JOYSTICK magazine: Shaun Southern interview

the big ZOO

SHAUN SOUTHERN
INTERVIEW: DEREK DELA FUENTE

Shaun Southern is the author of Supercars I and II, and the hugely successful Lotus Esprit Turbo Challenge. We went to Wales to worm the truth out of him.[1]

Joy - Who makes up the Magnetic Fields Team?
Shaun Southern - There's Doug Braisby, the boss, Andrew Morris and Jeremy Smith who are graphic designers, Peter Liggett who is a tester and me who is a programmer.

Joy - How long have you been together? How did you meet?
SS - Personally, I arrived there in 1983: I had written games that I offered to publishers through classified ads. It was a company called Mr Chip Software, located in Llandudno, which responded, and which has since become Magnetic Fields.

Joy – How did you start? What games have you written?
SS - I've been programming since I got a Vic-20 in 1982, and at the time I did the graphics myself, which were quite simplistic. Here is the list of what I wrote, but it is possible that I forgot some, because it only took a week to three months to write them: Jackpot. Duckshoot/Kwazy Kwaks, Olympic Skier, Ad Infinitum, Operation Fireball, Formula One Simulator, Hero of the golden Talisman, Tutti Frutti, Jetbrix, Bandits at Zéro, Trailblazer, Kikstart, Dizasterblaster, Speedking C16, POD, Laser-wheel, Arthurnoid, Kikstrat II, Cosmic Causeway , Dingbat and Super Snake Simulator. It is likely that you don't know most of them because they are largely cheap software distributed by Mastertronic, Alternative and others. We released some of them under the Mr Chip name. Others were published by Gremlin Graphics, like Trailblazer which is the best known, but that already dates from 1986.

Joy - What is the story of the company?
SS - Two and a half years ago, we changed from Mr Chip to Magnetic Fields, it was more evocative. This coincided with our entry into the 16-bit market. I had an Amiga since it appeared in 1985, but it was gathering dust, I only had the demo with the bouncing ball. Andrew arrived a year or two before, and started to produce graphics for some of the games I just mentioned, so the first product to come out under this label was truly a team product. There's something I need to explain to you: Magnetic Fields is located in Llandudno, on the North Wales coast. Andrew and Jeremy live three miles away, so they go to the office every day. The office, in fact, is just a room! I live in Oswestry, almost 100 miles away, so I work from home and only go into the office occasionally to discuss ideas with them. Recently, our teamwork has intensified, and I see them about once a fortnight. Gremlin Graphics is based in Sheffield. We see them almost every month to keep them informed of our progress.
Our first 16-bit game, Super Scramble Simulator, was the last one I programmed on the C64; other members of the team were responsible for the conversion to the ST and Amiga. At the very beginning, there was another programmer, Michelangelo Pignani. But he left several years ago to tour Africa. We received a few postcards from various places, but he hasn't returned yet. There's also Dave Making who works with us from time to time, he did the ST version of Supercars and the ST and Amiga versions of Super Scramble.

Joy - Who programmed Lotus Esprit Turbo Challenge?
SS - It was me, and the graphics were done by Andrew. Jeremy, who joined us recently, is working in part on the graphics for Supercars 11. Graphics are becoming more and more important in a game, and more and more time consuming, especially for games like Supercars. For the moment, we have two graphic designers for one programmer.

Joy - Your two most recent games are Lotus and Supercars. Were these projects of yours? How did you sign the license for the Lotus?
SS - Supercars was indeed our idea, but it was still inspired by several other games. When it was brought to Gremlin Graphics, it was almost finished. They simply suggested some changes and edited it. For Lotus, actually, we wanted to make a two-player car racing game, like Pitstop II, so we did some graphics, and while I was helping Dave converting Supercars to ST, I developed the routines that were used to make Lotus. We brought the game to Gremlin. who we didn't know had signed a license with the Lotus brand. So they asked us to transform our game into a game based on Lotus. And again, Gremlin suggested improvements and ideas throughout development, but to the extent that it can be considered an original idea, yes, it's our idea.

Joy - Why did you choose Gremlin?
SS - I wrote a game called Trailblazer a few years ago, back in 1986 I think, and we sent it to several publishers to see what they thought. Gremlin responded, was interested and our alliance was born.

Joy - As a development company, are your relationships with your publisher close, and is there a limit to the changes they can impose on you?

SS - Like I said, Gremlin generally leaves it up to us, but there are times when they intervene a lot. For example, I came home one day from a well-deserved holiday to find a list of over forty requested changes. But most of the time, what they suggest only improves the game. I think we start to know what they expect from a product, and they know that we are capable of achieving it, so relations are becoming easier and easier between us. Obviously, we respect each other, and we talk together quite a bit. For Supercars II, we had a meeting before starting the product, to avoid wasting time on things that would later need to be changed. But again, it's our idea with some improvements suggested by Gremlin.

Joy - Did the people at Lotus see the final product and were they happy with it?
SS - We did a product launch about two months before release at the Lotus premises in Norwich. They allowed us to drive a Lotus hard on their private circuit, and gave us a guided tour of the factory. Part of the licensing agreement was that the game had to be approved by the Lotus board, and so we actually informed them about the progress of the game. And yes, I think they were happy. I don't know how much the staff there liked video games before, but I was told that they spent their time playing them ever since. There was also a funny thing, which is that one of the options screens shows a Lotus seen from the side, and Andrew had copied it from photos that came from Lotus itself. And while everyone was raving about the graphics, there was a mechanic who came in, just took one look, said “the wheels are upside down” and left. And indeed, the hubcaps were designed upside down.

Joy - Were you provided with all the Lotus specifications and were they difficult to implement in the program?
SS - Lotus Cars sent us loads of details and we found photos of their cars in a garage in Chester. The options screens show just about every detail, but since it's just a game, some have been left out. We were offered data that would exactly simulate the handling of a Lotus, data sampled from a real car in motion. Changing gears, braking, steering, etc. But I repeat, it's just a game. The respect for the revs for the gear changes and the top speeds are faithful, but we had to modify the acceleration factor, multiplying it approximately by three. When you sit in the car, it seems very fast. On screen, if we respected the real speeds, we had the impression of dragging ourselves. The impression of speed now really sticks to reality. That is to say that 150 km/h on the screen gives roughly the impression that we have in reality at that speed.

Joy - To write a racing game, do programmers need to know how to drive, know mechanics and have some knowledge of aerodynamics?
SS - Well, I don't know how it goes for other programmers, personally I know how to drive (although this opinion is not shared by all my friends), I know enough about mechanics to do the mathematical calculations necessary for the program. There are no aerodynamic calculations in the game, but that might be a good idea.
As I was programming Lotus, I found that my actual driving tended to resemble more and more the way I played my own game. Very dangerous. Right now I'm writing Supercars II. and I spend my time making controlled slips in reality, because there are some in the game.But I'm not yet jumping over roads, fortunately! In case anyone is interested, my car is actually a truck, a sort of tow truck. Rear wheel drive, and fast enough for me. Let’s stop saying that programmers all drive Porsches (or Lotuses!).

Joy - Have you done a lot of research into the handling of the game itself?
SS - It's simple, if we had a gearbox, a steering wheel, brake and accelerator pedals, it would be great. But we only have one joystick. Up, down, left, right and fire. This means that either you're trying to write a simulator, and any pilot trying a joystick flight simulator will tell you, it's nothing very funny, or you decide to write a game, and then it becomes funnier. If you look at the car when it is turning, the wheels are not turning, the whole car is turning. And in this game it can be interesting to brake in turns, for example, which is far from the case in reality. In fact, I favoured the fun aspect of the thing, and I tried to copy (please don't sue me) the handling of Super Hang On. I know it's a motorcycle, but hey, apparently it works. It gives the desired effect. We always have to make a compromise between realism and playability That said, I think that playability wins.

Joy - Do you program on ST to do the port afterwards, or do you redo the program for each machine?

SS - Programming on the ST? Nah... I have a nice Amiga 2000 at 25 MHz with 5 megabytes of memory and a 40 megabyte hard drive (chuckle giggle). It is a recent acquisition. We only got the hard drive two thirds of the way through Lotus, but I couldn't live without it now. I programmed Supercars myself on the Amiga, and then I ported the code to ST, but most of it was re-written (because my program was so messed up that the other programmer had goose-pimples when he saw it).I wrote Lotus again on the Amiga. using azu maximum all coprocessors of this machine[2].
Then I wrote a version on the Amiga which did not use these coprocessors, and it is this version I ported to ST, and Dave helped me get it in shape and running. This is probably what will happen in Supercars II. All graphics are developed on Amiga and then converted to ST. Our policy is to make the best possible version on Amiga, and then to make the best possible adaptation on ST, at least a version whose handling is exactly the same. Graphics are less important.

Joy - Why did you choose to make a sequel to Supercars?
SS - We wanted to put in everything we couldn't put in the first one, and thanks to the success of Supercars, Gremlin gave us the green light. We also tried to include or correct all the criticisms that were made by journalists on the first one.

Joy - Are there certain facets of programming a car game that make it different from a classic shoot'em up, for example?
SS – Yes. The speed must be constant. This is an annoying feature of some racing games: when there's a lot on screen, and the same can be said of some Shoot'em ups, everything slows down. On Lotus, I used a sort of real-time system, where all movements are taken into account in real time while the screen is only refreshed when possible. Of course, it takes more memory, but the movements are as smooth as possible. This also means that the ST version is just as fast, the only difference being that it is a little more jerky.

Joy - How do you go about writing Supercars II? Can you describe the chronology of events for us?SS - Actually, the scenario is already done, but we drew the cars first, then we redrew them. We added everything we wanted to add, such as the two-player option, for example, and then we drew the tracks. What we now have to do is enter the track data, create the intermediate screens, create the championship part and finally the opening sequence of the game. All we will then have to do is test, test, test and retest.

Joy - Many people consider sound effects to be easy to make. Is this true?
SS - In Supercars, Ben Dagleish did the sound and music. In fact, we simply asked him for a certain number of pieces, knowing that we would reserve the best for the credits, and a certain number of sound effects that I simply placed at strategic moments like crashes, rollovers, etc. In Lotus, we were a little more ambitious, we included the sound of squeaking gears, etc. It was friends who gave us the sounds and since then they have started writing games, under the name Broadsword Software. We tried to put a Doppler effect on the sound by making phase shifts to give the impression of a car approaching then moving away, but during the two-player game, playing on four channels simultaneously, that takes too long. For Supercars II, I plan to use the Amiga's stereo capabilities to the fullest. I will try to put as many sound effects as possible. I would really like to take advantage of sound in a game, so much so that I would like to write a game in which sound is a crucial element in completing it, not just a gimmick added later to make it look nice.

Joy - What is the most difficult thing about programming a game?
SS - Well, I want to say that that's the initial idea, although when we look at the games we've made so far, they're general themes that we just stick with. To bring it up to date. This game is the most difficult, it is perhaps bringing all the little bits of the program together. And also to ensure that the player is permanently hooked, that there is no downtime in the game. We test a lot ourselves, not to look for bugs but to see if the game remains interesting to play. long term. Programming itself sometimes creates problems, but most of the time there are always simple solutions. Displaying a sprite quickly on the screen actually only consists of calculating the timing of each instruction and optimizing a bit of routine. I spend more time creating race courses than writing display routines.

Joy - There have been tons of car racing games released recently, both simulations and conversions of arcade games. What do you think of Nitro, Jupiter Masterdrive...?
SS - Nitro, uhhhhh... I admit that I haven't seen it. Jupiter Masterdrive, I saw a demo on ST. It looked fast, but I prefer the handling of Supercars, with the controlled skids and handbrake turns. Badlands is a strong competitor to Supercars II. I played it in arcades once or twice and it looked really good. I don't really understand why it got less good reviews, maybe people are starting to get tired of the genre. Indy 500, I read the reviews but I only saw it running on a PC. It's very pretty but a little too realistic for me. I generally prefer games over simulations. Iron Man, I've only seen it in arcades and I found it a little easy, but I'm impressed by the quality of the conversion on micros. The truck graphics are fantastic. Again, I believe this is a serious contender to Supercars I and II.

Joy - And what do you think of 3D vector simulations?
SS - Vector simulations? There is a very clear speed limit. Maybe I'm being picky, but when a screen takes more than four frames to refresh, it's too choppy. And the 3D graphics aren't very detailed, and unless I'm sitting in an arcade booth with a huge screen, I get claustrophobic looking at the monitor. So far, no computer game has clicked for me. It's always too slow or too jerky. That said, my favourite arcade game is Winning Run. This is what I call realism. But then again, having real controls like brakes etc. helps a lot. Like sitting inside a cabin with hydraulic cylinders and surround sound.

Joy - Do you think the public is tired of this type of game and what makes Supercars II special?
SS - I think the public will stay interested as long as each game that comes out is a little better than the last. Supercars I was already nice, even though there wasn't much in it, you couldn't create your own tracks, for example. And it’s this extra dimension that will make it special. We tried to make it as humorous as possible, it can't be bad.

Joy - What does a beta tester do on a game, and what kinds of bugs can appear in a racing game?
SS - Uhhhh... Beta tester? I don't know the term, a friend who studied computer science told me that it consisted of following all the possible paths that an algorithm can take. If this is indeed it, we are not doing a beta test. We just play to death. Many of my friends test my games, because they play differently and therefore they find different bugs. Gremlin also tests all versions before they are sent for duplication, for over thirty hours, even though I have changed almost nothing since the last version they tested. In Supercars II, the main bugs that gave me problems were some cars sticking together after a collision, or getting stuck in walls. But most of these bugs usually come from stupid errors, like typos.

Joy - What do you think of the console market?
SS - Doug says he's watching this very closely but we don't have any immediate plans. I'm very impressed with the Megadrive, but I hope the Amiga maintains its place in the market. It remains an astonishing machine. Computers are much more versatile, and games too, it seems, tend to move away from simple right-left movement.

Joy - What are your favourite games?
SS - Tetris (by the way, one of my cats is called that, and the other is called Monster), Super Pipeline, Sentinel (the creepiest game ever, because of the music when you die) , and all that kind of games. Populous is very popular with everyone I know. I would love to write a Shoot'em Up like this, in which the aliens could learn from their mistakes, reproduce, learn attack formations, etc. Something that would surprise even me when I play. But until that exists, I prefer two-player games, because you're never sure of the other's reaction. I also play my own games, at least until I finish them.
I played Lotus for two hours every day, until I finished the last circuit. I crossed the finish line in 10th place, while listening to Ace of Spades by Motorhead. The perfect atmosphere! But I don't think I've played it again since.

Joy - Do you think Supercars II was disadvantaged by the fact that it wasn't a movie license or an arcade conversion?
SS - For the first, it may be true, but I think that the Amiga market is very adult: in fact, good games sell, while bad licenses sometimes fail.
Supercars II is already benefiting from the success of the first, and Lotus' success can't have hurt it.

Joy - Are there any arcade conversions or films you would like to work on?
SS - Andrew says he wouldn't like to do an arcade conversion because he wouldn't have any freedom in the graphics. But a movie could be fun.

Joy - Who do you admire most in this profession?
SS - I hardly know anyone, but when I had a C64, I really liked Jeff Minter's games.
I don't know what he's done since then, other than his latest game, Defender. I find it relaxing to think that some things will never change.

Joy - What are the main limitations of the ST and Amiga, apart from memory?
SS - On the ST, the lack of blitter is the worst limitation, resulting in it taking a long time to display graphics on the screen. The capabilities of these machines require the use of large and colourful graphics. But that takes time.
Future Amigas may have more coprocessors and a faster clock speed. The technology exists for arcade games, and I hope it won't be too long before we can enjoy it at home. One of the limitations, if you can call it that, is that people buy arcade conversions when we know full well that they cannot, no matter what, be as fast and as beautiful as the 'original. So obviously, the developers remove bits here and there, and the games suffer. It's the same with 8-bit games trying to copy 16-bit games. Most of the time, this is simply impossible. We should take advantage of what the machine can do. It's the same for ST and Amiga: they are keyboards and mouse.
How many arcade games have so many? We should take advantage of it, otherwise it's wasted...

Joy - Do you think that we are far from having exploited all the possibilities of the Amiga?
SS - Well, if you look at the difference between the first games and the ones now, you will see that the graphics have improved enormously. It takes a lot of time to find new techniques and to learn how to exploit coprocessors and the architecture of a machine. That said, I think we're starting to get really close to the computer's maximum potential. I don't think Lotus could have been faster; at most, we could have added a few functions. When it comes to gaming enjoyment, the only way to improve it is to study the coolest 8-bit games. I didn't think, at first, that I would need the 512 KB of RAM, but you see, there are more and more games that only work with a mega. For now, the only thing I could do with all that ram is speed up loading times. Maybe one day we will see games that will only work on 25 MHz Amigas with 5 megabytes of RAM?

Joy - What are your next projects?
SS - We're not there yet, but I think there will be a Lotus II. And then, I would like to write a Shoot’em Up, please...

Joy – Thank you.

The original French language interview is here.

[1]  "Pour lui tirer les vers du nez" literally means to pull the worms out of his nose which is quite an image.
[2] I think the translation goes a little squiffy here but I don't know enough about the Amiga to work out what it should say.

Click here for the Mr Chip/Magnetic Fields article.