Hello again web browsers! Today I’m bringing you a long-due but very interesting one: Last year I happened to reach out to Robert Baker, one of the developers of the underrated gem that is Rollcage (which you might call “Wipeout on Wheels”; but does some more than just that), as well as being part of developing LEGO Racers 2 (which was a project given to Attention to Detail, where he used to work) and the fabled LEGO Rocket Racers, an exclusive game for the long-defunct LEGOLAND Windsor park that used Rollcage’s engine as a base.
On top of that, he also developed modern patches for both Rollcage titles called “Rollcage Redux” and “Rollcage Extreme“, which allow you to play the games in modern Windows and even with online support. As if his dedication to this classic wasn’t clear, he also is the lead programmer in GRIP Combat Racing, a spiritual succesor/remake of the aforementioned Rollcage; complete with a slew of new additions and a shiny new coat of paint!
Regarding LEGO Rocket Racers, I had been trying to write something about the attraction and the findings related to it; however a recent video by Coaster Connection did a brief but comprehensive insight of everything that was known about the attraction before it closed. However, it is definitely an interesting piece of hardware for its time, and there’s a good understanding of how it went between the video and this website (where some pictures were borrowed from).
Today’s interview with Rob gave quite a lot of interesting insight not only in the development of these games (Rollcage which released back in the 90s, GRIP which released not too long ago, and LEGO Rocket Racers which was such a mysterious game that had vanished at one point with the park itself), but also in his experience on game development, the way it was around that time, and his personal experiences since then. It really was a pleasure to have this opportunity with someone like him, so I hope that you enjoy this as well!
This interview was conducted around June 2020
How would you describe yourself to begin with?
Robert: Well, I live in Indonesia, in a wooden house built on the beach of a small island in North Sulawesi. I work from home, which is great, though it can often be a bit of a burden as there doesn’t seem to an off switch like there is when you work in an office. But certainly, I wouldn’t trade working in an office for this. I like the quiet of the beach and having no real neighbours, that helps a lot with the concentration. But I love music too and am beginning to get into creating some of my own in between my work projects.
Right now I’ve just come to the end of development on GRIP and have started to develop a game programming course for udemy which I hope to have completed in a couple of months. Apart from that, I love getting out onto the reef in front of the house here, and exploring the wildlife there – it’s just stunning. I have a couple of turtles out there that I call friends.
What would you say that are the type of games that you like the most?
Robert: FPS, for sure. But I like anything with an interesting storyline or art-style. Limbo for example, a great looking game which was so different to the norm. Loved it. Ironically I rarely play driving games, though I must have written half a dozen of them over the years.
What was you first experience with consoles?
Robert: I guess that would be the Sega Megadrive era. I was always a bit of a home micro guy back in the 80s and had kind of grown up and become a father and husband in the 90s, so I thought I had left gaming behind me. But with PCs becoming more and more capable around 1993/1994 with the arrival of DOOM, and the MegaDrive that somehow gave home gaming a bit more adult appeal compared to the Master System, I found myself getting dragged back into it. Certainly, I do remember having a lot of fun with Ecco the Dolphin, Sonic 2 and Desert Strike. But to be honest, my step-son pretty much dominated the console in our house, and rightly so – he was at that age.
How did you end up involved in game development?
Robert: Jetset Willy, DOOM, Quake – in that order. As a teenager I just wanted to create the kind of odd games that were then out there, the quirky Jetset Willy in particular. I didn’t have much money though, and only managed to buy a stock Spectrum 48K. Developing on it using only a tape-deck for storage was painfully slow. I did manage to write a couple of games, neither of them published, but it was a good learning experience into assembly language.
I then quickly settled down after leaving school with a wife, and a child, so I had to get a proper job. Which I did, programming business applications. Games programming was then relegated to just a teenage fantasy. I was still involved in graphics and audio programming in my spare time, the type of coding you need for games. But it was still just tinkering. Then DOOM got released, and everything changed.
DOOM really, really made me want to get into games properly. It showed me just how immersive games could be compared to what went before, it was the dawn of a new era. And then with Quake not too long afterwards, that really was it. I wanted to code something as technically impressive as Quake, it was massively inspirational. I was already on the mental road to changing career at that point – after DOOM was released I started doing a lot of research into graphics programming. But when Quake came out, I knew I needed to do something about it.
So I did. I took a chance and left a well-paid, sensible job in the programming sphere, and threw it all away so I could follow that dream of making video games. It was a huge decision – I had a wife and two children at the time. But it paid off. One of the better decisions I’ve made in life, and one I am so, so glad I took. It’s too easy to live a comfortable, but boring life.
How did you end up working in Attention to Detail?
Robert: Completely by accident, it should never have happened. I was in a bad situation with a so-called friend of mine where he was funding a game development effort we were both involved in. He went very quiet only two months into it and it took me a few weeks to get a hold of him only to hear he had run out of money – two weeks after he was supposed to pay me my second month’s salary. Needless to say, at that age I had no savings and now I was in debt and out of work. Some friend.
Work was scant at that moment and I needed to find something, and fast. Just about the only suitable job that came my way was with ATD, working on a driving game (a genre I didn’t particularly like), simply porting it from PlayStation to PC (which really wouldn’t have stretched my abilities). ATD was also an hour’s drive from where I lived, and I didn’t even drive at that time. It was not the most auspicious of beginnings.
I did learn to drive though, and within a week of joining ATD I had passed my test and bought a car. What I didn’t know back then but did learn subsequently was that I got a particularly poor reference from my last former employer. That caught me by surprise. I asked my then boss why the hell did he employ me if that was the case? He said he was just desperate for staff at the time and took a chance on me. It seems we were both in a jam and needed each other, even though neither of us was particularly what we wanted.
However, never judge a book by its cover. My boss said my former employer couldn’t have been more wrong after he got to know me, and he was glad he ignored the poor reference as it was just bullshit. I ended up as Principal Programmer at the company by the time I left seven years later, and it turned out to be one of the best places I have ever worked, with Rollcage right up there at the top of the most enjoyable games I’ve ever worked on. My boss there, still a good friend to this day.
What was the first game you worked on?
Robert: Guts n Garters in DNA Danger. It was just about as shit as the title suggests. I only really came in at the tail-end to help with the UI and installer. It was the company’s first, and last game. I also developed a 3D software renderer for them also during this period, for what was to be the company’s next game. The renderer itself was pretty much complete by the time I left just a few months after joining (because the MD was a disrespectful and overbearing arsehole). Considering it was my first job in the industry writing a renderer was a real stretch and I was very proud of it at the time. Interestingly, this same renderer was used in the Sydney 2000 Olympics game that ATD produced for PCs that didn’t have hardware acceleration. I would love to go back and write a new one today using all the experienced I’ve gained in the decades since. A completely pointless exercise of course given the ubiquity of GPUs, but it would give a lot of personal satisfaction.
What was the first game you worked on as part of Attention to Detail?
Robert: Rollcage. In particular porting it from PlayStation to PC. Much of that work was done very early on though and I found I had spare capacity to do some more general things in the game in later months too. I joined just as the new tech for the game had come to a critical mass and they were just starting to turn it into a game. So not right at the start, but pretty close to it.
What was the main idea that inspired Rollcage’s creation?
Robert: I think it literally was going to be Wipeout with wheels in its initial incarnation. It wasn’t particularly innovative in terms of ideas, though the tech we produced was progressive at the time. The real change came when we signed with Psygnosis and was assigned a producer in the shape of John Meagan. He was pretty cool, and it was his suggestion that we gave the vehicles massive, fuck-off wheels which allowed them to drive on walls and ceilings. Rollcage in essence, is down to him.
What were your best experiences when you developed Rollcage?
Robert: I would say when the game started to become playable on PC, that was kind of like magic there for a few days. Taking that PlayStation game and making it work on completely different hardware. It was bit like archaeology really. I personally didn’t really build anything, I just uncovered it. The work I did was uncovering this game on a new platform. It felt a bit like witchcraft, slowly finding the elements of the game other people had written wrapped up in my own work.
But also when we started MP LAN testing towards the end of the development, that was pretty special too. The PlayStation version had been finished a few weeks or so by then and so the team was quite relaxed. My time on the project was also coming to end after a lot of hard work. All that was left to do was get the MP game working solidly. So it was a happy time, testing and fixing, but mostly just enjoying playing it with your friends and there was little to actually fix as far as I recall. It felt a bit like the last day of term at school where you wear your normal clothes and play games all day. Technically we were at work after we had all worked so hard, but at that time we were just playing really. Dream job if you think about it.
How would you describe the relationship with Psygnosis on that time?
Robert: Very good indeed. Though personally I wasn’t really involved with them except for the odd brief conversation with the producer we had who visited from time to time. But I do know from talking to our internal producers that we had a great relationship with Psygnosis.
What did you know of Rollcage’s reception at that time?
Robert: Oh! Rollcage was received exceptionally well by the gaming press. I have so many magazine previews, reviews and articles about Rollcage that could only speak of it in glowing terms. I think we managed to bag a couple of magazine covers on PlayStation magazines at the time too. You’ve got to be something pretty special to accomplish that. I remember relating to other team members, that although this was very early days in our careers at that point, it probably won’t get any better than this for us. We had better relish it, because likely much of our careers won’t be this exciting. And indeed time has proved that for the most part they weren’t; that is pretty sad. The glowing status that Rollcage received in the press I’ve never experienced since. But at least were there, at least we tasted what that was. For most people in the games industry, that never happens. I certainly look back on those days fondly, and with gratitude.
The original PC version had network support. How did you get to test that feature?
Robert: Indeed. How did we? Networking in those days was nascent to say the least. We definitely tested LAN play internally, and that was fantastic. I remember testing serial cable play briefly but not at great lengths, we felt this was the least likely to be used by the public. Internet play was never tested at all by ATD as we didn’t have the infrastructure for it (the company itself had to time-share a 56K modem to the outside world – there was no broadband back then). I was told Psygnosis did test it, but I’m not at all sure that they did. But yeah, testing the LAN multiplayer was really, really good fun. We had a great time in the office, often after normal working hours, which was just as well the amount of swearing we would do when robbed of a win at the last moment.
Did you get to play the original Rollcage version online with others?
Robert: No, not at all. Even when I produced Rollcage Redux many years later specifically for improving the MP experience I never tested it with real players – I let the Rollcage community to do that for itself to be honest. I’m not really a multi-player kind of gamer these days, I much prefer single player games.
How would you describe the process or experience behind porting the game to PlayStation?
Robert: It was written on PlayStation and ported to PC. The experience was certainly a learning one. Looking back, I could have done a better job for sure in terms of making it cleaner. But, it was all about the performance back then. The relatively dirty approach we took was down to trying to keep things working as fast as possible, and with least impact to the writing of PlayStation game code. Coders on that game only wrote for PlayStation, so we were expected to just recompile it in the PC environment and have it “just work”. I would do things differently now, trade a bit of performance for cleanliness, but we still did a pretty good job regardless.
For the first few days my screen was filled with nothing but compiler errors. Taking PlayStation code and trying to compile it on PC is never going to work of course. It was a matter of figuring out which PlayStation functions and data structures we were actually using, and then writing support to replace those with equivalents that would work on PC. To start with, there was hundreds of these references to be replaced. A few days later I got that number down to just a handful. A day later I actually saw some triangles rendered on the screen. That was pretty special. There was no textures and not even any clipping either so the screen was a mess. But if you looked hard enough, you could see some recognisable geometry in there somewhere. It was a happy beginning.
For the next few weeks it was all about getting those triangles to look more and more like their PlayStation equivalents, getting all the right data in the right place at the right time. And even in those early days, it was about performance optimisation too. Unbelievably, a 200Mhz PC with a 3Dfx card was struggling to keep up with the 40Mhz PlayStation and its GPU. There’s a lot of assembly code in the original Rollcage for PC because of this.
Then came things like input controllers and audio.
Last I think was networking. We put quite a lot of thought into this and did a lot of research. In the end though, as far as I remember, despite our many tests with different algorithms for prediction and the like, pretty much the simplest solution turned out to behave the best on-screen.
What would you say of the PlayStation development scene at that time? And how did it work?
Robert: I honestly don’t know much about it. I know bits, but not enough to speak with qualification. I’m a PC guy.
It seems that a sequel to Rollcage was a no-brainer after its release.
Yeah, it kinda was, it was relatively plain sailing.
What can you say about Rollcage Stage II development?
Robert: To be honest I was never supposed to work on Stage II. At the time I was writing new tech to be used later in Lego Racers 2, though at that point it was being developed for a completely different racing game that we ended up not pursuing. I’d been doing that for just a few short months when I was asked to join Stage II because they were struggling with it due to lack of experience of the coders on the team. I didn’t really want to if I’m speaking truthfully, Rollcage, sadly, had cost me a long-term girlfriend so I was carrying that association around with me. But also I was actually enjoying working on clean, new technology. To go back to Stage II and repeat myself, reminding myself of my loss, wasn’t something I really wanted to do. But you know, gotta do what’s good for the company. Little did I know Sony would strangle it at birth anyway.
But yeah, we spent a fair bit of time cleaning up the code, the kinds of things I said I would have done differently earlier on. But from then on I don’t really remember writing anything specific on that game, except for a few graphical effects. I do remember doing the time-warp and shield effects for sure, oh and the credits sequence too. But really I was responsible for gluing everything together and making builds, outside of that I can’t recall much else apart from polishing the main user interface.
What extra things were planned for Rollcage Stage II?
Robert: I’m not sure there was anything that we didn’t implement. I could be wrong, I wasn’t involved in the early stages. For me though, the overriding impression was just to produce a commercial sequel within a strict budget. Rollcage we had already seen had underperformed at the cash till. For whatever reason, the excitement that the gaming press had shown the game wasn’t shared by the buying public. It did OK, but only OK. Stage II attempted to correct some of the mistakes we felt we had made with Rollcage and make it more appealing, but to me that actually made it the less exciting game, it emasculated it.
Was it easier or harder to port Rollcage Stage II to PlayStation than the first game?
Robert: It was written on PlayStation and ported to PC, which is why it looks like a typical PlayStation game. The porting though was easy, we really did benefit from all the work I had put in on the original Rollcage to make it dual-platform. As I said, it was plain sailing.
Did either of the games have any planned ports for anything other than PC and Windows?
Robert: No, not at all. Interestingly though, Ducati World which used the Rollcage Stage II engine was ported to Dreamcast.
I’m aware that ATD worked on the Lego Racers sequels; but also, between Rollcage and Lego Racers 2, there was a Legoland game called LEGO Rocket Racers; which used Rollcage Stage II’s engine. (almost all info comes from here)
How did ATD get involved with LEGO about that in first place?
Robert: We were already working with Lego for Racers 2 at that point. They knew our history, and what we could do, so they asked us if we were interest in doing Rocket Racers for them.
22. Outside of programming the game, did you have any other role in the creation of the ride?
Robert: Yeah, I was the guy that went down to Windsor a few times to raise the game up from the single-player game that we tested internally to the full multi-player game that it was always meant to be. All the hardware was down in Windsor, and it was the only place we could do that work. That said, I prepared for it well and actually the integration wasn’t so very difficult.
It was during that testing that I had a kind of unique moment in my career. If you can imagine that certainly back in those days, developing games was quite a closed affair. The only real feedback we had on the work we did was via reviews in magazines, but these were written by journalists, and not our actual audience. We didn’t really know how the public saw our games, or how they might have been affected by them at an individual level. It was all quite impersonal. We created games and put them out into the world, but what the world made of them we didn’t really know.
With LegoLand Windsor however it was different. Just as the software became beta-testable we started to allow select groups of families into the event to see how it held up. The lighting in the racing rooms was quite low and what light there was was centred on the vehicles the kids were driving. I was holed-up sitting on the floor in a dark corner of the room, with a monitor in front of me showing me what was going on with the server software and all the technical data. Nobody really noticed me sitting there in the dark, their focus was on the racing.
Oblivious to my presence, for the first time I heard kids talking about a game that I helped to make. Just little things you know, like how they loved a particular weapon, or the volcanic fireballs in the level, or that they just wanted to go around again because they enjoyed it so much and wanted another go. Little kids, you know, with excitement in their voices. It was really nice seeing something that I had helped to bring to life bringing some happiness to their lives, even if just for a moment. It was very gratifying and I felt privileged to have been there, I only wish the others on the team could have been there too but I was alone in doing the commissioning of the game sadly.
What can you comment about that ride at the original run in 2000-2004?
Robert: The Lego Racing attraction at Windsor was an interesting project. It came along at the time we were developing Lego Racers 2 and I had to be taken off it for a few months, again. It involved repurposing the Rollcage Stage II engine and creating a couple of Lego-themed levels out of it, complete with an all-new UI that allowed you customise your character and vehicle. The game itself was fairly easy to build, I think it only had 3 or 4 weapons and these were quite simple by design. The big problem was the networking of it, and this is what fell to me. The attraction worked by running two games concurrently. Where one game was in session with kids racing each other, another game would be being setup with the next set of kids designing their car and character.
There was a central server co-ordinating 3 different lanes, so this is 6 concurrent games. There was also the capacity to add another 3 lanes for the second level we had designed for the game. In the event this was never actually used and they just stuck with space-themed level. The thing with the networking was speed. This wasn’t a loosely coordinated network game, it was very tight. We had to effectively ditch DirectPlay which we had used for networking in other games until that point as it just wasn’t fast enough. So we switched over to plain old network sockets and used message broadcasting without any acknowledgements. This got us the speed we needed but wasn’t the end of our problems. We had to account for packet loss and timing variations in games too as clocks drift across machines. That sounds trivial, but it wasn’t. Initially the game was quite sensitive and took a while to settle down. It involved a fair bit of just sitting there watching it and looking at the data to detect issues and formulate fixes for them. But we didn’t have to fix much, just 3 or 4 things from memory. Then it was solid, for years.
I don’t really know much about how it all went after the commissioning was complete. Of course I went back to the Midlands and continued working on Lego Racers 2. I assume everything went well because they never called us back to fix anything until much, much later.
How much was involved to rework Rollcage Stage II into LEGO Rocket Racers?
Robert: Actually, it pretty much was Rollcage under the hood, just with a different veneer. Most of the work involved in developing the game centred around the vehicle/character building UI written in a completely different engine framework, and the networking which was all new. Less strenuous was the new weapons and vehicle handling.
I can’t remember how long it took us exactly, but I think it was on the order of three months. Three coders worked on it including myself, one for the UI, one for the new game content, and myself for networking and general infrastructure.
What did you think of the idea back then?
Robert: I wasn’t terribly happy about being taken off writing new tech again, but once I got into it and sunk my teeth into getting the networking right it began to warm on me. Of course though, it was a kid’s game and it’s difficult as an adult to get too fired up about that. But like I said, when seeing the effect it had on small children that was when it hit home and became relatable. After Rollcage, the game itself all felt very tame as it was designed specifically for small children. It was only when the kids actually got involved did I see that it was in fact a good idea.
After the original ride closed in 2004, you were later contacted in 2009 in order to work for the restoration of it. What can you say about that?
Robert: Yeah, that was rather unexpected. ATD had long since closed and working on Lego games was just a distant memory. To be asked to work on it again was a surreal experience, outside of its time and place. Something from the distant past reaching into the present.
In all honesty though, I can’t remember what it was I actually did during that time. I think it was just about getting the game to work well with modern Windows and also work on hardware that was different to that which it was originally commissioned with. So very likely it was all about upgrading the APIs and adjusting the timing code to be more flexible. But truthfully, I don’t really recall it very well at all. For me, the thing I remember the most was meeting up with my old friend James Dobbs again. He really was the heart and soul of that game down in Windsor.
What can you comment about the other staff you worked with for that?
Robert: I only really work with one other person on this, my main contact at Lego – James Dobbs. He was great. A real company man, totally all about Lego and the attraction meant a lot to him. I believe his wife worked there too. With every forward step we took with developing the attraction he got really excited.
And of course, the other coders I worked with at ATD I knew very well indeed. Our UI guy was an exceptionally quiet Scottish man, Geoff Browitt. I believed he left the industry after ATD to pursue a different career. The other guy is a good friend of mine, Lyndon Sharp, who I worked with again many years later at a different company. Both were solid to work with on this project and we got it nailed pretty quickly.
What was different in that 2009 re-run?
Robert: Nothing discernible as far as I recall. It was just engineering.
There seemingly were two set of lanes; one on the upper area and one on the lower area. One of those used a different track?
Robert: No, as previously stated, there was 3 lanes on the lower floor and there was due to be another 3 on the upper floor but as I far as I know the upper floor was never opened. The upper floor was to be used for an Egyptian themed track, but I don’t think this was ever seen publicly.
What can you comment of the assisting programs for controlling the whole system? (Startup Console, etc)
Robert: Wow, I’d actually forgotten all about the Startup Console. I just dug around my source code here and found it. Yeah, it was something I had written. I’m guessing it was running on the host machine for each lane. More than that, I have no recollection at all. You see, this is what happens when you get to 50.
Do you still have any source code of that one (or any tools to edit the tracks?)
Robert: Yeah, I have all the source code for the game and the runtime tools. You never know when LEGO might be in touch again, although probably looking for something a LOT more modern next time. I have nothing for track editing, for that we used the standard Rollcage track editor which I was never involved with.
Do you still have tools for editing Rollcage tracks and the sort?
Robert: No, there was only ever one coder who was involved with that, I don’t think I even used the track editor the whole time I worked at ATD.
What inspired you in working on the patches for Rollcage (Redux) and Rollcage Stage II (Extreme)
Robert: Well, I got an email out of nowhere sometime around 2014 asking if I could help in getting Rollcage to play nicely with modern Windows. It was sent from the head of the Steam Rollcage group and apparently there was a whole bunch of guys still playing this game online. As I still had the source code and because Rollcage was a particular favourite of mine, I thought why not? The idea then of polishing it up more was a natural one and I really did enjoy doing that for a while. It was nice seeing the game getting some attention again after all those years had passed. I mean, it looked dated as all hell, but it still played fantastically well. I was swimming in nostalgia. I’d kind of left the video games industry at that point, having recently moved to Indonesia. It was a welcome trip down memory lane for me.
What can you say that were the biggest changes or additions in both patches?
Robert: I don’t think there was any big graphical or gameplay changes at all. It was just upgrading APIs and polishing things really. It’s difficult to do much without the editing tools. I was working on a new version of Redux for a while before GRIP came along. It was using proper floating point numbers for much of the engine instead of the fixed point numbers PlayStation had enforced. This resulted in it moving a lot more smoothly on the screen and I felt this was quite a big leap forwards. But in the end I never really got to finishing that because GRIP landed my doorstep. Once I started work on that, I didn’t really have the opportunity to look around until 5 years later. And now, it seems a little pointless completing that work given that GRIP is already everything we wanted Rollcage to be.
Did you try out the online feature with either game patch?
Robert: A little, but not much. I largely left that to the Steam group and they reported back to me with issues.
How did ATD get reached out for LEGO Racers 2?
Robert: I’ve no idea.
What can you comment about the development of LEGO Racers 2?
Robert: I enjoyed work on Racers 2, it was a good time. We had a very talented team and we were working with new technology on new consoles. Things were certainly taking longer than we were used to though. We had recently transitioned up to C++ and using modern development tools. But it took a while to get things right.
I don’t remember developing anything other than technology for that game though. Rendering tech, audio tech. A fair few graphical effects. But not really any gameplay stuff. But as I said, I kept getting seconded onto other projects so my input wasn’t as great as some other coders on that particular game.
I do remember one particular time when the technology was coming together where I was watching a snow blizzard fall over one of the lakes in the artic levels with the sun setting low over the horizon so you could see its reflection in the water. The water itself was animated with waves and ripples so the whole thing seemed quite alive. All of this was tech I had written myself. I think that was the first time I felt we had actually achieved a real step forward with the technology and that things we going to work out. Before that, it had seemed like a real uphill struggle.
It became quite a rush towards the end to get it completed on time, and it felt rushed and unfinished by the time we hit the deadline. Luckily LEGO had already accounted for that and gave us another 3 months to get it right. That was good management, and I was pretty happy with the result at the end of it all.
And I very much do remember my girlfriend’s kids playing it relentlessly for months after release – they loved it. So yeah, that was pretty cool.
Overall though, I’d say that game was a large learning experience. It would be fair to say that was the first game where we started to write professional quality code. The industry was growing up, and we had to be better at our work. That code was by no means perfect – the rush of creating the game didn’t permit that. But it was definitely a new start in the right direction and that felt good.
And it was nice finally working on a game that was contemporary with other PC games. Rollcage and Rollcage Stage II, and LEGO Rocket Racers were all old PlayStation 1 games running on PC hardware. They didn’t look as good as other PC games around at the time and to me that always made me feel a little inferior, like porting games from PlayStation was a bit of a curse holding me back from developing the exciting graphics PC was capable of. With Racers 2, despite it being aimed at the children’s market, it did feel like we were finally doing what we were supposed to be doing on PC.
What did you think of the original LEGO Racers?
Robert: I never played it.
What can you comment about Drome Racers development?
Robert: I wasn’t particularly happy about doing yet another racing game at ATD, I had already worked on four up to that point. But, Drome Racers was an interesting step forward from Racers 2. It was aimed at a more adult market so we could relate to it better. I was working on it from start to finish so I felt like I was properly invested this time. And the aim was to make everything look better than it did on Racers 2. So, there was a lot to be happy about and I do remember being pretty content working on the game.
The specifics though, really do evade me. I remember having a lot, and I mean a lot of tasks on my schedule. But the only thing I remember much about was working on the lighting and shadowing code for the game. This was a real leap forward from Racers 2 and much more on the cutting edge of what other PC games were doing at the time. I felt like we were pushing things about as far as we reasonable could in that direction and was pleased with the results we were seeing.
We had a hybrid vertex / lightmap approach. We used precalculated lighting at the vertices where detail in the lighting wasn’t important which was for around 80% of a given level I would estimate. For the remaining 20% where detail was important we used precalculated lightmaps which smoothly blended into the existing vertex lighting. This was really quite a feat at the time and as far I know something that had never been done before.
I also remember coding up dynamic lighting for the vehicles based on the precalculated static lighting. So for example, vehicles would become darker as they entered shadowed areas. This was done with again a novel technique that again, as far I know was unique and utilised a web of 3D vertices which sampled the light at various points in a level and then we interpolated those values across the vertices to solve for a particular vehicle location. It was all pretty complex for its day, but the code itself was lightning fast and resulted in realistic looking lighting for next-to-no cost. I’m fairly sure I have a document somewhere that describes this whole thing in detail.
Oddly, looking at footage on Youtube in preparation for this interview the car shadows seem to missing from most recordings, I have no idea why. I was actually pretty proud of having realistic, soft shadows at the time and that wasn’t easy to do. Though those character models were pretty appalling even then. I don’t know how we got away with that.
Things went pretty smoothly with the development, I don’t remember there being much stress or any big problems. But I do remember walking away from project at its completion feeling like the game we had developed lacked soul somehow. I was happy with the work I had done, but not particularly proud of the game we had released. What was missing I don’t know. It just needed to more for fun I guess, and perhaps a little more adult than it turned out to be. But I’m sure a lot of kids had a lot of fun with it, it’s just difficult to appreciate that as an adult.
I can say I remember feeling that I missed the Rollcage days, even then. Drome Racers was a decent enough racer, but it wasn’t Rollcage. I really wanted to do another Rollcage if we were to work on racing games at all. I wanted to get back into the game that I loved to play, and I wanted to do it as best we possibly could on PC. Little did I know it would be ten years later before the prospect of that happening would come around, just at a time when I thought Rollcage itself was ancient history.
What can you comment about the other ATD games released before liquidation?
Robert: I was principal programmer on a game called Ion Runner at the time ATD folded. The company had more than halved in size as it couldn’t support the salary bill it was used to. We knew we were living on borrowed time for a good year before the end. Ion Runner was our last, ditch attempt to put the company back on its feet, but I don’t think there was many of us left that felt confident we could do that. We had no other games in development, with this one we either succeeded, or the company was finished.
It was a game in an Anime style, and had a pretty unique art style all of its own. I liked it. I felt like this was a game I could really sink my teeth into and we did get quite a way into it. I remember us having the main character and his vehicle all working well and looking good. We had a couple of weapons in the game and quite a few enemies, fighting against you too. We may have had a couple of fully complete levels too before the end. It was really coming together and for me, it was the best game I had worked on since Rollcage. Had we not had the continual stress of not knowing how long our jobs would last I would have said that ATD might have been going through a bit of a renaissance during that period. But sadly, we did have that stress. The company was slowly dying as staff dribbled away for more secure employment. I didn’t blame them. As the months went by and we were unable to secure any publishing deals, despite our best efforts, it did feel quite desperate.
And so yeah, about a year after we started Ion Runner the company closed. It was a sad time. It was. But also, a time for renewal. Our time at ATD had run its course, and now all of us needed to move onto pastures new and see what the future held.
Was there a LEGO Racers CC/3/4 in the works? If there was, is there any surviving material, code or anything about that?
Robert: We did have a Drome Racers 2 in pre-production for a few months. It got shelved for Ion Runner. The only detail I recall of that was it was going to feature a track editor in the vein of what would later arrive in Track Mania. But very few people worked on it, it never really gained any traction. I don’t think we had any working code, just design ideas.
What other games or things you did since ATD’s liquidation/closure?
- Call of Chulhu: Dark Corners of the Earth
- Renderware 4
- Rail Simulator
- Diner Dash
- GRIP
How did you think in working on Grip Combat Racing? (which is clearly an homage/spiritual succesor to Rollcage)
Robert: Well, as a result of developing Rollcage Redux, Chris Mallinson got in touch with me and said he was interesting in doing a Rollcage sequel, and was I interested in joining him? Honestly, although the idea was certainly of interest, Chris had never worked on a full game before and I was patently aware of just how much effort it takes to develop games these days. I thought it was almost certainly doomed to fail if we did try, Chris just didn’t have anywhere near enough experience and we would need a large team to get where we needed to be. So I declined, for a while.
But the longer the idea sat with me, the more I wanted to try it. After 2 or 3 months I figured why not give it a shot for 6 months? I’d never used Unreal Engine before so I wasn’t clued-up as to how much time it would save us compared to what I’m used to working with internal tech. After 6 months though, I very much would know and we could decide then whether it was worth continuing development, or whether it would be too difficult. It was also quite a useful period in determining just how capable Chris was going to be in this endeavour too, I’d worked on far too many projects where partners had overstated their ability and left me holding the baby. I didn’t want to end up in that situation again.
But I should never have doubted Chris – he turned out to be one of the most reliable and multi-talented people I’ve ever worked with. We never looked back. It was best risk I ever took.
How many of the original Rollcage staff joined on Grip development?
Robert: Just myself really. We did have one of the old Rollcage producers working with us for a while, but it didn’t work out. I think he had been out of the industry for too long at that point and wasn’t as useful as we had hoped he might be. Lovely fella though, so it pains me to say that, it would have been great if we could have made it work.
What were your best experiences in working with Grip?
Robert: GRIP has been quite a ride. It’s been fantastic that I could revisit Rollcage and develop the best version of it that I had in me. It was also really nice being able to work up ideas and get them into the game without having to convince a team of people first. Of all the games I’ve worked on GRIP certainly feels like the one I gave the most to. Sure, a lot of people did work on it over the years. But as almost all the code in the game was written alone I definitely feel a huge sense of ownership for it. For Chris and I, it’s certainly our baby.
There are many areas of the game then that I could point to and say I’m proud of the work we did there. The weapons for sure were a lot of fun to develop. Even today, I keep going back to those, polishing them up trying to get them as shiny as can be. The cinematic camera the end of races too was great to develop. It really was immensely satisfying bringing that together, showing off all the things we have created in the game in a very cinematic manner. I loved it. AI – I definitely feel a lot of pride in how competitive we’ve made our AI bots and that we even managed to better what were considered to be very intelligent bots in Rollcage. I’ve even ideas now about how they can be improved further. It’s been very satisfying seeing them come along and offer the player more and more challenge. If anything now, players complain they’re TOO good.
But more than any of these, I had a huge amount of satisfaction developing the AirBlades, this was an idea I’d had on the back-burner for a while.
Both Rollcage and Wipeout were from the same publisher (Psygnosis then Sony), released in the same time period, and inevitably comparisons were drawn between them. Wipeout got most of the kudos, and rightly so, for being superbly well presented, having a great aesthetic and a then uniquely impressive soundtrack. It entered the mainstream consciousness in a way few other games did, it was certainly part of the popular culture in the mid to late nineties.
Rollcage too, got a tremendous amount of exposure and very positive reviews at the end of the nineties, but it never quite succeeded in getting the popularity or public attention in the way Wipeout did. Sometimes reviewers would write that Rollcage was Wipeout with wheels, which is a fair compliment. But for those of us that developed Rollcage, where we believed it had the superior gameplay and it deserved to be number one, saw it the other way around – and Wipeout as Rollcage without wheels. Either way, we liked to think that Rollcage was Wipeout+, and Wipeout was Rollcage-
It was all friendly rivalry though, we knew people on the Wipeout team, one of them even came to join us a couple of years later and likewise the Producer of Rollcage went on to head the Wipeout team after Attention to Detail folded. Nevertheless, we felt there was a wrong to be righted there. Wouldn’t it be good, if so many years later, Wipeout could finally, genuinely be bested by Rollcage’s younger, more demonic brother – GRIP. I was really excited about that idea.
The secret, wasn’t to set out to ape Wipeout, what would be the point of that? No, the much more interesting idea, was to explore the middle ground between Wipeout and GRIP, to produce a playable hybrid with the best features of both. It was fertile ground that no-one had yet explored.
Of course, we had to satisfy our core audience first in delivering a worthy successor to Rollcage, and this was always our goal with the first iteration of GRIP. That alone was a difficult proposition and took us a considerable amount of time to polish it to get it where it is today.
But the concept of Airblades was a part of the design, in my own mind, from very early on. We didn’t want to muddy the initial reception of the game with two competing vehicle types, and perhaps have a game where people didn’t quite know what it was. So we established it firmly in the Rollcage mould to give it instant recognition, and satisfied the those fans that were looking for a modern reimagining of that great classic.
But later, we took it up to a whole new level. We took the dynamism of the classic GRIP vehicles and injected that into what is traditionally a restrictive antigravity vehicle model. This is no longer the brake left, brake right, racing on rails of old. We merged the bombast of GRIP with the fluidity of antigravity vehicles and produced something unique in the racing world. Super-smooth vehicles that you can really throw around and drive with real skill, while fighting with the heavy weapons that GRIP provides.
And because I developed the AirBlades towards the end, it was a great way to finish the game off. All the hard work had already been done, and I could take my time to get these things right. It was a happy time, for sure.
Were any direct homages or ported tracks from Rollcage 1/2 in Grip?
Robert: Certainly, but just one really. Keen-eyed observers will notice the track called Impact is pretty reminiscent of Contact in the first Rollcage. I was very happy when we put that one in the game – Contact was one of my favourite tracks in the original and it just felt right having it in GRIP. But also, if you look around carefully in the track called Hive Horizon, you just might see some memorabilia around the place.
I have seen that you have a personal website, with even a blog you still keep updating, as well as some few photos and other stuff you have made.
Something that also caught my interest about that was the Ultra game you made (which is inspired/based in Tempest), but would be interesting too to know about the other things you made there.
Robert: Hmm…well, Ultra was something I started not long after ATD folded. I think I was searching for a new beginning, and not quite sure what direction to head in. I was finding work, but it wasn’t really exciting me. I had a couple of arcade machines in the house at the time; one of them was the original Tempest machine from 1981. I absolutely loved that game, and there’s not like playing it on original hardware with its spinner control. And so I thought let’s do Tempest, but better.
I worked on it casually for a couple of years in my spare time. I was happy where it ended up at the time, I thought it had good gameplay and looked pretty cool. These days though I could do so much more; I’m tempted to have another go at it with Unreal Engine, but time is a bit of a problem, so I’m not sure I’ll ever get back into it. But yeah, it was filling a void at the time. So it served a purpose.
I remember it getting reviewed by Retro Gamer magazine back in the day. They loved it and gave it a really high score compared to most, and it was pretty strange having one of my own games reviewed like that, with my own name in print. Normally of course it’s the developer you work for. It felt pretty special.
Regarding about working again in LEGOLAND Rocket Racers back in 2019, I answered about meeting James Dobbs again and how that was actually what I remember the most about that. But I kinda forget to mention that I also met Graham Owens again too at that time. In my mind somehow they were separate occasions but they were in fact the same. So I’d like to say it was really nice spending time with both of them.
You never know, if Graham ever read what you write about it he might feel unloved if we don’t mention his name. He’s a really nice guy.
That’s nice to hear!
Now that you mention loving Tempest, I wonder if you had heard or tried out Tempest 2000 (or any of its variants or clones, like Typhoon 2001 or Tempest 4000). I remember playing that game a lot on emulator because it was like Tempest but they jacked up some stuff and added powerups; I would guess that if you did know of it, you loved it too.
Robert: I did yeah. I had an Atari Jaguar for a while and played Tempest 2000 on that briefly. But this was around 2002 so quite a while after it was released and it did feel quite dated already. But you got to love Jeff Minter who is a bit a cult hero in the UK in programming circles.
I remember trying Typhoon and thinking it was OK. Tempest 4000 though, I just saw footage of that, never played it. It does seem strange that Jeff keeps making the same game over and over again though. I mean, he calls them different names, but they’re all on the same theme.
I never played Tempest 4000 either, but mentioned it as it was another T2K-based game.
Robert: It would be really nice to see him break out and do something different for a change. He’s obviously got some imagination.
I worked with a guy once who has a friend of Jeff’s. I felt weird knowing him through a friend like that, decades after admiring his work back on the Commodore 64. I bought Polybius recently, apparently it’s quite a trip in VR. I just need to find some time to fire it up and give it a go.
About the Jaguar emulator mention, its interesting that someone made an emulator specifically designed to run T2K as good as it can.
On the topic of emulators, what are your thought on those?
Robert: I went through a period from 2001 to 2010 where I was all about retro gaming and emulation. I was swimming in the whole thing really. I loved the 80s microcomputer boom in the UK, and playing arcade machines just as much. I had finally had some spare income in those years and bought a LOT of retro stuff.
Then around 2010 I pretty much sold the lot when separating from my ex-girlfriend and was about to go travelling for a couple of years. These days, I just keep my old ZX81 and ZX Spectrum around the house, and a recent copy of MAME on my work PC. I’m happy with that. Having all the equipment was just a phase really.
That’s nice to know! I’d love to know your thoughts about emulator developers (as I was looking interviewing some), and on the footnote of your travels, to know about those travels if you want.
Robert: I’ve a lot of respect for emulator devs, it’s a pretty specialised job. I was quite into it for a while, just analysing the source code for MAME. But honestly, I don’t have the patience for it personally. It’s a special breed of programmer that does. I couldn’t be a hacker for the same reason, too much detective work and not enough forward progress for me.
So around 2009 it became obvious my only daughter would soon be leaving home. I was pretty sad about that idea but of course it’s just life. But about a year before she was due to leave for university I thought it would be a really, really good idea if I took a few months off work and went travelling with her before it’s too late…And so we did.
We went very close to your part of the world, and spent 3 months in Ecuador, Peru and Bolivia. Compared to what we had done before, this was REAL travelling. It was life changing, ande had such a fantastic time over there, it’s, so, so different to home. We loved it. I was actually planning to come back over there and visit Columbia and Venezuela, until Venezuela deteriorated. Such a pity what was happened with your government.
Since GRIP, what else have you been working on right now?
Robert: I’m still on GRIP, incredibly. We’re putting together our last patch as we speak. I’m also working on an online programming course based on the game too, trying to teach coders with a real-world game example which is pretty novel. It’s turning out to be a ton of work though.
That’s interesting to know!
Robert: We’ve did some work on GRIP 2 for a while. But I just couldn’t find enough enthusiasm to continue with it. It was more my business partner’s idea that got dragged into. I think 5 years on one IP is more than enough for now. I really need to do something different for a while.
At least the concept was finally brought back to modern times with GRIP itself
Robert: Yeah, I’m happy with it. I’m quite content to leave it where it is for a few years now (and there’s a new track coming this month). We do have another project on the boil too, which I’m not properly with yet. I can’t really talk about it though because nothing has been announced yet.
Its interesting to know that up to this day there still are patches being worked on for the game; I find developers maintaining their stuff for years interesting to an extent, like the support for Worms Armageddon, Bungie patching Halo to use new servers after GameSpy died, etc.
Just by chance, what games have caught your attention recently?
Robert: Would you believe DOOM Eternal?
Fair enough!
Robert: Yeah, some Ace Combat 7 too. Also finally got to playing Abzu with the kids. What a wonderful game that is! Like a work of art.
Now that I think about your taste for Doom…Have you ever dabbled in mapmaking for Doom 1/2?
Yeah I did, way back when. I entered a PC Zone magazine map design competition. I think my name was in the magazine somewhere for having entered, but I placed nowhere near the winners. I rightly so, they were very good. This was back in 94 I think.
I thought about it again recently, having another crack at it for fun. But then I played Sunlust and Eviternity and thought why bother? Man, there’s no competing with those.
Do you recall what was the first emulator you ever used?
Robert: Yeah, it was for the ZX Spectrum, around the 1996 time period. I forget the name of it though, it was one the old one’s which isn’t around any more. Then I caught onto MAME around 1999 I think. That was pretty special; some pretty impressive engineering right there.
Was there any particular game you played often through online multiplayer?
Nah, I’m not a multiplayer kind of gamer. I much prefer single player.
By chance do you still keep contact with some of your friends from back then?
Yeah, I was, until quite recently. I still see my old boss who was technical director at ATD fairly often, once every year or two. But it’s been 4 or 5 years since I’ve seen anyone else. It’s difficult when I live so far away now.
What did you think when I asked for this interview/chat?
It’s an interesting thing you know. Since I produced Rollcage Redux a few years ago a number of people have asked me questions like this. Small interviews, that sort of thing. It’s quite strange for me, having been an introverted programmer for decades without any real contact with the public. I don’t feel like a celebrity or anything like that, but it is quite weird having people you don’t know interested in you. It’s all good though brother.
I’m looking forward to get further in these interviews and research for writing up articles; there are so many fascinating things out there that I’d love to talk about 😀
Sounds interesting. And no problem, good to speak to you too.
If you enjoyed this, you can check out all the other interviews I have done with other game developers and modders, and there’s also some few more interviews coming as part of a new article I have been working on recently 🙂
If you liked my content, sharing this with others helps me out; and you can lend me a hand through my Ko-fi page as well! Any feedback and ideas are appreciated, so don’t be afraid to drop a message either here, my Twitter acc or through Discord.
Special thanks to Justin Reynolds (Ko-Fi support), UTTER_BODGE (helped me on the LRR research), and of course, Rob Baker (real chill man) for the day of today, and thanks to everybody that has been following my misadventures this long. As today I finally got off a huge weight from university (at least for the next following months), I have a lot of ideas waiting, planned and also on the works, so hopefully there will be a lot of new cool stuff upcoming for this entire month! But until then, see you next time…
ARCADEX MACHINA 