Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
For example, assume a Japanese game which has 20,000 characters. That sounds a lot, but would actually be a game with very little text in it. The accepted formula for estimating English word count is to divide the number of Japanese characters by 2.5. In this example that would be 8000 English words.
Depending on the translator and client, this would either be priced based on source characters (Japanese) or output words (English). Let's assume the translator charges a reasonable (liveable) amount on the low end of the scale, say 3 cents (USD) per single source Japanese character or 6 cents USD per output English word.
Charging by source would cost the client around 600 USD. By output word, 480 USD. Most translators prefer to charge at whichever gets them a higher pay cheque. So we'll assume that the client is being billed per source character.
They now need to sell enough copies of a game to get that 600 USD back, in addition to costs of development, licensing, Steam distribution fees, payment provider fees, and so forth.
Now imagine that the translator is charging on the high end. That could be upwards of 30 cents per Japanese character or 60 cents per English word.
Maybe they're lucky and found a translator that will accept 1 cent per Japanese character or 2 cents per English word -- the bottom of the barrel. These are the people who generally plug the source text into Google Translate and leave it as is.
That last sounds good and cheap, but the quality will be so bad no one will want to buy the game.
Not entirely true. There are plenty of games sold on Steam outside of Japan that are not available in English. And even more that are only in Chinese.
That and Asian game studios don't really need translations to reach huge audiences, there's loads and loads of Chinese players so if you're a Chinese studio, you don't really need internationalization to reach a large market. Japan isn't too different in that regard.
I'm not sure whether this plays a role, but I also gained the impression (working with Chinese) that they don't really want to communicate in English as well. But that really could be just my (false) impression.
In that case, I'm not actually viewing it as a serious attempt to publish it outside of their respective markets. Not enabling the regional censoring in Steam is free, while everything they need to do on the software is not...
There's also the simple fact that certain idioms, phrases, etc, do not translate smoothly across languages via direct translation. You have to understand the meaning and context of a construct in one langauge and then substitute it for the appropriate one in the new langauge.
When you don't do tha, you get Engrish, or some of the most Hilarious NES era translation gaffs.
There are some times where translation adds to the original. I remember reading an article on games translation in PC Gamer where the main character in a Yakuza game ends up in a fight with a room full of adult babies. Translated into English their war cry became "Pacify him !". Neat, huh ?
S.x.
Not everything is looking at a global audience even if its is on a global store.
No. It is that simple. 1 kanji, 1 hiragana character, 1 katakana character, it's just one character. That's why it's not a word count. Words can consist of 1, 2, 3, 4 or more kanji, or 1 or 2 kanji and several hiragana characters, or several katakana characters, or several katakana characters and several hiragana characters. There's no distinction made. They are all 1 character.
That is why the industry standard is to count Japanese characters and not Japanese words. English clearly denotes words (most of the time) with spaces. So it's easy to count them most of the time. That's why we charge by word when English is used for billing.
When estimating word count for clients, to give them a workable quote, the industry standard is to divide by 2.5 for Japanese into English. Estimates are generally made based on source word count as it's harder to count output words when you have no output.
If you have translation tool that allows it, you can get a rough machine translated output for rough word count quotes, but it will be highly inaccurate depending on multiple factors.
I've worked with clients that prefer to pay based on output English as well as clients who prefer to pay based on source Japanese characters, and even clients who pay based on the number of items translated (and not the word count therein -- great when it's a small word count, not so great when it's lengthy).
The famous translation goofs such as "All your base are belong to us" are not the result of bad translators (well, not all of them) but (and specifically in the case of All your base are belong to us) because the publisher thought one of their Japanese employees with little English knowledge and a high schooler's Japanese to English dictionary was enough to localise their game.
And yes, I've dealt with clients who can't speak a word of English but are still convinced that their "translation" is the correct one.