished game)?

I always leave my stuff localizable simply because it's not usually hard to do. Something as simple as loading all game text from a separate file is usually enough.

I always see videos of indie games on NicoNicoDouga (basically the Japanese "YouTube") where they'll try to play the games but get stuck on parts that require a lot of reading in English.

They really seem to enjoy I Wanna Be The Guy though. Then again, there's not a lot of required reading in that game, it's nothing but really hard gameplay (which has no language barrier).

On the other hand, I saw someone on that site trying to play Spelunky but getting stuck on the tutorial due to it being in English.

I just wonder how much more popular indie games would be over in Japan if they included a Japanese language option. It'd be interesting to see full playthroughs or speedruns of them on NicoNicoDouga.

Now here's an interesting question: what about for a dialog-heavy RPG? Is localization worth it? And if so, for what languages?

I ask because the strategy RPG I'm working on right how has a lot of dialog. (I just recently finished writing up the dialog from the start of the game through to the end of the first mission into a script format for voice acting purposes, and that section alone takes up 33 pages.) I'd love to reach untapped markets if it's worth the trouble--but only if it's worth the trouble.

I guess it depends. is your game designed to output a database and read from it all the strings? Then it may be real easy even if there's a lot to translate. After that, it should just take time.

I can only speak Spanish and English Fluently, but it begs the question if indie games can profit from localization. Maybe to Japanese? they sure love their Dojin games.

Ah yes, I forgot that I'm in a quite complex case  :tired:

I have to load modules (containing dll/so + other resources) that contain those localization files, requiring a bit more organization in the way I load the localization files ^^;

But that's right, it's not that hard in the end.

The real problems start if a game assembles sentances from parts and the grammer needs adjustments.

For example, a template like: "You hit the %s"

Assuming the game will insert the monster name for the %s. In some languages the exact word for "the" depends on the inserted word though (french, german, maybe more) and just translating the sentance template will not give the right results.

This becomes worse the more complex sentances a program tries to construct. I think most of these problems can be solved, but it doesn't look easy to me - at least I had one project where I tried to keep everything localizable but the sentance construction tunred into a nightmare.

Also, don't forget number formats. There are many ways to write thousand separators (some languages want none), and the decimal point can as well be a comma in some languages.

I write dialog-heavy games, too.  I've gotten lots of requests from people asking if I planned on translating, in addition to people offering to do it for free!  But, sad to say, they underestimate the work involved and in every case they disappeared without finishing.  Even as short as the Blackwell games are, they easily have several thousand lines of dialog each.  I get the feeling that if I could get them properly localized in German they would do very well, but I don't have the resources to do it properly.  There's nothing worse than a badly translated game!

Spain also has their share of Adventure game fans. They still produce many of their own. Germany has always been Adventure game friendly as you say.

Maybe we need a game translation community site. a standarized access/database for game text and translation tools for fan translators and indie devs. anybody interested in that? maybe this translation portal could financiate itself with ads...
maybe i could build up such a site if there are enough people interested in that.

Yes, please!  :handshakeL: ;D :handshakeR:

That already exists, I've been on one some years ago, but can't remember wich.

with regards to dialogue heavy games, wouldnt that actually make it more worthwhile?

the game i am looking to localize has minimal text ("start", "go" and some short tutorial messages) and most non-english speakers have either known enough english to get by or have been able to largely pick up and play without the tutorial messages anyhow.

but if dialogue was a core aspect of the game, surely that would make localization even more worthwhile? ie my market is only somewhat constrained by people not being able to understand the tutorials, but a dialogue game is absolutely locked out of a country if it isnt in their language.

Well, it would make the game much more accessible, but it would also be way, way, way more work. So the question is, is the additional benefit worth the additional work?

I'm making a strategy game and it's already worthwile, because it targets geeks. And geeks are universals, citizens of the world  :gentleman:

You have to "know your audience". Knowning your audience is the second guide to make right decisions for your game. The first one being knowing the fundamental theme of your game.

Once you know your audience, you'll know if it's worthwhile.

I like the idea previously of putting the game out there and letting fans/volunteers translate, so it makes sense to prepare your game for localization from the start.

We will be doing this with RedRiot, simply have all the text loaded from a language file.
 (The hassle is making sure longer translated words fit, especially in menus)

And I was wondering about translating then releasing, but I think releasing the Beta, then asking for volunteers to translate makes more sense, then we launch the full version with as many translations as possible, and it remains open to more.

I think it's quite worth it.
This data it's from the casual games market, but i got it on one of my Linked contacts in the industry, and his numbers were something like :

German Market : 540 million dollars
British Market : 340 million dollars
French Market : 270 million dollars

This is just about casual games, but i would get the indie market would not be to much far away. Also depends on the type of game, from what i have read somewhere, Germans are quite addict to Strategy and Adventure games.
No idea about the brits or the french.


Bruno

I'm not sure those numbers mean anything. They don't state that these countries only buy games in their own language, do they?

British can speak English .. that is the language most indie games are made in anyways.

The French and the German both have a good share of people who can read English, so it shouldn't be that a pressing matter ...

I'd assume Japanese/Chinese to be more important localization, also Russian. Particularly because those languages needs different fonts, and the people there not only need to learn the English language, but also Latin characters.

Yes, they state that this countrys buy games using their own language.
I didn't mean British, i mean Spanish, sorry :)
If any of you already made localizations of casual games, and have acess to their sales, then you would see how important localization is (for casual games, i have no idea for indie games).
Russian/Polish markets might be also important, no clue, in Russia i think games are sold much cheaper than in the rest of the Westen world, but i assume as the market is huge, it's probably worth it.
BigFish currently has support for 5 languages (English,German,French,Spanish and Japanese), but this year they will increase this number to 10 languages.

I wouldn't say that...

Lot of online French people knows english but frankly it's really not much. I think only "geeks" speak also english.

A game with minimal text doesn't really need localization. A big RPG will benefit but the cost will be higher. I would recommend coding to make localization easier, but launching in your native tongue first and then localizing if the game is a success.

Strawdog Studios localized their iPhone game Turbo Duck into Fre, Ger, Ita, Spa... the game sold more in Japan (unlocalized) than any of the localized versions did.

Canadians from Quebec understand and use english fluently, in France it is a different issue and I agree with Klaim's statement that only "geeks" or players community will play games fully in english.

By the way, I'm french and I will be glad to offer my help to translate games from english to french if needed (just PM or e-mail me).

From the opinions I gather about Germany, practically everyone who is of the age to be playing video games can read and/or understand English... some players just feel like it's their god-given right to play the game in German. As pointed out above, the French are a bit more proud of their language and thus not as likely to know English.

Russian, Chinese, Korean, and Japanese localizations sound potentially very fruitful, in that order. I have a feeling the Japanese generally only ever play doujin games and don't dip into the western indie scene, but I'd love it if someone proved me wrong. The Russians are in a lucrative situation; they're quite a big market of which many do not speak English that is more likely to check out and play western titles.

What I feel is more that most French don't really understand the need for learning anything other than French in their school days, making them bad at learning new languages. Really, it's more that most Frenchs aren't used to learn new languages. I think Japaneses have somewhat the same problem.

In fact, as France is the most visited country in the world, I guess most of the people knowing english well are : 1. geeks & tech people 2. commercials 3. having restaurants
I know a lot of power-gamers that just don't know english very well.
They just don't understand why I suggest  them to buy th eenglish version of Mass Effect instead of the french one, but it's obvious for anyone knowing french and english and trying to see the difference.
I think it's really problematic because the french education is in crisis since few years and I fear for the coming generations. At least they can learn a lot by being connected... but you know, most of the french young people can't event spell french well.

I'm not a language laywer but it's really sad because the language is like a protocol for communication. We lost a lot by having so much young people that can't communicate fast en easy with other older generations (or maybe even with younger).

same thing in germany. especially when your target group includes children younger than 16 or people who work in non-IT areas. in addition in germany all foreign main stream media is dubbed (music excluded :))...

What do you think? Should I deliberately not translate my game into French, just to encourage the French to improve their educational system?

No (sorry for the late answer).
Don't forget it's to help YOUR game that you add localization, not to help others (and French are often egocentric).

So in your case (and I'm in the same one in fact) I would ALLOW localization and then focus on english, and see later for other languages maybe helped for free by community members.

I'm quite overwhelmed by such a massive lot of bullshit about Frenches.

Proud of our language? Well no, you are the ones that find it romantic or whatever, most people talk English here, if not a fluent English, and I don't think anybody would have problems understanding the 2 or 3 words that appear in a shooter game. Now for rpgs, it could indeed be bit different since the vocabulary and grammar used can be a little more complicated, and big parts of the scenario could be lost.

If it's for a rpg, i highly recommend you to translate it(French and German is the way to go i think).

I can only suggest you my help for the French part , I've already translated some games, the last one in date being a famous game on tigsource, Jumpman.

Has anyone mentioned the input differences yet? It's quite possible that you'll have to recode your input system if it didn't take into account localised keyboard layouts. For text only translations that don't have too much dialogue, I don't really see why you wouldn't translate it assuming you have a solid game to start with.

Personally, I use OIS (C++ lib) that allow to easily manage inputs whatever the keyboard settings.

As a French person I can say that keyboard layouts are WAY TOO OFTEN overlooked . I have long stopped counting the number of indie games (esp. Flash games) I could not play because they required WASD for movement. Just look how these keys are laid on a French keyboard : http://upload.wikimedia.org/wikipedia/commons/thumb/5/55/Azerty_fr_zoom.svg/800px-Azerty_fr_zoom.svg.png >.<

No matter how fluent people are in English, they won't remap their keys just to play a game. I think this is even more crucial than actual translations.

Why...

Why do you have two M's?

Also what do game developers with wasd controls typically do for French people?

The lower M is a mistake in the chart, this key does not exist (and therefore the rest of the row should be shifted left). This one (http://www.neroucheffmichel.be/html/clavierHtml/clavier/clavierAZERTY05.jpg) looks better.

If the game is localized then WASD usually translates to ZQSD. When it is not, the best solutions are to offer alternate control schemes (arrow keys based for example) or customizable key inputs (in the options screen or w/e).

Man, that sucks. Because there isn't really a lot of conciense about that.
All you get is the "Once upon a time ... DVORAK dragon...which was killed by the QWERTY knight"

I like the way wxWidgets handle Internationalization because it doesn't break your flow.

You just call a function "_" passing the text in the default language, e.g.

Then a little program (poEdit) extracts all the _(whatever) from the code and allows you to translate it and save translations to ".po" files, that can optionally get loaded later by wx. If you don't, no problem, because it defaults to whatever string you passed to the _() function (really fast).

It's not the fastest/best/most optimal way but it's the more intuitive and easy to implement. And if you have something that you use all the time, you should cache it on a variable.

Apparently that ".po" files are pretty standard (GNU gettext catalog format) and even Python can use them


Regards

It's not wxWidgets, it's GetText library : http://www.gnu.org/software/gettext/

However I think it's bad for some type of games.
The first problem is that it assume you will put text in code and then translate later. If you do that, when you have some short sentences (or one word) that might be used in several context to mean different things, in other languages that might be different text that should be displayed. But then you have only one id, the original text.

However there is almost to other alternative at the moment. One promising library is being developped for boost (C++) but it will not be there until a lot of time. (you can still get in the mailing list if you want to participate to it).


I think a simple localization library + tool might be helpful for everyone here. The main thing that makes it hard to build is that the implementation of localization is really game-relative. I'm making a game where all the texts should be loaded at runtime start because texts are provided as language packs for future extension. Other devs would be happier (and would not worry about memory) with statically compiled texts (making several binaries, one for each language).
Making a good translator tool is also a hard task but it would be so useful. I've used some before, but only in companies I worked for so it's not free or open-source.

So I take it that if a game is installed on a French system then the key bindings will automatically go to azerty style? Is that how, say, Source engine games do it?

That really depends on the engine/implementation.

I use OOIS (C++) for keyboard and mouse inputs and it takes the local by default I think.


Some engines will provide advanced features to manage localized keyboard, but it's not useful for every games.

The best thing is to make sure there is a simple way for the user to re-map the keys to actions.

I was totally kidding, dude. :P

I honestly think TSOG simply has way, way, way too much text hard-coded into the game to do localization effectively. It's a shame. The next time I make a game, I'm going to try to have all the text in an easily accessible external file.

Googling OOIS doesn't give the website so linky pls

This one : http://sourceforge.net/projects/wgois/
It's used in sample applications of Ogre rendering engine.

One easy way for French people etc with differing layouts is to simply add English to their Windows' language bar as an input language. Then you can just switch to the English keyboard layout when you run a game with non-customizable controls without having to remap anything, because your A and Q are now "as they should" even if the keyboard doesn't physically look like it. I usually change the input language to my own whenever I'm dealing with foreign keyboards because I can touch-type so some letters being in the wrong place doesn't matter at all.

Oh that's a good solution. Thank you.

I use it too but remember that it's a solution for advanced users. Other potential customers will only assume tha the game will provide some way to make it work with non-english keyboards.

Localization will be worth it if:

- You're targetting casual gamers - Casual gamers are... casual. Their English proficiency is usually much lower than hardcore gamers. Your store description alone will filter them out if it's in English only, so localization matters a lot for them.

- Understanding the game is essential - If you have to explain gameplay mechanics or if you're telling a story, consider localization or people might get frustrated and leave bad reviews because they can't play the game or fully understand it.

- There's a potential audience for your game in foreign markets - Do some market research to see if similar games do well in certain markets, if there's a lot of competition, if you can have visibility, etc.

- Your game already has some visibility - If it doesn't, localization will still help it reach more people, but it won't do miracles either and you may not get ROI.

- It's well executed - This point is especially key if your game is narrative-heavy. Sloppy translations ruin the experience, so make sure you get quality translations.


If you're interested in reading stories on that topic, we recently asked 3 indie studios if localizing their game was worth it: http://www.leveluptranslation.com/single-post/is-it-worth-localizing-an-indie-game  (http://www.leveluptranslation.com/single-post/is-it-worth-localizing-an-indie-game)

Level up translation look very nice, i plan on localizing my game at them. There are some big markets in the world that are used to having their games localized, like Russia, Germany, France, Spain etc...

You're welcome anytime, Fat Pug Studio! :beer: Give us a shout when your game's ready for localization ;)

France, Italy, Spain and Russia do have a relatively low English proficiency, so localization is almost mandatory if you want to make the most of these markets.

Yeah, i've been following you on Twitter for some time, you have some nice articles on localizations, i'll be sure to contact you as soon as the game is finished :)

Thanks a lot, really glad you appreciate the content from our blog! :handthumbsupL: