Tuesday, September 23, 2014

Episode 91: Our First Roguelikes

This is episode 91 of Roguelike Radio, where we talk about the first roguelikes we made. Talking this week are Darren Grey, Eben Howard, Jeff Lait and Nicolas Casalini.

You can download the mp3 of the podcast, play it in the embedded player below, or you can follow us on iTunes.

Topics discussed:
  • Nicolas' first Angband variant, inspired by the dragonriders of Pern (which lay the eventual seeds for ToME)
  • Ido's CaveRover, which is still a cool game
  • Jeff's POWDER, which is still awesome, and was originally meant to be a Nethack clone
  • Eben's Quarker, where you play a quark interacting in strange and beautiful ways with other quarks
  • Darren's Gruesome, where you play a grue hunting adventurers in dark places
  • What motivates people to start making their own roguelikes
  • The desire to creates worlds for others to play in
  • The desire to play with new mechanics
  • The lure of making the next ultra-awesome feature-packed roguelike (Avoid the lure! Make a small game first!)
  • Copying others, and starting from a derivative place
  • Challenges we faced when we first started out, and lessons we learned as we progressed
  • Ido's Roguelike Blueprint
  • Richard Garriott's first published game
  • Go make a game! Go on! Now!!!

Join us next time on Roguelike Radio for a return to Brogue!


  1. Ha! 10,000 if blessed! I chortled.

  2. Oh man, I'm excited to listen to this. Can't now as it's 2 am and I need to be up for classes at 10:30...maybe if I traveled at a high enough fraction of the speed of light while doing so...

    Err, but yeah. Quite excited to hear this, as I am myself sort of stalling in the process of creating a roguelike. I have to tools to start finally, but I haven't thought of actual game design very much yet. I've got a mildly simple idea I've been playing around with in my head revolving around doors, but I'm not feeling quite inspired with the idea.

    Hoping to get some insightful thoughts from you guys, just as soon as I have the time. Thanks for keeping this podcast going! It always gives me ideas and the like when it comes to this stuff. I probably wouldn't be as far along as I am (admittedly not especially far) otherwise.

    Going to bed now because this is getting long and I think it's starting to stop making sense...goodnight?

  3. Replies
    1. But I had made a platformer before that with Gamemaker and 10 years before that I had made a Castle Risk game in high school.

      So a crappy little random shooter was my first RL. Then a couple Rogue inspired card games, then SUN CRUSHER!!!, then KlingonRL and The Littlest Princess.

      I do think I'm ready to make a big game now. Finally.

  4. Very true about the feeling that libraries are cheating. I gave in to the fact that I didn't have the will power to learn /everything/ needed to create my own roguelike from the ground up and tried learning libtcod with C++.

    That actually didn't end well, but I learned a huge amount of information about libraries in the process, and why some things work the way they do. I ended up going with BearLibTerminal, made by Cfyz of the Rogue Temple forums, and things have gone much better since then, likely aided by the fact that I can bother the library's creator when needed.

  5. It might be interesting to have a similar episode about the first roguelikes you played, and if that influenced what you look for in other roguelikes you play, and if that influenced you in the sorts of roguelikes you design.

  6. Finishing a small game project is indeed hard. I've found that I start by working my ideas and code outward like a spreading tree from a few starting points. If I don't at some point define an end stage, these tendrils never work back toward a single trunk. Eventually I get bored of my own hair, because the development process is without outcome.

    I'm now going to write a small roguelike :D Thanks for the motivation. This should be a task in tertiary education systems.

    I also realised when listening: I've already made something that's somewhat roguelike. A top-down grid maze SDL game that's turn based with AI, done as a highschool project. It's been sitting around for years and I uploaded it a month or so ago to github.

    Source code: https://github.com/Veyrdite/echidna_menace
    Screenshot: https://raw.github.com/veyrdite/echidna_menace/master/screenshot.png

    Assuming you have a *nix system handy: compilation is easy. Code is (overly) simple -- read the description for more of the gorey details.