| Game Bytes: An Interview With Vijay Lakshman | |
|---|---|
| (link) | |
| Medium/Format | Magazine Interview |
| Date | November 30th, 1993 |
| Interviewee(s) | Vijay Lakshman |
| Interviewer(s) | David Pipes |
AN INTERVIEW WITH VIJAY LAKSHMAN of Bethesda Softworks
Conducted by David Pipes
ED. - This is the first of a two-part interview with Bethesda recently by our own David Pipes during his recent visit to Rockville to meet with chief game producer, Vijay Lakshman and company president, Chris Weaver. Hope you enjoy the insight into this exciting game company.
Bethesda Softworks is located in the Maryland high tech corridor, not actually in Bethesda anymore, but rather in Rockville. It occupies a suite of offices configured to the needs of the staff, with darkened, no-sense-of-time programming areas; bright, well-lit workspaces for the artists; and an an awful lot of rampant greenery in the buildings' atrium, just right for taking a short breather.
GB: I notice that Bethesda Softworks is a pretty small company...
VL: Yes, we try to keep our project teams to a maximum of 2 programmers, 2 artists and a designer. Sometimes, depending on the project, we might throw more programmers at it, but we like to keep it to five people to a project. At other companies, I've heard of projects that get, oh, anywhere from 14 to 70 people to finish it, and I think that's a little extreme.
GB: How many technical people do you have here?
VL: There are about 12 people. Six programmers, 2 designers, and 4 artists. That's it. And of course various other people who come and go on a part-time basis, just to help for specific projects.
We're pretty casual here. The company is really great about leaving us alone to do our work. We have 24 hour access into the building, and they are pretty cool about letting us work the way we want to. But of course, we as a group agreed that we want to stick to fairly regular hours, because there are so many people who need to work together to produce a game. The artists need to talk to the designers, and so on. So we all try to be here between 9 and 6, no matter what.
It's pretty cool. Right now, our workdays are a lot longer, trying to make sure everything is working right.
GB: This is for a particular game that you are pushing hard on?
VL: This is for Arena. Right now, Terminator Rampage is very close to being done. It's a pretty smooth-running project, it doesn't need much more work. We're happy with that one right now. And Delta Vee isn't scheduled for release until December, so we aren't really under the gun for that one yet. The only other major project is Arena, and since role-playing games take about twice the testing of any other game, because of all the permutations that possible, we're trying to get at least a month and a half of straight testing. So to do that, we need to push and get it integrated soon. A lot of people are in on the weekends and at night, to make sure their modules work.
I really like working here, because it is a small group and everyone is pretty talented. I'm chief designer, and it is good to know that if I think of something, the chief programmer is good enough to make it happen, and the artists are talented enough to make it look right. It's also good that a lot of us are D&D nuts, roleplaying nuts, so that when we talk about what we want to do with a game, we all understand each other. All in all it is a nice group, because we all know each other. It is a small company, we never want to get so big we don't recognize people walking down the hall.
I think it's a good way to go. It is a very Japanese mentality.
GB: Tell us about the technical side of things.
VL: What languages we use depends on the project. I would say that about 90% of Arena is written in assembly language. Julian is very adept at assembly, it is almost a second language for him. He feels that C has too much overhead, it runs too slowly, and so he likes to do things instruction by instruction. That is his philosophy, so the junior programmers who work with him also use assembly, to fit the modules together easily.
Greg, the chief programmer for Delta Vee, is a C nut, so he does all his work in C. Since he's the only programmer on that project, it really doesn't matter. Of course, if something needs to be really fast, he'll code it in assembly, but if it is just a menu or something like that, he can do it in a few minutes in C, as opposed to taking more time to do it in assembly.
GB: Do you reuse a lot of libraries, like graphics libraries?
VL: Oh yeah. Usually our project cycle goes from January to December. Then from October to December, we do library work, updating old code, making sure modules work correctly. Everybody shares the work, because when you think about it, Delta Vee, Rampage, and Arena, even though they are different games, all work off of the same kind of engine. They are like flight simulators, where you are moving and turning in space. The only difference being the addition of pitch in Delta Vee. So the games share a lot of functions and functionality.
All the programmers work on both '489's and 386's, and we look at the code carefully on both. We try to make sure that all our code runs fast on a '386, because while the market is moving forward, we'd like to make sure that all the people who still have '386's are happy. We don't want to leave them behind.
GB: What is your background as Chief Designer? You are obviously not some 20-year veteran manager who has risen slowly to the head of things. (VJ is indeed younger than this interviewer, an athletic type who speaks easily of teaching full-contact Karate, and who has been digitized into several games to date. Yes, that's right, Bethesda has a tendency to turn its' employees into costumed extras at times.)
VL: Actually, I'm just starting on my third year here. I graduated from college with a degree in Computer Science and a degree in English Literature, and I went to work for the Department of Energy as a manager. I was looking through a gaming magazine - I'd never even thought about doing games for a living, I was actually writing a book - I was looking through this magazine because I'm a gaming nut, and I saw this company named Bethesda Softworks. Now, Bethesda's not a name you pull out of a hat - it's a place, so I figured it must be around here. I called them up and said, "Look, do you have any job openings, I'm a writer and I do do this and...". They said they weren't looking for a writer but they really needed a producer. Did I think I could do something like that?
I said "Sure!".
We worked on a few projects, 2300 and the sequel to Terminator, and I sort of fell into the position of designer. We just didn't need a full-time producer; everyone is responsible for getting their own work done on time, and there is only so much to do in scheduling and coming up with lists. But we really needed a designer, and I liked doing it.
As a producer, I talk to the person in charge of programming the project, and we decide on how long the various parts should take. Then I just talk to them every week or so, make sure they are on track, barring unforeseen problems, of course.
GB: What would you recommend for someone who wants to work in this field?
VL: I actually had unique experience, because I walked in with no experience, except as a writer and a manager. I got the job because they needed someone. I sort of hung on to the niche.
I think the industry is only going up, it is only going to get bigger. The biggest things I can tell anyone who wants to get into it, well, it really depends on the field. For someone who wants to be a programmer, learn C and Assembly, especially Assembly. If you want to be an artist, work on a computer. Pick up a computer paint package and get familiar with how to work a compute . You might be able to draw, but if you can't turn a computer on, it is sort of useless.
And if you want to do anything else, like writing or design... The way I hired an assistant when the workload got too much for me to do on my own was to write a very short story, and I had people complete it. I let them have just a few pages, and when I read through all of them, I hired the person who wrote the one I liked best. So I think you should practice writing, read a lot, play a lot of games too.
We'll probably be looking for a new designer in January. Last time we opened the door and looked for a new designer, we got 200 resumes. Out of those, I picked five that looked go, and we hired one of those.
It worked out pretty well, because we need people who really enjoy what they are doing, because it is work. Trying to be creative 100% of the time is tough.
GB: When I talked to Sid Meier, he made the comment that a lot of people who design games for fun should do it on their own, not try to get involved with a company, because once you do you are in the trappings of cooperate culture. You have to worry about what your boss wants you to do, you send ideas out by putting them in a memo, and later you hear that one was incorporated in some game you never even heard of before. It's work, but it is no fun.
What I'm hearing from you is the opposite - small groups, pay attention to people's ideas...What is the difference here?
VL: I think the difference is that here, not to pay anyone on the back or like that but they really do try to hire talented people. Because they assume the people they hire are talented, they trust us, as far as ideas go. Chirs Weaver is unique as a president in that he is wise enough to know when he doesn't know what he is talking about. He has good ideas, and he is a good person to fan a span a spark into a fire.
Generally, a game is created with a one-page idea. Chris says that I can't explain it to him in forty words or less, it probably is not worth doing. I talk to Julian and we try to see if it will be exciting - do we *want* to make a game like this? If we do, then I write the idea up and send it to Chris. It is pretty informal, and he usually lets us go with want we want to do. And we develop a schedule and all that.
Once he has said "If you think it will work, go for it," he wants to know what is going on, and occasionally he has input. He knows when to butt in and when to let us alone. In general the process is very informal.
GB: Since you have been here for about 3 years, then, are the games we are bout to see form Bethesda coming from a different perspective than the previous ones? Are you shifting from one area to another?
VL: Well, put it this way. The new games are from, well, me. I wasn't here when they did Terminator and Wayne Gretzky Hockey. The reason there is a shift is that we've hired someone to think up games. So, we have three totally different games, and each one sounded like fun to do.
Actually, Delta Vee is the end result of about 9 different iterations of a flight simulator idea that never really held. I'd come up with a one-pager, and half of the people would like, half were so-so about it. We were pulling our hair out, because we'd get a few months into an idea, art was done, coding was being was done, but there was no game. People were getting upset because they didn't know what to do. They didn't want to write something that would turn out to be something completely different.
Finally we had a huge meeting, and everyone wanted to know what was going on with the project, and I just decided that I didn't want to do the whole game, which was a role-playing science fiction game called 2300. I was working on Arena, and that was the same sort of thing. It is hard enough to be creative on a game with 400 towns, cities, villages and hundreds of quests and stuff. Take that, and ten you to do one for space opera, and it's like...screw it, we'll do that next year!
So I came up with an idea to do a game that put you in a solar system encased in a globe, like a penal colony or something. People try to run from this field that encases the system to escape through wormholes in the field. Everyone thought that was sort of neat, so we thought about it for a bit, maybe you could go faster as you got closer to the walls, like that. And I took that idea, an wondered what would happen if, in the future, virtual reality was like it is in cyberpunk, if people had personas in a computer matrix. Corporation would hire people to test their virtual defenses, and pay them to make these runs. Forget the prison system, see if you could break *into* the company through virtual representations of data lines. And everybody loved the idea, and boom! we designed it out, and everyone is pumping along on it. It was just more fun.
I think that would never happen in a bigger company. It's not the way we do things.
I do think these games - Arena, Delta Vee and Terminator Rampage - are the best games Bethesda has done so far. There is a lot of talent behind them. We have a serious commitment to bringing the fun back into role-playing, into any game. Not the 350 page manuals, you can now really fly an F-15 sort of game. We wanted to something where you can sit down and just get into the game. If it is a flight simulator, you just sit down and fly. In Rampage, you have a pretty simple objective.
The most involved one is Arena, which is taking the most time. It is a very involved operation, putting together a role-playing game. It is pretty simple with pen an paper, but not so simple on a computer. It is not someone just thinking it up and telling you what you see.
GB: What is the most important element in designing a game?
VL: Well, after everyone agrees that an idea is good, I try to set up a game flow. I try to outline it, make a loop so you go from here to there, from starting to winning to dying. Then everyone can see the screens, this is what can happen, this is what the screen will look like. Do things come up in menus, or just pop up.
Once we have the general game flow thought up, then an artist and I sit down and we storyboard the game. He creates a visual representation, a walkthrough of the game. He Then art can start, since they know what will be needed. We try to tink of all the kinds of things we need to show - objects, places, like that. While this is being worked on, I sit down with the programmer and we try to figure out what is involved, as far was what is needed - do we use polygons or bitmaps? Are are looking for speed? What will I see when I a playing? And the programmer tries to decide on the best methods for brining that vision to life.
That changes continuously, as some technical avenues we take prove to be dead ends. We decide somethign is taking too long, or it won't work the way we want it too, so we backtrack and try another way. And sometimes that changes the design a bit. Like if we can't have 50 enemies on the screen, we have to redesign the one that doubles each time you hit it
While the artists and programmers do all this, I try to flesh out the different screens, put in specifics. What the buttons are, how the character points are alloocated, what kind of weapons are avilable, all the enemies, the damage they do. Within a month we hopefully have a complate design. It always changes, no one ever believres we *have* a design, everyone hates you during the project because someting always has to change. But we have to decide, do we want to do the best game we can, or do we want to stick to the paper? Everyone has their own philosophy. What is on the paper is easiest, but when you look at something that's a few months old, you can tell another idea is much cooler. A good programmer will be able to feel the changes that do make sense, and filter them out and convince you that this is the way you ought to do go.
Later on, as everyone knows what they are doing, peopel will see the effects of some things, and if we didn't foresee them we have to go back and accoutn for them. But as long as people keep the overall vision, it is pretty easy to keep moving forward. The hardest part is to transfer what I see in my head to what others are doing on the screen. But it is fun too.
GB: Are you doing mostly single-player games?
VL: We really want to make Delta Vee a game where you can fly against other people on the network. The game is just made for that - so is Rampage, on or two time-soldiers going through the complex shooting at terminators and stuff. Arena, we'd like to have the first role-playing game where different people can have different characters on a split screen. If you can read Elvish, it looks like English to you, but just runes to me. You can see me, I can see you. That would be neat.
At first we had networking involved in these, but because of time constraints - "About a week before delivery, we decided to drop the direct neural interface" - we had to back off, and go for that in the future, maybe as an upgrade or a seperate package. I think that a lot of companies decide in advance what their add-ons will be, but for us, we don an entire concept, and then as we go along we find we don't have the time, so we spec out an add-on disk. We dont plan them, they just happen. We aren't holding out on people, it just that if they want a game by Christmas, we can't do networking this year. But we are still working on it, and all three games are designed for network play. The hooks are there, we just need time to put the modules in.
(At this point we break to look at the games, or at least what is done that point. Terminator rampage is a glossy, fast-moving first-person perspective of life and death in a Terminator-infested office building. The objects are beautifully digitized even down to the ejection of shells and the destruction of various items of office fruniture and decor under fire from the extremely satisfying variety of weaponry. Added to that is a sound effects library which is quite startlingly real, and an excellent frame rate. This is clearly the next generation of first-person shoot-em-ups.
Delta Vee was still at the prototype stage. It is a gouraud-shaded 3-d flight sim, with again beautifully realistic light-sourced graphics The frame rate was very high, even with a lot of detail, and I would look for it in the mid to high 20's on the average '486-33. It uses bit-mapped polygons and has a very cyberprunk look, like flying through the inside of a large machine or perhaps a spaceship. All I saw was the flight sim part, which was very impressive.
Arena is well along in production. It is a first-person perspective role-playing game, again with very good graphics and sound, and a party to accompany the player. Players will control multiple characters in combat, in real time. They will be able to plan tactical actions and instruct the others to follow through, then watch it happen as they fight. The world is huge, and includes both indoor and outdoor locations, with people wandering around doing their own things. Walk into a house or shop at random; three will be rooms, furniture, maybe people to talk to. One can play the game for a long time without ever entering the plot, like Darklands, fighting in the Arena or doing odd jobs to earn a living. But there is also the traditionally quest-based plot to drop in and out of, with lots of multi-level dungeons, each its' own self-contained world with monsters and puzzles. All first-person perspective, realtime. This is an impressive monster of a game. Not only can you be any of several races and more nationalities, you can even design your own magic spells. There are many other innovations, big and small, in this game, and I for one am looking forward to it.
(Afterwards, one more question...)
GB: How do you handle testing?
VL: Actually, this is the first year we've had three different products to test, so we are learning as we go along. We have a lot of beta-testers come in during the last month or so and put in 8 hour days playing the games and assembling bug lists. Then at night and on the weekends, the programmers clean up the programs, and the next day there is a fixed version for the testers to work with.
GL: Thanks for the insight into a very busy group of gamers, Vijay! I look forward to speaking with Chris!
VL: My pleasure. Thanks for coming over.
This interview is Copyright (C) 1993 by Game Bytes Magazine. All rights reserved.