Thursday, October 02, 2008

Good programming doesn't mean great software

I recently found some of the best C game programming from the previous decade; the source for the game Angband. The code has seen quite a lot of revision and it does what it does well. However, the game is basically a huge waste of time. If there was ever a game to identify with hard-core, hack-n-slash, geeky role-playing, this is it.

Chief faults:
1)Interface design -Too Cryptic, requiring knowledge of macros to work well. 2)Game play -Extremely tedious and permanent character death is part of it's culture. 3)Story -Uh, you make a character and you go around killing stuff. That's about it. 4)Scope -One town and huge frigging dungeon to crawl. 5)Frustrating -Character Death occurs too frequently and load/save is considered cheating by it's community.

Moral of the story, good programming does not make a good game.

2 comments:

  1. Having played roguelikes for ~20 years I know what I get out of them, but can also see where you are coming from:

    1) Yep, I don't recall any variant that purposefully tried to introduce the commands through a tutorial or redefine the interface to give context sensitive commands. I have introduced a number of people to Angband or variants and most still play today, however the first 20 minutes is explaining the commands / interface.

    2) Tedious is a matter of play style, character choice and game choice. Running ('.') is your friend and I would argue tips the scales the other way. There is more opportunity to tweak the game to your playstyle so that you get to the more 'interesting' parts of the game quicker (whatever you define as being interesting). This is sorely absent in recent MMOs and RPGs due to the higher levels of realism that they want to convey. Unfortunately most of these timesavers are for the inquisitive as you will need a good understanding of the game to tweak it appropriately, thus falling into problem (1.

    3) there are mods with many storyline elements (ToME, UnAngband, etc), however the appeal of Angband is with the uncertainty of what is around the next corner. It tugs very hard on the exploration gene. I like quests, I like open worlds, but I also like randomness. The holy grail for me would be a random quests with a consistent, cohesive random world.

    4) Try different variants for open worlds. They add a different level of exploration, not necessarily better exploration.

    5) It's all about the highscore. Almost always your death is pretty stupid and could have been avoided. That drives me back to the game to compete against myself to do better. I'll go through phases where I will purposefully ramp up the difficulty for a more intense game, then other times when I batchfile away my savegames because I know that the character I'm playing is as good as it gets and will save myself another 100 rerolls to get to where I am. Even though the culture may expect permadeath (and I certainly enjoy playing that way sometimes) I see no real need to FORCE it. On the flipside I would thoroughly recommend new players starting with permadeath ON as it is a different style of play that is becoming ever more rarer in today's gamespace.


    Your moral is actually the opposite to what I perceive. I continually come back to roguelikes because they are almost pure gameplay. There is nothing else, no sound, no graphics, no realism to focus on; just the gameplay. Even recently after bitching and moaning about the lack of gameplay in Spore I fired up a copy of UnAngband to see if it was just me being a cranky old gamer. Nope, the old games can still suck me in for 5+ hours and are as addictive as ever.

    On the code front it has only been in the last 10 years or so where there has been a concerted effort to open up the code and clean it up. Many, many variants came and went because, although they were free, the code was not published. Many variants were simply 'experiments' and didn't have the rigorous standards you would expect from open source code these days. In fact I know that there are legacy sections in Angband that were hacks to get a new feature in , but have hung around because it still works. ToME is possibly the best redesign I've seen as they have established a modular framework for the game to exist on (and be modded from).

    ReplyDelete
  2. Thanks for the insights vrbones. I might give some variants a try.

    ReplyDelete