We sat down with Technical Director Mike Mika to talk about the making of Tarzan's and all things Game Boy.
Q: Tarzan is really Digital Eclipse's first truly original title. Your previous efforts were almost all based on classic videogames -- was that a conscious decision or simply a natural development of a maturing game developer?
Mike:
Well, both actually. Digital Eclipse has done very well by making its mark as the first commercial emulator group. Our company's relationship with Midway has always been very strong in that regard, and when Midway asked us to do their Game Boy Color launch lineup, it was time to finally initiate our plans for growth.Q: What are some of the difficulties you ran into when you started designing a game almost from scratch?
Mike:
Well, initially the largest obstacle was in trying to do the game in just under three months. Ideally, a sixth month project time would have been much better. Luckily, we had been working on new technology that would allow us to focus more on game mechanics than actual coding and problem solving. This gave us an opportunity to create massive amounts of animation and link it all together without sacrificing control or gameplay. After all, it was a Disney title, and we felt we had to at least attempt to live up to the old Virgin Disney titles such as Aladdin for Genesis or Lion King.Q: How many people worked on the game all in all?
Mike:
Too many! We had about five times the programming staff dedicated to art. So, two programmers, about ten artists and three people designing levels. We work with a large collective of talent out of house, too. So the numbers varied. Luckily we have some very talented producers on staff that kept everything under control.Q: Is that a typical size for a Game Boy development team?
Mike:
Well, that is a two-pronged question. Tarzan was a great number of people for a Game Boy game, and if given the proper amount of time, a team that size could regularly produce extremely compelling and original games. However, most publishers work on a time frame of three to four months. For some reason, Game Boy games are perceived as easier to produce by publishers, so they must take only a fraction of the time to development when compared to, say, a PSX title.I would contest that. I think the best games on Game Boy often come from a decent schedule for development, which includes a design phase, research and development, and final assembly of the game itself. But since Game Boy has taken off so fast lately, it has created a very competitive market out there for developers. If we say we need six months to make a game, publishers simply go to someone who claims they can produce a game in six weeks. Problem is, if we as developers continue to lower our standards, the amount of garbage on the shelves could destroy the market for Game Boy games altogether! So, to answer the question, we would prefer teams of Tarzan's size. But, typically, you have about four to five people on a title, one being the main programmer.
Q: How closely did you work together with Disney on Tarzan?
Mike:
Disney Interactive is an amazing and accommodating company to work with. They worked very hard to make sure we had everything we needed, from reference animation to actual stills and captures from the movie. From the start, we worked very closely with Disney. They were very open to our design ideas and had plenty of thought invested in it.Q: In the same vein, how much freedom did you have in the creation of the game? Were there some things you couldn't do?
Mike:
Activision literally let us run wild with it. They needed a company who could do it all considering the time limit on the project. There were, of course, many things we couldn't do. We had plans for riding birds and playing as Tantor. Also, we had plans for more gameplay goals. We had hoped to have characters interact and propose objectives. The original design was quite ambitious. Unfortunately, the sands of time fall fast in three months.Q: The "FMV" opening intro is pretty sweet for a Game Boy game. How did you manage to fit all the artwork and animations into the game?
Mike:
We requested double the size of the cart. Actually, those animations are pretty small when compared to the size of the in-game animations, thanks to some crafty techniques used by Jeremy Mika to produce the full screen animation. We really needed to accommodate the in-game art first, and the only way to do that was to get the next cart size up from 8 megabit, and that was 16 megabit. So we had some space for the intro, which was actually finished as a prototype for technology and given to Activision and Disney in the first week of the project. All the art and animation was painstakingly painted pixel by pixel by our art staff. Boyd Burgrabbe, our Art Director, was literally working day and night during the entire project.Q: Do you think it's conceivable that we will see Game Boy Color titles in the future that heavily rely on lots of cinematic scenes and FMV?
Mike:
I can't say anything about our plans, but I definitely envision entire games using highly compressed playback. If I told myself that six months ago, I would have laughed. Why would you want FMV on Game Boy? On any other system it is usually a nuisance, but on Game Boy it is almost a miracle that it can be done. We've developed a number of new techniques to seamlessly transition and build drama within our games via FMV. Now we can get up to 20 minutes of video playback on a 16-megabit cart.Q: There seem to be quite a lot of sound samples in Tarzan. We've heard animal noises, drum beats and even a few beats of real music. Considering the cart sizes for Game Boy games are still rather small how do you squeeze all these samples in there?
Mike:
Robert Baffy, our Musical Director, managed to lose 10 man years on Tarzan to produce those sounds. Actually, we knew we wanted a lot of sample playback since we received so much praise for the sound in Klax. However, Game Boy samples must be 4-bit in quality and played back at a set rate. So there is a huge problem when down sampling versus quality, let alone in sheer size required by samples. Robert has a technique of generating samples at 4-bit rather than trying to downsample 16-bit or higher sounds, so the final result in quality is much better. We were able to mix sound effects on the normal channels on Game Boy while producing samples at the same time, and that helped boost the overall soundtrack for Tarzan. And thanks to Game Boy Color, we were able to keep the game fast while processing the sound.Q: Was there anything you had to leave out because the cart was too full?
Mike:
Not really. It was the time frame to finish the game that left a lot out.Q: How important do you think battery back-up or Rumble Paks are for Game Boy games?
Mike:
I think they are very important. Tarzan has rumble support, but did not ship on a rumble cart. If it sells really well and makes Activision a lot of money, who knows, maybe they'll rerelease it on a rumble cart. The fact is, it is costly to manufacture. If Nintendo® ever releases a rumble pack for Game Boy that you can place a cart into, then you will experience the rumble in Tarzan. We had hoped to fully support it, but in the end, mainly the boars and Clayton produce a rumble. Battery back-up would be great, but again, it is costly. I really do hate passwords in games, and I hate having to force consumers to use it with our products. If the Game Boy market continues to grow and game sales continue to explode off the shelf, then maybe publishers would be more willing to pay for it. Or, perhaps, if consumers can make it clear that they are willing to pay the extra dollar or two for it, then maybe they'll raise the MSRP to pay for it.Q: Tarzan is not backwards compatible with the regular Game Boy. What are some of the advantages that you gain by going GBC-only?
Mike:
DMA! The new hardware DMA allows for extremely fast exchange of data from ROM to RAM. We're able to swap in tons of animation on the fly. Also, sound benefits from faster writes thanks to the double speed CPU. All that combined, you actually get more than 2x the speed of a normal Game Boy. At times, you are probably running at about 5x the speed.Q: What do you think are Game Boy Color's uncharted territories? What innovations can we expect in the future?
Mike:
Well, the Holy Grail is fast 3D. It isn't impossible, and there are great strides in this area from developers we've spoken to. We have some tricks up our sleeves, that you will probably see next year. Hopefully, you'll be playing games that will be reminiscent of old FX Chip games on the SNES. It is totally possible.Q: How hard is it to implement Game Boy Printer support? We noticed that you guys pulled it off before anyone else -- in Blitz. Do you have a special connection with Nintendo®?
Mike:
Well, one of our programmers is a hardware guru. Once the printer came out, it wasn't long before he had it up and running. We've used the printer in just about every one of our projects since. Our favorite use is in Klax, in the hidden game 'Developer Dress-Up', in which you get to paste our heads on various bodies. It is what kept us sane for the duration of that project.Q: How did you get into developing for Game Boy in the first place?
Mike:
I started by downloading information from the Internet. There were various documents on Z80, which is almost identical to the Game Boy processor. I started tinkering around with registers, looking into the data in the Game Boy cartridges and along with the help of others (Who now work with us), I was able to get a game up and running via my own Radio Shack equipment. At that time, our Musical Director Robert Baffy and I started actively pursuing Game Boy development and ended up at Digital Eclipse.Q: Do you find it limiting to develop games for a handheld?
Mike:
Very much. There are a lot of things you have to consider when developing a handheld game, primarily resolution and sprite sizes. Converting gameplay from a console to a handheld is a particularly daunting task. What works on a television hardly ever works on a handheld. You do not want to create a character too small to see on the little screen, and likewise, you want to give players enough time when an enemy appears on screen to react. That is why there are so few shooters on Game Boy. Aside from that, working with limitations can be fun when you do something you didn't think possible with the hardware.Q: Are you planning to branch out onto other systems, like NeoGeo Pocket Color or Game.com?
Mike:
We'd like to, but the development environments and third party resourcing is very limited on those systems. We really do like the NeoGeo system a lot, and it is relatively easy to move from the Game Boy to NeoGeo on a technical level. The NeoGeo is definitely more powerful.Q: What's next for Digital Eclipse?
Mike:
Several Game Boy titles, including some original titles fully developed by our in-house teams. Not to mention some unique and original PC and Next Generation console games in the pipe for next year. We don't mind being handheld specialists, but we do have numerous other platforms we're developing for too. However, we do not intend to slow down our Game Boy production any time soon... Let me just say that you'll be seeing many more games from Digital Eclipse this Christmas.Interview by Peer Schneider
