Quote Unquote

Taking Independent Videogame Developers out of context since 2010.

c:\indie games\ // Oddities // Twitter


The Last and Final Word: Linley Henzell

Linley Henzell barely needs an introduction. His first game was Crawl, an ambitious roguelike, which remains in active development by others to this day, even though Linley has long since stopped his own involvement. His shmups, which include Overgod, Garden of Coloured Lights and White Butterfly have attracted a lot of attention over the years, not only for their polish but their vast array of ship and weapon combination choices, unique appearance and tough difficulty.


Old enough to have played games on an Atari 2600 for reasons other than nostalgia.



What do you do?

My actual job has nothing to do with games or programming, or even really with IT except that I use a computer to type things. I write games in my spare time.

Development tool(s) of choice?

I use Code::Blocks to edit code and GCC in the form of MinGW to compile it. The Allegro library covers graphics, sound and input. The GIMP for graphics. Sounds come from a variety of sources, mostly SFXR and a heap of random VSTs run through Psycle.

I have to admit to being somewhat ignorant of what you did, game-development-wise before you gained recognition for your shmups. Firstly, how did you get into game development and secondly, can you give us a little bit of background about what you developed before you started focusing on shmups?

I learned to write BASIC on my parents’ first computer, which was an “IBM compatible” XT, and then a C64 that I got for my tenth birthday. The most exciting things in the small town I grew up in were the two arcade machines in the local deli, with a rotating roster of games like Double Dragon, Juno First, Xain D’Sleena and a mysterious one with everything written in Japanese (later identified as Kiki Kaikai), so naturally I wanted to make games.

I got some help from an administrator at my primary school who made educational software and taught me how to write little proto-games in C64 BASIC, but I always ran into the hard limitations of the language and after my parents bought a 386 with DR-DOS and a vast 105MB hard drive I got them to buy me a copy of Borland Turbo C++ as well. I liked PCHack, with its random levels and savagely difficult emergent gameplay, and decided to teach myself C by writing a roguelike.

The roguelike turned into Crawl, my first and (by far) most popular game. I released it under the same version of the General Public Licence as Nethack 3.1.3, thinking that as Hack turned into Nethack, maybe after I’d stopped working on Crawl someone else would take it over and keep it alive. Happily this did happen, and the Stone Soup dev team are doing great things with Crawl despite the nightmare tangle of spaghetti code I left behind.

After Crawl I picked up the Allegro games library and started working with graphics. I wrote a couple of really bad Liero clones then a couple of arena shooters (Lacewing and Overgod) before I started writing shmups proper with Excellent Bifurcation. Allegro is a bit old-fashioned - procedural C, 8-bit graphics support, low-level access to most things - but it does everything I need it to and I’d find it difficult to work with anything else.

You have concentrated your efforts entirely on shmups for a number of years now. What fascinates you so much about the genre that keeps bringing you back to it?

Lack of artistic skill: it’s a lot easier to draw top-down views of abstract spacecraft than it is to design and animate human figures. The way that, for a genre with such a rich history, shmups get almost no attention from mainstream developers (and the shmups that do get developed, rarely seem to make it to Australia) so I feel like I’m doing my little bit to help keep them alive. The way that the basic framework of shmup gameplay, which is approachable and instantly recognisable to anyone who’s ever played a videogame, allows a lot of experimentation with mechanics and style. Things like that.

Who or what are your influences, within the independent gaming community?

I used to download and play any indie game I could find, but to be honest I don’t play that many of them anymore. That’s partly because my 6-year-old computer is getting a bit long in the tooth and has trouble displaying anything too sparkly and mostly because I just don’t have as much time as I used to. But the visual style of my games is heavily influenced by Kenta Cho’s shmups, and I learned a lot from pitting my games against Cactus and Monorail in the competitions shmup-dev.com used to run. Jph Wacheski is another indie developer with a style I like, especially his beautifully eerie Cathode Raygun.

And other influences?

Non-indie: I really like Tetsuya Mizuguchi’s style (e.g. Rez) and can’t wait for Child of Eden. And I’d love to see Platinum Games do a shmup. C64 games like Armalyte, Cybernoid 2 and X-Out are a big influence, and Ultima IV is still the best example of effective and expressive minimalist graphics I can think of.

Do you have any interest or any plans to attempt any other genres in the future?

The game I’m working on at the moment is a 2D top-down homage to Freespace 2. With little fighters buzzing around giant space battleships, while they tear each other apart with glowing energy beams, that kind of thing. I think it’s going to be called “Angry Moth”, unless I think of a better name. That’s still sort of a shmup, though, and when it’s finished, I’d like to work on something a bit more involved.

I have an idea for a combined 4X strategy, RTS and arena shooter game which will be a sort of cross between Star Control, Liberation Army and Master of Orion 2. But I also want to return to turn-based gaming, and I have some other ideas for an intensely political grand strategy space colonisation game with multi-player and simultaneous turns, like a mixture of Master of Orion 2 (I really like that game) and Solium Infernum. That would be a vast amount of work, though. One day!

If you had to choose, which of your games are you most proud of and why?

Crawl! It was such a huge amount of work, and it’s still alive more than a decade after I stopped working on it.

But of my shmups, probably Garden of Coloured Lights. It has the most consistent and coherent sense of style. I really like those outlined bullets with their little trails (even though I borrowed them from rRootage) and the music from Stage 4 still haunts me.

Do you feel that it is important to try and ‘innovate’, even in a well established genre (that is, not to re-hash the same old game mechanics and ideas that others have done in the past) or are you happy just to put out a well balanced, polished and playable game? Do you feel that your creations innovate?

I don’t think they’re very innovative. In fact I’m struggling to think of a single really original idea in any of my games. Mostly I isolate particular aspects or mechanics from a number of games I like, then put them all together in a way that seems like it might be fun to play, rather than think of anything new. Crawl was just a mishmash of details borrowed from other roguelikes and RPGs, although I think I mostly picked pretty good details to borrow. Excellent Bifurcation was a split-screen display modified from the one in an old DOS game called Flightmare wrapped around a set of basic shmup mechanics. Garden of Coloured Lights was another basic shmup mixed with ideas from Rez and rRootage.

I’d love to come up with something truly innovative but unfortunately whenever I have a genuinely original idea for a game it turns out to be either not much fun to play or beyond my abilities to implement. Like the ecosystem / evolution game I tried to make: so many awesome ideas, such a tiny amount of entertainment.

In a increasingly faceless population of bullet hell and arena shooters in the independent gaming community, you continue to put out some fairly memorable entries into the genre. How do you make your shmups stand out?

By making them hard. So hard that people write reviews of them on Sourceforge complaining that they’re inaccessible to “casual gamers”.

My favourite gameplay mechanic is roguelike perma-death: a character who took hours to build up can be destroyed forever by a few poor decisions and a single turn of bad luck. When you can’t just reload a save from two minutes ago again and again until you get past any obstacle, decisions become meaningful and the game stops being a quasi-interactive movie and becomes a game again.

A designer who demands that players subject themselves to that level of cruelty has to offer something in return, and a good roguelike feeds the player with a constant supply of interesting decisions and rich, preferably randomised environments. It’s harder to do that in a shmup, but I balance the difficulty out with partly random stages that require a flexible, tactical approach and by offering the player a choice of different starting conditions and playing styles.

Have you been satisfied with the feedback you have gotten from the independent gaming community and others in general?

Yeah, people have been really helpful. I’ve always gotten the most support from communities dedicated to a specific genre of game, in particular shmup-dev.com, and the people at rec.games.roguelike.misc back in the late heyday of USENET when I was writing Crawl. It’s places like these where you get fewer one-line “hey, cool game” comments and more in-depth critiques of the gameplay. But I’ve had great feedback from many different places.

How do you approach the development process and how much time do you generally spend on development?

I don’t really know how long it takes - maybe three or four months of five to ten hours a week. I’m a lot faster than I used to be, partly because now I have a large library of old code to draw from and partly because practice has made me much better at getting the overall program design right from the start, which saves lots of time. If I were still an unemployed student, I’d be able to churn out a new shmup every month or two.

My approach to development is to do the hard technical stuff - untested algorithms, complex functions, Hellspider arm growth, anything else that would require a complete re-design if I couldn’t get it right - first, then work on making the game basically playable before I start adding too many features. It’s much easier to motivate myself if I can actually play the game, no matter how incomplete it is.

Has development time increased as your games have become more polished and technically advanced, feature-wise?

If anything, my games have become less featureful. When I stopped working on it, Crawl was a vast, sprawling monstrosity full of details that most players probably never saw. The second game I released, a bad imitation of Liero, had a ridiculously complicated menu interface for customising every aspect of the game and even did multi-computer multiplayer (over serial cable, unfortunately). A lot of learning how to design games has been developing a sense of which features to leave out, and that makes development faster and easier.

Do you have any advice for anyone, thinking of trying their hand at game development or any new developers, who are still learning the ropes of their craft?

Learn to draw! I wish I had years ago. And for selfish reasons, it would be good if people made more complex, deep games. I like those.


  1. gam3z reblogged this from quote-un-quote and added:
    Interview de Linley Henzell, créateur de Crawl (le fameux roguelike devenu “Dungeon Crawl Stone Soup” par la suite) et...
  2. nihilocrat reblogged this from quote-un-quote
  3. quote-un-quote posted this
blog comments powered by Disqus