Sunday, November 30, 2008

I know kung-fu

There is one thing that I've been wanting to get off my chest.. and it's that I'm tired of being forwarded to the site japanmanship.
It's a blog from an (American ?) game developer working in Japan.
I did read some of the posts and while much of what is said it's probably true in many ways, it's still negatively biased and overall sensationalistic.

I had my complaints about working in Japan in the past. Things certainly do change with different companies and even in the same company depending on a project or a time of they year (more or less busy, more or less interesting work).

The blog in question does specify that many issues are not necessarily only Japanese ones, but then it still goes on criticizing the Japanese situation like a crazy one."Look at me I work among green little people !".. c'mon it's not so special. No need to go all the way to Japan to feel the pain of office work. Sometimes, when I feel like complaining about my work place, I browse over to and I don't feel "alone in Japan" anymore. BS is everywhere.

Myself, having worked for several years in the game industry in the US, I actually believe that Japan is no worse.
Game dev in US implies a lot of overwork as well.. it's also a lot easier to lose jobs and there is no shortage of incompetent people. In fact there just isn't much experience, because many eventually float away to jobs that are more rewarding economically and that require less working hours.

On reason why I prefer working in Japan is that here one ends up working with people that have a lot of experience and that have worked on a number games before.
Productivity is probably not so great on average, also considering the time spent in the office. Japanese see the work place like a second home.. it's in the culture can't blame them for that. Though it's true that game developers in the US like to stick around to play LAN games..

One horse that has been beat to death and then some more, is the development technology one. Much knowledge about software development is clearly in English, a language that most Japanese cannot speak fluently.
Most Japanese people, being often humble, blame themselves for that and look up to the "great" game dev of the U.S.A. like a landmark of productivity. However, I don't think that anybody really holds the key of superdevelopment.
Bosses higer up want better productivity and thrive at the idea of reusability.. but reusability is not so simple. The idea of a generic game engine that will make it easy for many other to develop a game is less viable than it seems.

Many tried and failed to make all sort of engines. An engine, the more generic and unrelated to the game developer it is, the less impressive are the results.

Myself, I think that using those canned engines is usually a bad idea.. but I realize that some people are truly interested in just writing fun games and some game engines out there may actually help productivity. Though that doesn't come for free. External engines cost good money and still require a lot of support.. which means wandering around message boards, sending e-mails or making phone calls to the other side of the world.

What's productive in the end depends on the specific case, but most importantly on the people working on something. Some like the idea to rely on other companies to handle the "system programming", others, like me, would rather spend time crafting a new wheel than trying to fit somebody else's in my cart !


  1. Mr. Kaz,

    The Japanmanship guy is from England, and I believe he worked as an artist. Given that, it's likely that his experiences are going to be different from that of a programmer.

    I think there is some merit/truth to the words he writes; even some of the Japanese I work with realize that things need to be better or could be better. A project I worked on had it's bugs tracked using M$ Excel and not a proper bug tracking software ... at the time, there was clear resistance to the idea of using such software ... I guess because it would disrupt the "flow" of how things have been done ... but I do know that a lot more time was wasted keeping track of bugs that way ... and I can say we were "lucky" because the project wasn't so big ... I've heard about the nightmares (and wasted days!) that have popped up from this manual method. Personally, I think there's no excuse for not using bug tracking software in this day and age. At least in this regards, things are "better" in the US development scene.

    Another (current) advantage out the US scene over the Japanese one is that it's a lot easier to move between companies (if need be) ... I'm not talking about the "I wanna make more $$$, so I'll just quit, etc", but (depending on your situation) it might be trickier to move between companies in Japan ... though I'm noticing that this is slowly changing.

    The pay in the States is of course a lot higher than in Japan; there's really no debating that. To a certain extent, it's not so much the issue of money as the issue of the culture of working over time (which exists in both industries); it would be nice if Japanese game companies paid their employees a little bit more than they do now. I think on this point, it might be difficult for Japanese game companies to attract talent from outside of Japan (at least from the States anyways) because of the large different between pay. They'd only be able to do it for those individuals who won't mind making less because of the desire of wanting to be in Japan being greater.

    Ironically, it seems that there's a repeat in that time where Japanese game companies are branching out a bit and making collaborations with the Western devs (either buy funding a studio, starting one up, or some kind of collaboration). I'm not too sure how well this kind of thing will work because there's always (ultimately) some kind of big friction between the Japanese side and the non-Japanese one. Unless there's a mutual understanding on both side and "cultural breathing room" given, it generally will fail after a while.

    I'm with you in terms of rather writing the "system code" or the engine to run the game rather than some off the shelf middleware. I can see the merits of using middleware, but I'm just a low-level kind of guy and like to build wheels and put things together. Japan could probably make use of it's own "native" game engine, but as you noted, there's a lot of overhead with maintaining that ... I think this is one of the reasons why Capcom hasn't bothered to license out it's "MT Framework"; they'd probably make a lot of money off of it, but support ... then again, who knows. Maybe after a few more iterations of games on it, they'll change their minds and release it. If that happens, that will probably cut deeply into the number of UE3 licenses from Japan.

    I think ultimately it's really a question of how does one want to spend the time of one's life. If a job is not allowing one to meet one's goals, it's probably a wise thing to look elsewhere within the same industry or leave it. That takes a lot more courage to do ... especially if there's a family to take care of and bills to pay.

  2. I'm not sure about the pay. I personally don't see much difference, but that depends on the person as well I guess.
    Changing jobs more frequently certainly helps to get pay raises. In the US it's easier probably because there are a lot of smaller studios.

    As a foreigner though, I can assure you that it's actually a pain to get to work in America. Work visa are given per company, while in Japan one can get a visa once and change jobs at will..
    Immigration in Japan has always been great and quick for me. No hassles, no waiting in line for hours at the freaking Los Angeles immigration office..

    What bothers me is that people sometimes get so involved and seem to only have complains about Japan.. there is no need to make a racial issue out of everything.

    About engines.. support definitely would take a lot of resources.
    Specifically about Capcom, the little I saw didn't impress me. We'll be teaching them something ;)


  3. Well, I took a pay cut to work at the place I was at ... and it wasn't a small one. But as far as the changing jobs bit ... I wasn't trying to refer to the changing jobs to increase salary, but changing jobs because of life circumstances like needing to move somewhere to be closer to family or whatnot. Switching jobs to boost one's pay will only go so far I think.

    I think the visa hassle is only as bad as how well the company one works for (at least in the States) ... and unfortunately, I can't say that many game companies are good in that area.

    I can't speak for others about Japan and whatnot. I personally never experienced any racial issues; only annoyingly outdated development process issues ... I'm still trying to recover from the M$ Excel sheet for bug tracking! 8P

    Looking forward to the "lesson" you are working on to teach the MT Framework people ... he he he ... I just hope they'll hire a good hair designer to teach your modelers how to make fashionable hair!


  4. Too much to read!!!

    Can you guys write shorter paragraphs?

  5. Mr. Ragin,
    When 2 years ago I sent around my resumes, I got a negative answer form Capcom, not even a job interview.
    I wasn't really hoping to get in as I didn't really know the company.. however, as I realized that they have been making news for gamedev technology in Japan, I set myself to help my current company to overtake Capcom.
    I'm pretty sure that I can help to do that in a way or another..
    I really hope that I can make a difference.. I like the idea of a coder's "butterfly effect".

    Capcom may have some complex game engine (really ?), but their graphics seems rather average.
    Next gen real-time 3D is going to be harder than ever before and I'll be very surprised if they can come up with anything better than us 8P

    P.S. As usual, I speak for my outspoken self and not for my current employer 8P


    Mr. Paul,

    Hai !

  6. Ooooh ... the gauntlet has been thrown. I'm looking forward to seeing what you'll produce! I'm deadly serious about the hairdresser bit on the modeling though!! Nothing like virtual bad hair!!

    I think worldwide very few places are going to be ready for the next-gen consoles. It's already gotten to a point that it's obvious that throwing more people at the problem won't make things easier or better. That PS2 project I worked on back when we worked at the same place immediately made me realize what real challenges lie ahead, so now I'm always trying to look at least two years ahead and try to guess what direction things are going to go.

    I still think the future is ray-tracing (or some form of it) ... but we'll see. ;-)

    Mr. Paul, sorry for the long messages!!

  7. Oh what a strange coincidence: Computer-generated hairstyles get real

    Mr. Kaz, I expect nothing but the most realistic hair (properly styled!!!) from your R&D efforts!!! 8P

  8. The image-based portion if not so useful.. however the deep opacity shadow maps work well.
    We have pretty good hair now. Including animation and procedural shape parameters right off Maya. But it requires a lot of GPU power to render and CPU to animate.
    Also very good anti-aliasing is important.. unless one wants to go the translucency route to simulate thickness, but that's even more difficult with the freaking z-buffer.


  9. If you are using multisampling, have you tried alpha to coverage?

  10. I tried that and it wasn't so good. It basically merges all translucency "layers" into one and introduces some ugly sampling pattern.