I have been trying some roguelikes and games with roguelike elements.

Roguelikes seem shaky on the concept of a difficulty curve. They tend to have difficulty cliffs. Start, maybe a tutorial, good luck. And you know I’m all about the new player experience; creating an unwelcoming experience for new players is just poisoning your game, making sure it dies except for a dozen grognards who populate the last forum about the game and curse out new players looking for help as lazy idiots who should go play Angry Birds because they’re not willing to put in the time to learn how to play properly. Some roguelikes have heard about difficulty levels, but those tend be levels, not curves; either it adjusts the height of the cliff or makes it a plain. And too many games think that adjusting the numbers is an interesting way to scale difficulty.

Roguelikes seem shaky on the concept of balance. Players enthusiastic about roguelikes seem to be of the opinion that it is okay for difficulty to randomly vary between “doable” and “not even theoretically possible” because it is random. That’s the full justification: yes, you have effectively been given a 1000-piece puzzle with only 998 pieces, but you never know how many pieces are in the box until you put them together, and you’re not supposed to be able to put every puzzle together, so just keep reloading until you get a puzzle that does have all the pieces. I have seen advocates of balance via save scumming. I rather find it a large design problem if you can make all the right decisions and still lose. Which is to say, there was no “right” decision, so you had no meaningful decisions to make. But people like slot machines, too.

This is certainly not all roguelikes. Sturgeon’s Revelation still applies: 90% of everything is crud. But roguelikes that fail often tend to fail hard and do so in ways that are not immediately obvious whether the game has hidden depths or is just broken. I’m perfectly happy to invest small amounts of time in the face of randomization, but then you find creatively horrible ideas like “how about a 4X with roguelike elements?”

: Zubon

4 thoughts on “Roguelikes”

  1. I’m not a fan at all, although for me its mostly the general style doesn’t click. I generally hate ‘action RPG’ games, and roguelikes often feel to me like stripped down RPGs with barebones action; kind of a ‘worst of both worlds’ situation.

    That said I did enjoy Faster Than Light (I think that’s a roguelike?), which has some of the elements you mentioned above (sometimes the game just decides you shouldn’t win), so I guess it all comes down to how it’s presented.

  2. Any game that has nontrivial random starting conditions strongly tends to have balance problems in individual instances: not just roguelikes, but e.g. games like Civilization. I am a fan of nontrivial random starting conditions, and I am willing to pay the price of imperfect balance.

    Sometimes games go beyond that to add quite a lot of sloppiness or complete brokenness about balance, which is unfortunate. I have played a lot of Dungeon Crawl Stone Soup, mostly in the 0.5 version that happened to be installed on my old laptop but also some on the telnet server around maybe version 0.11. It has enough good things that I enjoy it, but I find it unfortunate and noteworthy how unbalanced the endgame is w.r.t. melee characters vs. AOE spellcasters, in ways that look as though a considerable amount of creative effort has gone into actively causing the problem. Various of the big threats in the endgame seem to be intentionally optimized to screw with melee characters even at the price of immersion. Enemies that swarm you, OK, that’s an immersive way of screwing with melee characters. But enemies that swarm you while some hang back, *and* target you magically without interference from intervening enemies (“smite targeting”), *and* attack completely ignoring the armor that is supposed to be one of the major advantages of your character (and that in one’s imagination does after all physically stand between you and the incoming attack), *and* attack in weird ways that are specially designed not to make any immersive sense but purely to neutralize any melee advantage in hitpoints (most especially by reducing any target’s hitpoints by half on each attack, but causing mutations that qualitatively reduce the target’s effectiveness also partly qualifies)… Once is happenstance, twice is coincidence, three times is enemy actions, and four times is, um, too high for Og the Ogre to keep track of easily but surely makes him want to retreat from the dungeon and go looking for a designer to smite. It’s still a fun game, especially before the endgame, because they do a lot of other things right, but to me it’s a fun game despite the designers pursuing this perverse trash-the-balance design priority, not because of it.

  3. There was a time, from the late 80s into the early 90s, when I was very much into NetHack, one of the grand Roguelike games. I had the source code, built my own versions, and even have a bound print out of the sprawling Usenet guide to the game. (Having a girlfriend who worked in a print shop helped on that front.)

    I loved the sprawling dungeons, the ability to leave behind scratchings or corpses that I might run across in future games. And when you were on a roll, it could be quite the game. I remember the glory of finding a well stocked store and getting my pet to steal anything I couldn’t afford.

    In the end though, the combo of the bad-to-impossible starts and the hardcore “you died, time to start again from scratch” aspect of the whole thing wore me down. And I never, in all those years, managed to do anything that might count as “winning.” All my characters ended up corpses, a lot of them not very far from where they started.

    It is kind of a fun memory still. Every once in a while I get on Google to check out the state of the code base. But in a lot of ways it is a relic of an earlier time, one of those things that came about as much because of the constraints we faced on computers back in the day as any conscious design philosophy.

  4. With the harder roguelikes, each game can be considered a chapter in the meta-game of finding a strategy which will ultimately score a win (usually, or occasionally, depending on the game).

Comments are closed.