Interviewing Cas Cremers, Parallax MSX Software founder and Information Security Professor at Oxford University

After a lot of effort and many setbacks, here in Distrito Entebras we bring to you a very special interview, we talked to an historic developer from the Dutcth scene of the late 80s and early 90s, Cas Cremers. It has been a pleasure making a collaborative interview made by two, you can read at its end who were the askers. This is the literal dialogue which tooks place:

«Back in time, the Dutch scene had a major quality to the MSX catalogue, mostly for the second generation machines. The evolution from the demoscene groups into videogaming developers was, in the MSX case, a curious mix, sometimes they paid homage to famous Japanese titles and manners; sometimes brave projects in a more Westerner way. One way or another, the hungry MSX2 users at the time didn´t ignore them. But there were some games which were superior in all aspects; their final results were astonishing, with a high audiovisual quality, and their mechanics and global ambition were at the same level as the commercial software – games that there were only a few – in several cases they were better. Those were the Parallax MSX Software, which in several cases it had a staff with some members, in most of the cases there was a single behind them, a factotum man called Cas Cremers. That quoted ambition, among with that perfectionism and inconformity related to the commercial games in the old japanese standard, it unleashed a bunch of games like Magnar, Blade Lords or Akin«.

«After his unfinished obra magna – despite we can enjoy a promo version – Core Dump, the Dutch developer Cas Cremers (which interests, MSX stuff too, are preserved in his «old» personal website), is still related to computers. On these days he´s a Professor in Security Information at Oxford University, but he gave us some of his sparing time to answer us some questions. A great chance to get closer to a genius of 8 bit software development, for the 2nd generation MSX machines.»

[Frank Romero] – We know that your beginnings on MSX productions was Sigma Group, making demos which included graphics and musics mostly ripped from Konami´s games. Did you know well the MSX demoscene in Holland?
[Cas Cremers] – I got to know the early Dutch MSX demoscene quite well during the Sigma group period. Of course, the scene was then still quite small. When I later started making games I did not really keep track of the demo scene, but I got the impression it grew substantially after we moved out of it. For myself, programming the demos just felt like practicing for programming larger things that I wanted to create.

I would recommend young people to chase after their dreams, and encourage them to pick up technical skills – it’s not as hard as it might seem, and it is fun to do. Sometimes people are afraid of trying to accomplish or learn something larger, because they are afraid of failing. I think an advantage that I had was that I never thought “I can’t do this”, but rather thought, “let’s see how far I can get”, and wasn’t afraid to put in the time. It turns out that with a lot of effort, you can often go surprisingly far.

[Frank Romero] -Would you recommend to go to Computer Parties to the nowadays young people?
[Cas Cremers] – With respect to Computer Parties: I’m not sure what happens at computer parties nowadays, but I personally learned a lot from trying to program and design things together with friends. The discussions helped us a lot to inspire each other and try to push further.

[Gabriel López] – Back in time when you were an MSX game developer, did you have contact with the rest of the Dutch scene? Did you ever go to users meetings, such as Zandvoort?

[Cas Cremers] – When I started out with the Sigma demos, we did go to the local MSX scene in the south of the Netherlands, which had very regular user meetings. Later we traveled to places that were further away from us, such as Amsterdam. I attended Zandvoort several times. But this was before the internet was so well developed, so we were not in contact with other groups as much as everyone is now. It was fun, but it wasn’t realistic to build any collaboration from it, so we just visited large meetings once it a while to see some of the new stuff. But these visits didn’t change much of what we were doing anyway. I think I reported on going to several Zandvoort meetings in the online Core Dump development diary.

[Frank Romero] – Your games are over the top at several levels (Graphics, Music, Playability, Game Designing). Even in the present days it´s unbelievable so many quality is there in your games. Were there planned from the beginning, or did you code some parts of the game and later you modified them after testing was made?
[Cas Cremers] – My original plans were actually always much more ambitious than what I ended up making, and I typically had to accept the fact that I was not going to be able to create the things I had in my head. Most of the time this was because of a lack of time or other resources, but also not enough talent in all the aspects to produce what was in my head.
I created most of the larger games together with Patrick Smeets. Since these were relatively large (for that time), a tremendous amount of work had to be done next to my other activities. Often we worked a few hours per weekend, and I would sneak in some night hours during the week, as I was in secondary school (and later university) during the week. Clearly, not everything was planned from the start, since the projects would certainly take a year to emerge. So there would typically be some starting vision that lacked detail. During development, lots of things would change, and some of the more ambitious ideas had to be reduced to things that could be realized. In this sense, it was quite unlike modern large-budget team development, where you plan the entire project well in advance, and then allocate resources and work it out as closely to the plan as you can. This was much more organic.
Sometimes the direction of the organic development also depended on how far we could push the technology. For example, Core Dump’s split windows only became possible after some time. Once I knew we could do them, we could dream up some new weapons that would only make sense because of it (proximity mines, guided missiles, etc) I of course also spent quite some time on making the tools for development, such as a drawing program for games, a map editor, a music editor, and a sound effects editor. These things would drive new ideas as well.

[Gabriel López] – Anyone can trace influences from some game sagas or famous genres in your games. Was there any other famous game or saga you ever wanted to bring to MSX but you couldn’t?
Cas Cremers] – There were of course many games that influenced Parallax games, but I never felt we were just “bringing” other games to MSX. That was not the goal.
I had many ideas for more complex strategy games with computer opponents, but my ideas there were a bit ahead of what the technology could actually do at the time, and these never emerged. At the other extreme, I was also interested in “pure” game mechanics. This, as in Blade Lords but also in Scalar (an online Flash game I created) which tried to distill the idea of “risk versus reward” in games down to its simplest core.

[Gabriel López] -It’s really impossible to find Arc nowadays, be it legally or as a copy. How did the protection system work on that case?
[Cas Cremers] – I don’t completely recall the specific details of Arc’s protection, but I think it was a simple stateful circuit connected to some specific in/out port. If it’s not a port, it must have been a designated memory address. Depending on what combination of values you wrote to it, it would yield a different value upon reading.

[Gabriel López] – Is there any way to get the full game today?
[Cas Cremers] – I don’t know how to get the full game today. Ironically, I don’t have it myself anymore.
(Note to the readers: There are 2 avaliable Arc demos in digital magazines Clubguide Picturedisk. The first one (un number three) is a playable autodemo and the second one is a playable demo which it can be found on number five. There was a «compilation cartridge project» with all Parallax games – Arc included – but Sunrise, the main responsible for that, stopped its activity).

[Gabriel López] – -We can see some kind of evolution from Vectron to Magnar and Black Cyclon. Was it intentional, or was it only you refining your techniques?
[Cas Cremers] – There is some evolution going on from my early top-down Arcade games, but I was also just trying to build better games. I did always feel I ended up making something that was much worse than the vision that I had in mind during development. This feeling then drives the need to have another attempt to that particular game right. Now, looking back, I’m happy with each of the games in their own way, but still critical of many specific aspects.

[Gabriel López] – Of these three games, which is the one you’re the most happy with?
[Cas Cremers] – Vectron will always be special to me since it was my first game, I was still quite young, and I had literally only the bare basics for development. A Philips MSX 2, a book or two on Z80 programming, the VDP manual, a hex editor, and a disassembler. I had no assembler. So, while the later games were much better, I’m quite happy with this as a first game.

Magnar was a return to the top-down view after making Arc in the meantime. It was more mission based, with much more involved action and weapon types than Vectron. It’s quite substantial in its item types and “depth” to the game.
Black Cyclon was my original attempt at trying to think about more dialogue and a more substantial plot. In this sense, Black Cyclon has early ideas that ultimately led to Akin. But in Black Cyclon, these ideas somehow got lost during development and juggling social life, studying, and making games. By then I had gotten much of the programming quite well under control for that type of game, and I didn’t see many possible improvements, so I needed a new challenge, and went for a scrolling game next.
I will always find many things wrong with my games (or whatever I produce) when I revisit them. I don’t think I have a particular favourite. Black Cyclon’s second level works quite well though, now that I think about it.

[Frank Romero] – Let´s talk about Blade Lords. This little gem was a surprise for me, it´s so playable and the graphics are so well finished. There is a real legacy there, all the Games & Watch heritage there, it´s a rival even from some Arcades at the time, like Nightmare from the Dark for SNK. How did you think the sword dynamics in that game?
[Cas Cremers] – Blade Lords surprised Patrick Smeets as well. He returned from a year in America and found I had made a game on my own in the meantime.

After developing Magnar, I was trying out new ideas for a rendering engine for a game, and so I had to write some code to test it. We had been playing Taito’s Bubble Bobble a lot, and of course Konami’s King’s Valley 2 appeared around this time. I played against my brother a lot, and so somehow the idea came about to have some sword throwing mechanic within the more cooperative Bubble Bobble idea. This ultimately lead to the core idea behind Blade Lords. It was clear from the start that the intent would not be to directly kill the other player, but only to gain a temporary advantage by stunning. While this most directly stops the other player from getting to some bonus item in time, it actually allowed for implicit kills of one player by another, if timed correctly with enemy patterns. It took us a lot of playing to fine-tune the basic idea though. I experimented a lot with different settings for the player inertia, which ended up owing more to Super Mario World than to Bubble Bobble. This inertia makes chasing the other player much more interesting. Similarly, I experimented with having the option of holding multiple swords (up to ten!), but that made the game sometimes too confusing, and distracted from the simpler chase mechanic where you either sprint to the closest sword or bonus. If there is only one sword, there is a different risk/reward mechanic for throwing it. So for the final game, I disabled the option of multiple swords, though I think there is some hidden debug menu somewhere to get that back.

[Frank Romero] – Why did this game included so many differents stages? Was there any idea who left out in the finished version?
[Cas Cremers] – The idea behind the multiple stages in Blade Lords was to have some variety in scenery and enemies. I chose to have different “worlds”. I did experiment a lot with different enemy types (and the previously mentioned controls and sword options), but sticking to the simple core seemed to be best.

For a Blade Lords sequel, I considered many ideas, certainly in terms of level design and enemy types. I was thinking of a Super Mario World-style map with different possible paths and secret exits, but I never got round to coding anything for that.
I actually started making a Flash web version of Blade Lords with improved high-resolution graphics. I got quite far with the project, and made the first few worlds playable with high-res graphics. In fact, I recall it was easier to make such a Flash-based version than I expected. However, I never completed this version, even though I had some beta-testers already playing it.

[Gabriel López] – About Blade Lords, did you get to design or develop anything, or was it just the idea of the sequel in your mind?
[Cas Cremers] – I created Blade Lords entirely on my own. Afterwards, I thought of a sequel, since it turned out that so many people liked it, and wrote down many ideas on paper, but never started programming anything concrete for it.

[Frank Romero] – Your next production was Akin, a huge creation about Sci Fi themes. Even the plot is impressing in this game. Cyberpunk elements, Violent Sci Fi and an Alien Conspìracy among other elements. Which were your inspirations from this game? Maybe films like Aliens or Mangas like Ghost in the Shell?
[Cas Cremers] – Akin was inspired by a lot of Science Fiction. Indeed, I had seen some Ghost in the Shell by that time, but not a lot. I would not be able to name a single specific influence. Many ideas came together at this point. The objective of course was to make something unique. I haven’t thought about it for quite some time, but I recently saw a Youtube video of the opening sequence of Akin with the first few minutes of playing. I think, the way it opens, it still feels quite special. I’m happy about that.

[Frank Romero] – If we pay attention to this game, it seems more a demake from a Genesis/Megadrive game than a 8 bit computer videogame. I like the designs of Akin, with the main character (a Mecha) and the corridors of the Space Base, the musics which created sense of tension at some moments in the game…did you make Akin thinking in commercial productions at the time like Super Metroid or even Mega Man?

[Cas Cremers] – Of course we had played Super Metroid and Mega Man, but the vision of a more ambitious story driven game was really the core idea behind Akin. I think it was influenced graphically quite a bit by Konami’s Mecha game Cybernator on the SNES. Patrick loved that game. We were influenced graphically by many Japanese games at the time, including of course Metal Gear 2, even if that influence is not that obvious. In the end I think that although graphically it owes to many games, it was a unique attempt to weave more story, tension, and character into a platform action game.

[Frank Romero] – Where did you learn to make so such great effects and narrative? They are really amazing. Usually coders and engineers don´t have so many artistic sense of script and story.
[Cas Cremers] – Thank you, that’s very kind. It was just a matter of trying, making mistakes, and refining. I guess there is a certain stubbornness to keep trying if the first attempt doesn’t work out. I never felt “locked in” a certain category (say, coder, graphics artist, or game designer). If I had an idea I would just try it, make something that wasn’t so good, try to figure out why it wasn’t what I hoped, and retry. I guess one has to be a bit stubborn to make this work and not give up.
My main struggle was typically with creating the music. I have no background in music whatsoever, and certainly never learned to read musical notes. Therefore I had to first find out about these things in the first place, before being able to create a music editor from scratch. Once that was done, I just started to try composing things, which was an interesting experience. You end up reinventing the wheel a lot, but that’s fine, as you learn things in a much deeper way. I always felt that the music was the weakest spot. The hit-and-miss ratio typically was very bad: I would come up with twenty music tracks or more just to pick one for a game. Patrick always had a much better ear for music, but I think I just invested many more hours in experimenting. It was only when I did the music for Akin that I was happy myself about most of the tracks.
With respect to narrative, I had been reading a lot of novels when I was younger. This included some Science Fiction but mostly generic literature. I was very drawn to the idea of a deeper narrative, but I never really got to the point where I could comfortably weave such a narrative into a game. (Who knows, I might still try to write a novel at some point, just out of curiosity to see how far I would get.) For Akin I really started trying to integrate such aspects, and I could somehow imagine that it would work within an MSX game. Core Dump had a lot of narrative. However, looking back now, I only see how much was still lacking, and to which extent some story aspects were superficial. For Core Dump I did plan some interesting character development, but only for a handful of characters. With more time and resources I could have done much better, especially for Akin.

[Gabriel López] -About the infamous hard drive crash, which ended Core Dump abruptly, how finished was the project at that time?
[Cas Cremers] – At the time of the harddisk crash, Core Dump was very far in development but also still far from my vision. I estimate it was already around twice the size of Akin by that time, if note more. I had a large plot outline written down, and I still had much of that to incorporate. Many game maps – similar to levels, but Core Dump wasn’t linear – had already been created. But given my ideas and ambitions for Core Dump, I think it would have taken me at least another year to finish. My motivation was also going down, perhaps I wanted too much, had too little time next to everything else I was doing, and the MSX audience was steadily decreasing.

[Gabriel López] – How many times have you been asked about Core Dump?
[Cas Cremers] – I stopped counting the number of times I have been asked about Core Dump. It’s a pity I never finished it and a lot of good work got lost. I know people wanted to see it completed. The demos only contain a fraction of what was developed. But over time, I got used to the idea that I tend to produce much more than ever sees the light of day. That’s okay.
Sometimes I see on Youtube that some people are still interested in the Parallax games. This still gives me pleasure. It’s actually very strange to see how long the interest stretches for something you made so many years ago.
This has been a development for me as well. When I was younger, I did not realise in which way the positive feedback mattered to me. It’s in my nature to create things, regardless of other people’s opinions. I was always grateful but mostly confused and surprised when people came up to me and expressed that they enjoyed the games or artwork or whatever. Maybe there is some disconnect there that psychologists would enjoy.
In retrospect, I think the Parallax website and development diaries were also very useful to people. It’s essentially what you would now call a blog. I would be happy to think that through Parallax MSX games I inspired some people to pursue their own ideas. Perhaps that’s the best possible outcome.

[Gabriel López] – You really pushed the MSX2 to the limits, did you ever think of developing something on MSX2+ or Turbo R?
[Cas Cremers] – I did consider at some point developing for the MSX Turbo-R. MSX2+ did not add many interesting capabilities to MSX2 from a game perspective, especially for the type of games that we were developing. The Turbo-R was technically much more interesting, but by the time this became a viable option, the overall MSX market was already substantially decreasing, and the Turbo R market was even more limited. It just felt as if very few people would even see what you were making, and I decided it wasn’t worth it.

[Frank Romero] – And finally, do you feel nostalgic about making a game in the present, or even designing a new one?
[Cas Cremers] – I loved making games, especially the creative aspect of it. And there is a special feeling to seeing someone else playing a game that you created. Now I´m doing research, which involves a lot of creativity as well, so in that sense I continued.
I still enjoy playing computer games once in a while and I own several game consoles. Having said that, I honestly didn’t feel the urge again to create my own.

Modern computer games typically have become huge projects which require large teams of people, which is quite different from what we did. There are still indie games though, but ultimately I was also intrigued by pushing the technology at the edge. To do that nowadays one would need large teams of specialist, and the cost of developing say, a graphic engine, needs to be gained back over many games. I absolutely do feel the urge to create things, but computer games are only one of the many things one could create, «there are so many things to do in the world!». While I had a great time creating computer games, I now get to explore other exciting areas.
[Frank Romero, Gabriel López] – Thanks for your patience, Mr. Cremers. We´re pleased and honoured to have talked to you.

The interviewers:

Frank Romero «Entebras» (Distrito Entebras, Pixelstorm)
Gabriel «Gaby» López (MSX Extendido, Metodologic, La Hermandad)