Slime Squisher
Posts: 368
Joined: Thursday, 11th April 2013, 21:07
Re: Robot - new hard species
1) Get rid of upstairs.
2) Get rid of hunger.
3) Get rid of rest-to-heal.
4) Get rid of luring. Numerous approaches have been discussed here and I still think the one with the most promise is reveal-black-to-heal with a passive HP leak. It's elegant, it works, but it turns the game into something very different than what DCSS has been. (Real life interfered the last time I was coding this and I never got around to picking the project up again, but the prototype code and design docs still exist and anyone interested in reviving this can feel free to shoot me a PM.)
5) Decrease the experience return from late-earlygame and midgame. This can be achieved a bunch of different ways; compressing dungeon layout is one, decreasing experience return from the monster tables on the relevant floors is another.
Oh, yeah, also while we're on this step,
5b) Go through the monster spawn tables meticulously to get rid of dead weight spawns. This has received some attention in mainline already, I think (haven't played recently enough to say how well this has gone), but it needs to be done across the entirety of the game if your goal is to make the game always-on challenging. Note that doing this would come at the expense of gameplay lulls that might otherwise *improve* pacing by providing some interspersed calculated downtime. The proper approach here is probably not as simple as cutting trash everywhere, but rather cutting it where it's too common.
6) Rework available item generation to make inventory choices meaningful. This means cutting and merging item types. It also means cutting available inventory space.
I probably haven't mentioned everything necessary to fix the game here. You'll notice that a lot of this stuff has already been addressed in Hellcrawl. This is no coincidence. Hellmonk is a good player who understands how broken mainline Crawl's combat systems are, and we discussed this at length early in Hellcrawl's development process. I think the result was the best bit of Crawl development to happen since DCSS first got tiles.
Unfortunately, you can't fix the game with a set of species apts and preexisting mutation conducts. DCSS has more fundamental problems than can be addressed that way, unless you want to go write all the code necessary to rework the base game into something not-broken and then wrap it inside of a species check for some insane reason.
edit: Also unfortunately, if you follow this line of design through to its conclusion you can easily come to the decision that fixing the game requires enough of an overhaul of the way things work that you'd be better off just putting together a spiritual sequel in Unity or something. This doesn't seem very compatible with the existing dev process, but it would probably result in a fun game.