... Bill Swartz
By Greg Wilcox
Anyway, if you’ve never heard of Mastiff before (or only know them from their two PS One budget releases, Dirt Jockey and Easter Bunny’s Big Day), get ready for a nice surprise or two. A quick trip to the company’s web site (www.mastiff-games.com) reveals that the staff there has been involved in the production of at least 50 other popular console or PC titles for the US and Japanese marketplace. Some of these (and some of my personal favorites) include Biometal on the SNES, both PS One Tenchu games, Covert Ops: Nuclear Dawn, Guardian’s Crusade, and for the PlayStation 2, Sky Odyssey, Shaun Palmer’s Pro Snowboarder and Bloody Roar 3. Here’s some info on Bill, directly from the site: Bill Swartz (Head Woof)Over the course of 12 years, Bill took Activision Japan from a single desk to more than $65 million a year in revenues, making it both the only foreign capitalized game firm to succeed in Japan and one of Activision's most profitable units. Previous to his role as Managing Director at Activision Japan, Bill was Marketing Manager for Koei America and part of Koei's development group in Japan. He is a cum laude graduate of Duke University. |
Swartz: Once it’s been decided to bring a product to the US, the three major factors are time, budget, and audience. You might look at a product and have a discussion that goes something like this:
GameGuy: “This is an awesome game but it’s too short. And the original developers say they’ll work with us to build a new stage.”
BizGuy: “Are you nuts? We’ll risk missing Christmas, and the hard core fans don’t want us to touch the game. Just do the cheapest localization you can and get it out the door.”
GameGuy: “Cheap localization? No, no. The fans will kill us. But I bet even the hard core will like this new stage, and Christmas is a real long way off.”
BizGuy: “So, how many units extra do we need to sell to cover the improvements?”
GameGuy: “Let’s do the math…”
I think you can sort of see where this is going. The good news is: The industry has matured to the point where most companies know they do well by doing good localizations. Profit comes from releasing the best product you can, not by cutting costs or rushing stuff out the door. However, the exact balance will change depending on the project.
DP: Can you give readers an insight into the translation process?
Swartz: The way it’s normally handled in the industry is to take the script and text from the game, give it to a general translator who does a direct translation (believe me, they aren’t pretty: think word salad) and then a producer re-writes the text. It’s put into the game and that’s it. I hate this with a passion. Hate it. It’s the way good games get destroyed.
The big problem with this traditional approach is loss of information and loss of writing quality.
A general translator working through text with little idea of (or interest in) context will probably produce a translation that contains less than 50% of what’s in the original, and it’s usually a document that makes little sense. An American producer will then rewrite the document so it sort of hangs together. But because that person is typically is not in a position to understand the game deeply and frequently is still in the dark about the context for much of the material you wind up with a series of guesses and assumptions piled on top of a translation that has taken away much of the original meaning. It sounds extreme, but it happens in most Japanese projects released here. You’ll also notice that nowhere in the process is a professional writer.
The right way to do it is start with a professional entertainment translator. This is not the kind of word salad producing general translator usually used (“hey, it’ll be re-written anyway and it’s cheap!”), it’s someone who could (and almost always has) produce books and movie scripts for publication. Next, make _very_ sure they have context. This might mean having them watching the game on video, or might mean having them play through. This also might mean adding a line of explanation (“line used only if NPC kills red monster with bare hands”) with every line of text. Then a bilingual editor polishes the result and, depending on the product, a professional scriptwriter also looks it over. This is a very labor intensive and expensive approach, but when you play the game you know where the money and effort went.
DP: If the game requires it, how are voice actors chosen?
Swartz: Start with a casting call. For the auditions for La Pucelle: Tactics (just done last week) we had 10-20 SAG (Screen Actors Guild) actors reading for each major part, so at the end of the process we had about 70 actors and a DVD filled with them reading parts for the various characters. We then took the DVD, listened to all the auditions for a part and reached a consensus on our first, second and third choices for each role. It was a lot of listening with our eyes closed, looking at pictures of the characters and debate. Is the acting convincing? If so, is the age right for the character? How does it fit the other voices? Stuff like that.
DP: Other than text and dialogue, what other changes can be made?
Swartz: Introducing a new version of a game, be it in a different language or for a different platform, almost always means another bite at the apple of production. And it’s a doubly delicious bite…you have the benefit of user and press feedback from the original, and you have a better idea of where you want to go than if you were working with, say, a beta version of the original product. Sometimes you have more time, too.
As for the kind of changes that can be made, the sky is the limit. One great example is Tenchu: Stealth Assassins, which I produced in my former life at Activision. The product as we found it was absolutely brilliant, full of brand new ideas that were well executed. However, it was also rough in a number of ways. It had camera problems, AI problems, graphics issues and was a bit short. We (Activision and the original developer, Acquire) spent a little time and fixed these issues, added levels and made a bunch of other incremental improvements. The changes were enough to make what had been a moderate success in Japan into a world wide smash.
I’d also add one thing here: translating text and dialogue is usually more than just copy typing in a foreign language. There are two issues here: one is language, the second is quality. Frequently the text and dialogue in a game are put together near the end of development when time is scarce and changes frequent. That’s just not a situation likely to produce great writing. A new version is a chance to tighten up the writing and maybe improve the consistency.
But even when working with good writing, a well-done translation can mean making major changes. Someone once said that going from Spanish to English is like altering a suit. You may take in a little here, let out a little there, but it’s pretty much the same suit. Going from Japanese to English requires taking the suit apart, reweaving the cloth and putting it together again. To give the English reader the same experience as Japanese reader often means stepping back a little, revising structure, inserting implied context, and sometimes creating totally new pieces for (to give an obvious example) things like puns that just won’t translate.
DP: Wow- that pretty much takes care of the pressing questions. One more thing, before I forget- is there a strategy guide being produced For La Pucelle: Tactics (and who are the lucky folks putting it together)?
Swartz: Short answer is yes, very likely. We are in talks with a couple of publishers now and I think we'll come to terms with someone pretty soon. Once we do come to terms we'll give all the support we can to provide fans a high quality guide. In terms of difficulty, I don't think you'd absolutely need a guide for La Pucelle: Tactics, but for a number of people having one would make it more fun.
