RollerCoaster Tycoon creator on the resurgence of management games


Management games have had quite a turbulent ride over the last two decades, but for the moment the genre seems to be on the up. Cities: Skylines, Prison Architect, and Planet Coaster, among others, give a wide range of creations to simulate and control.

We’ve even seen the return of RollerCoaster Tycoon, in the form of RollerCoaster Tycoon Classic, a revamped version of the first two games by the original designer Chris Sawyer. Initially released on iOS and android, the game has finally made its way onto PC. We took the opportunity to speak to Sawyer about his work on RollerCoaster Tycoon and to get his opinions on the genre’s status as a whole.

Before starting, it’s worth noting that, although, RollerCoaster Tycoon is Sawyer’s most popular title, it wasn’t his first foray into the simulation genre. Prior to that, he’d already made a name for himself by being the lead designer on Transport Tycoon (a favourite of Adam’s), a game where players build supply routes and move goods around a map. Following this game’s moderate success, he began exploring new challenges and soon settled on the idea of a theme park management sim, where players work through a number of scenarios while trying to turn over a profit.


Sawyer recalls, “I spent a lot of time on research at the beginning but mostly from books and videos rather than trips to theme parks, and the game’s toolset sort of grew organically depending on what really interested me personally and also what was possible within the limited power and capacity of the PC at that time (around 1997-1998).”

PC limitations would be a big problem for the game. Sawyer had to be conscious of available memory during every decision in order to achieve an acceptable performance level on a wide range of PCs. He states, “The isometric landscape and limited “piece by piece” roller coaster construction method with fixed angles and sizes came about because they worked well with the power and memory available…[Though] there’s not a lot I’d change if I could go back and re-write the game without the power and storage restrictions – the simplicity and restrictive framework in the game I think could be part of what makes it such a good game.”

These limitations were also responsible for one of the game’s most memorable features: the ability to torment and kill guests, either by drowning them or demolishing rides. In both cases, these were allowed because the alternative required more graphics and memory rather than a deliberate design decision.


Sawyer says, “It’s quite amazing, and scary, seeing how inventive and devious some players are with trying to torment and kill guests, when that really was never the intention! I never set out to deliberately allow such violence in the game.

“[Particularly] for the drowning guests, it was laziness on my part – the alternative would have been that guests would swim to shore and climb out of the water, but this would have involved a lot more programming (and graphics) than a simple animation of the guest waving for help and finally disappearing.”

RollerCoaster Tycoon was never intended to be anything more than a niche project, but it became a hit, when it finally released on PC in 1999, so a sequel followed three years later. Talking about the follow up, Sawyer says, “It was actually a much more straightforward project compared to the original game, though the quantity of work involved was a lot greater due to the vast number of extra roller coaster types and other rides and stalls we added. Looking back with hindsight, I think I spent too much time on the mechanics of the second game and not enough time on the playability side, particularly the parks/scenarios included with the game and the lack of “progression” and challenge/achievement.”

Sawyer fixed this for RollerCoaster Tycoon Classic by editing the park scenarios from the first game to work with the second’s toolkit. This means that the experience is balanced for two different types of players: those who like simulations and those who the prefer progression and management aspects of the first.

After the second game released, Sawyer more or less disappeared from the industry, acting as consultant on the third title and releasing only one new game over the next decade: Chris Sawyer’s Locomotion. There were a number of reasons for this. The desire to spend more time on his personal life. A general dissatisfaction with the types of games becoming popular. And legal issues with his publisher Atari over unpaid royalties.


His decision to take a break from the games industry coincidentally occurred around the same time there was an ebb in the popularity and quality of simulation and management games. I asked Sawyer about this. “Perhaps the management/simulation games were all getting too big and too ambitious and were becoming too much like construction kits rather than being challenging games? With the increasing power and memory of computers 10 years ago there was an expectation that each new game would have more detail and more size than the previous one, and for a management/simulation type game that’s a slippery slope I think – more detail and more size doesn’t necessarily make a game more fun and possibly appeals to a much narrower audience.”

Nevertheless, the genre has had a resurgence of late, with new technology and modding communities like Steam workshop arguably helping games like Cities: Skylines and Planet Coaster reach a more widespread audience.

Sawyer says, “The ability to share and create user-generated content and the availability of third party tools has been the biggest change in the games industry over the last 10 years or so, and the games that embrace it without it compromising the game itself have been massively successful because of it.” He adds, “It’s a very tricky thing to manage well though, as it’s a fine line between allowing a game to potentially be opened up and watered down to the extent that it’s only a toolkit, and being overly controlling in terms of what’s allowed or not allowed, and also trying to protect copyright and revenue generation of the original game.

“It’ll be interesting seeing how things pan out in the next 5 years or so – it would be great if we could use the best of the mod/creative/sharing community to enhance the game playing experience while still ensuring the game itself remains as fun and as challenging as it was when originally designed and created.”

With the genre now riding a high, I finished by asking Sawyer about his plans for the future. Would we be seeing an original game from the developer anytime soon? Sadly, his answer – at least for the moment – is a no.

“I’ve been offered a lot of opportunities for working with publishers (or development teams) on their new projects but that’s just not the way I work, and I don’t think I’d be all that effective working in that environment either. [So] not at the moment, but I’d never rule it out!”


  1. Carra says:

    I was always impressed by the fact that this game was programmed by one man… in assembler.

    • Xerophyte says:

      Yeah, Sawyer started in the early 80s, where working in machine code and assembly was a lot simpler and relatively common. Some of his first projects were apparently PC ports of David Braben’s games, including Elite II, which were likewise written in assembly so I imagine that was a pretty natural fit. Elite II remains one of the most absurd feat of video game programming that I can think of.

      Anyhow, keeping to x86 assembly as your language of choice in 2002 is a bit eccentric, but he clearly knows it extremely well and if it ain’t broke…

      • TillEulenspiegel says:

        Writing everything in assembly is like cleaning your entire floor with a toothbrush instead of just using a damn mop. You could do it, but you really shouldn’t. Even old versions of C offer far greater productivity.

        That said, some incredible games have been written by very unusual (and in some cases just plain bad) programmers. But generally that kind of stubbornness is unhelpful.

        • Xerophyte says:

          Nowadays, sure. I get enough of a headache when I have to break out the vector intrinsics, I definitely do not want to handle registry allocation manually. The compiler can do that annoying stuff for me, thanks.

          In the 80s? C is still easier to write but not significantly so, and although optimizing compilers were making inroads you could expect that your compiler will be very dumb. x86 is still a relatively friendly dialect, and ARM and 68K are both significantly simpler. I admit it’s about a decade before my time and I know very little about the Acorn side of things, but at least all of the greybeards I’ve met who were part of the Amiga games or demoscene at the time wrote literally everything in 68K and avoided the actual Amiga APIs like the plague in favor of talking to the hardware directly. Elite and Elite II probably wouldn’t have been possible in C, they pushed their platforms to an insane degree.

          By 2002 — when Rollercoaster Tycoon 2 was released — the x86 platform was of course way more complicated and the tools mature enough that sticking to primarily assembler for a new project is no longer a very good use of time. But Sawyer knew x86 very well, and the Transport Tycoon terrain code was battle tested and could be adapted to amusement parks without needing to reinvent any wheels. He managed to crank out a stable, high-performance and cracking good game in two years with that approach, so who am I to criticize?

  2. Michael Fogg says:

    I don’t play sims/tycoons that much, but I think the main way the nu-school games like Prison Architect differ from the classic approach is that they take inspiration from roguelike genre and implement procedural generation to creat unstable systems. The difficult to predict interaction of the games’ many systems create a situation where everything is constantly on the verge of crisis. Ths sometimes feels random and unfair, but also serves to deliver a lot of drama. On the other hand the classic SimTycoons were built on more rigid rulesets and the resulting systems were more stable and predictable.

  3. Borgadzim says:

    To this day, my favorite game is still Transport Tycoon Deluxe. Nothing comes close to that one. OpenTTD is amazing, but I really hope that one day someone makes a spiritial successor for it.

    • grimdanfango says:

      There is:
      Transport Fever – Clearly a TTD homage in many ways, it’s considerably more engineering-focused, with very challenging tracklaying that may not be to everyone’s tastes. Personally I love it though!

      Mashinky – a grid-based one, very clearly re-implementing the TTD look and feel, the construction works on the exact same kind of grid/height-steps layout, but with an added realistic mode you can switch to that makes it all look prettier when you’re not building. Haven’t tried it yet, and still in development. Heard it’s good, but still a little thin on featureset.

      • zegenie says:

        I’d also add “Rise of Industry” to that list. With a graphics style that resembles the look of TTD/X, but a slightly different approach to the core game play, I find it a refreshing take on the genre. It’s still in EA/alpha, though, but with a clear roadmap.

    • BlacKHeaDSg1 says:

      So many times i tried to play OpenTTD but i just can’t stand diagonal train tracks and confusing UI. They could learn something from Sawyer and clear out UI.

      • onodera says:

        Isn’t the UI the same as in the original TTD, but with a few improvements?

        • Darloth says:

          Not exactly the same, but very, very similar in almost all respects. You do tend to have more windows open in OpenTTD because there’s more options, though, so that criticism may be somewhat valid. Personally I think their UI is fine, I’ve seen far worse, but as always preference may vary.

          The original had diagonal train tracks as well, though.

    • MrEvilGuy says:

      If openTTD could include realistic traffic issues and realistic bus simulations like city skylines, I’d be much happier with it. On that note, city skylines should have been made into a transportation game with competing corporations, not a city builder. The two games need to merge.

      • JustAchaP says:

        You should try out Cities in Motions. Made by the same people that made Cities Skylines.

  4. gesmin says:

    Thank you for this interview and my thanks to Chris Sawyer. The mere mention of Transport Tycoon rekindles fond memories of a game I spent countless hours playing and trying to perfect. The freedom of play, sense of exploration and progress, attention to detail in graphics and sounds and welcoming interface and purpose captivated me. And I think it marked my introduction to multiplayer too; with friendly rivalry and no arms in sight.

  5. joekiller says:

    Re: devious play.

    I loved transport tycoon. Playing it long enough, I had several truck train accidents and eventually my train took out a competitor truck accidentally and I observed there was no consequence. Just a newspaper. I seized upon this newfound technique for competing and setup miniature ramps on both sides of a competitor’s road with a train depot at the top. I’d roll out a cheap engine that would ramp up and down it over and over. The end result was pretty much killing every single truck that would ever roll in front of it. Sometimes you’d get 4 at a time, the maximum due to spacing. There would be piles of the smoking wrecked trucks and my train just kept swooping.

    Because there was no consequence for “accidents” and the ai didn’t recognize what I was doing, ai would continue to try to setup road routes and I slowly would bleed them dry; squashing any new company and stifling growth of anyone who was already established.

    The technique was end game.

    • Darloth says:

      When I was younger, I too would torment the Transport Tycoon AI in a similar manner.

      My approach was slightly more subversive. First, I’d build a level crossing in front of a major truck or bus route, and park a train on it, manually stopped.

      Then once there was a nice long queue, build another blocked level crossing a suitable distance behind – so that the busses or whatever have enough room to move. This way, they still cost maintenance moving up and down the section of road, futilely trying to reach a depot. Soon the company would go bankrupt.

      …I have since ceased my pointless cruelty to unthinking AI, both on general principles and because it isn’t as much fun now I’m older.

  6. piphil says:

    I have spent way too much time playing Sawyer’s games. I think being one-man projects, they felt tight and focussed. They were games about systems and efficiency, especially TTD.

    This is the first interview I’ve seen him do for a while though. He became a bit of a hermit media-wise, especially in the period he was dealing with Atari. Which is a shame, I’d like to ask him a lot of questions about TTD, and what he thinks of OpenTTD? Unless someone has already asked him?