The state of DCSS is interesting because it's like a software Ship of Theseus. It's been actively hacked (and slashed) upon for so long that it is hard to claim that its the same game, even if the git commit history shows a direct lineage otherwise. Or, at least, the aspects of the game that I once so greatly admired (which I wrote about here [1]) are longer present. Only a superficial similarity remains. A very opinionated group of core developers with a minimalist design philosophy has driven the project for years now, and the game now is not the game I knew.
The DCSS community has vocal minority that's against any change to the game. Often they forget that old versions are still available for download, playable offline and on most online webtiles servers.
I for one appreciate the underlying philosophy of the game and the overarching intent for any change: to remove tedium, grinding, trivial optimal play, remove the requirement of spoiling yourself to beat the game - the latter point making it basically the opposite of a game like Nethack.
That's an unkind characterization. People aren't just complaining to complain - from their perspective their gripes are legitimate. I'd categorize myself as an explorer type of player (https://en.wikipedia.org/wiki/Bartle_taxonomy_of_player_type...). So I like to explore how game mechanics interact with each other and new levels. Lately, the devs have removed lots of details and nuances in crawl. That makes the game less interesting for me because there are less things to discover. For example, mummies are kind of pointless now since no race needs to eat. Removing features may be good for achiever type of players though, because it forces them to come up with ways to beat the game without those features.
It's fine not to like the game or some of the changes but I don't really get the resentment towards the devs that I read in some of these replies though.
If there's one thing where I believe that "if you don't like it, just fork it" is warranted it's this type of game. It's not like it's production-critical for anyone, so there's no deadline and you can take your time. It's a rather minimalist 2D game, no complex game engine knowledge is required. When I looked into the code a few years ago I found it very readable so it's a nice target for customization.
If you feel like modern DCSS doesn't go in the right direction just fork whatever version you preferred and go from there.
Now I 100% love the direction of DCSS, but yeah mummies should have been deleted and possibly ghouls as well. But more generally, DCSS is and continues to evolve to be a tactical combat game. Complexity in other areas is continually being stripped out.
> Often they forget that old versions are still available for download, playable offline and on most online webtiles servers.
Typically players get upset because they like the majority of the quality of life changes, and new content, but don't like that something was removed that they can't enjoy in that new version. Community knowledge also deteriorates over time; you can ask questions about recent versions, but antiquated versions require meticulous code diving / commit log scanning to figure things out.
Sounds like an instance of 'bitcreep': https://www.gwern.net/Holy-wars#bitcreep Yes, all the binaries are still there and runnable in a VM, but everything around them can't be run in a VM, and those changes impose negative externalities on those who were happy with the previous way of things.
Dwarf Fortress has similar “breakpoints” where major releases that had to drop save file compatibility have separate “version” pages on the wiki to support those who remain on the older version to continue running the fort or mega project they’re working on.
Interesting, that sounds like the philosophy of (modern) ToME, which takes some of those ideas to extremes, to its benefit I find! For example, potions are removed in favor of Inscriptions, that are equipped to limited slots and have cooldown-limited usage, but otherwise stick around forever.
I find ToME has much more unique classes too, doing a lot of interesting things, even if luring monsters into a 1-tile wide corridor is still the safest way most of the time...
Another possibility would be to add some things as game options, e.g. add game option that you can set at the beginning of the game if foods are enabled/disabled, if certain classes of items or other features (e.g. religion) are enabled/disabled, etc.
With a lot of software, it may be the case that you may want some of the changes but not others. (Fortunately, you can fork the project and make your own modifications, but you might not want to do that, for whatever reason, but if you do, then it will be a nonstandard version.)
I really like the philosophy as well. But there are probably less than a handful of players that actually managed to beat DCSS unspoilered. It's a really hard game, even knowing the various tricks.
26 years and counting! It's passed through many able hands.
My chief lingering complaint with crawl is that the game is too long. Early game is tedious and last I played seriously the midgame tended to drag on. I'd like to see the standard game contracted to twenty-seven levels. Squeeze down Orc, Lair, the Lair branches and so on.
It is a common complaint among a lot of experienced DCSS players that the game is too long. I have 10 wins under my belt and I stopped playing mostly for this reason. About half way through the game your victory is essentially guaranteed.
I disagree about the early game though, this is the most exciting part of DCSS, and the hardest.
It's definitely a different game. Personally I love the changes and mostly agree with the devs. Food was for most characters a pointless mechanic once you had a basic level of game knowledge. However there are also some interesting forks which preserve some of the old mechanics.
The changes remove some of the most memorable moments:
- Eating questionable meat for random powers
- Losing your great items permanently
- Not getting poison resistance and putting off the hive for ages
- Not being able to carry as much as a weak character
- Putting a stash on Hive:2 so monsters can't get your items
- Being a mummy summoner and having a full screen of summons (no hunger requirement)
- Hill orc priests would wander the map gaining converts
A lot of these things became part of the game muscle memory to avoid. They were hilarious and emotional as a new player. They are now obsolete. I have mixed feelings - I always felt enough mitigation existed in the game and I enjoyed learning about it
Item destruction I found okay because randarts were immune and it made them feel special. They could have went down the route of mitigation but I respect their clean design ethos
Weight management - yes, hard to have too much nostalgia for that one
I really miss being able to play a SpEn backstabber. It was the highlight of my DCSS experience. Such a fun, if slightly overpowered, combo. Then they took invisibility out of the enhancer starting book.
I started replaying DCSS last week after a hiatus of about two years, SpEn is one of my go-to backgrounds and it's still quite fun IMO. You can worship Dithmenos to get increased stealth. It's not as OP as invisibility but it's still quite fun once you get your stealth to a high level.
As I mentioned in the sibling comment, there's a vocal minority in this game loudly complaining about any change. My suggestion is to decide for yourself, play an older version if you prefer, or any of the dozen forks, all available on most public servers.
Off the top of my head, they have recently:
* removed the whole hunger/food system: no need to eat, no need to butcher corpses or collect food, magic doesn't make you hungry.
* removed cursed items: an issue only in the early game, trivially managed by careful identification of unknown scrolls
* removed unremarkable species or monsters
* removed trivial spells such as Deflect Missiles or Regeneration
I haven't played rogue/hack/nethack games in a couple of decades, so I'm out of it. But: no need to eat? That was a key part of the game (though the cannibalistic part of that was sometimes weird). Cursed items, on the other hand, were a PITA so I wouldn't miss that part.
About hunger management: it was annoying, it didn't bring any strategical or interesting gameplay, and if you were unlucky with RNG you'd just die. One or more inventory slots had to be used for food. It would heavily penalise newbies or cautious players that were taking the game a little slower than most. Would force you to cast lower level, less fun, spells so that you'd not starve because you cast one too many level 9 spell.
That doesn't sound fun to me, and don't miss it for one second.
I haven't played in a few years but always felt like crawl's hunger system drove faster games -- where in NetHack I could usually find a way to survive indefinitely and camp out and explore slowly, crawl always felt like the Star Wars garbage compactor pushing me forward.
I think I liked it but I'm not sure it helped me master the gameplay (I sometimes felt rushed even when hunger management wasn't obviously an issue).
Food is far more important in nethack than it ever was in DCSS, it was never a key part of the game in DCSS, just an annoying (and sometimes spoilery) minigame.
Just to be clear, the death clock is hidden and is intended to be impossible to trigger if you're not actively cheesing with one of the very few exploits that depend on waiting for a very long time. IIRC to tune the clock, they took the longest non-exploit game on record, and doubled the time that took.
While I like many of the changes in DCSS, I have stayed on 0.24 because I'm wary of how the game will play without food. I kind of enjoy struggling through the early-ish game without using permafood, and then seeing a pack of orcs and going "Hey they look tasty, maybe I can eat some of them before I starve to death!"
That being said I want to play Djinn eventually so I'll upgrade when that time comes.
I had a miserable time trying to contribute to this game. The code wasn't the problem; it was the team. Lots of bad personalities. During my short stint, two of their most prolific team members rage quit in profanity-laced exits. It was some Office-Space style shit. And after my own rage quit, I humorously got invited to a chat room full of people like me: people the DCSS devs stepped on. And there were many.
For what it's worth, bad behavior can be excusable if said team were full of talent (think Linus Torvalds). But with DCSS, it's a handful of medium-talent stewards who inherited a popular game and turned it into an exclusive honor society of pointless change. I wonder if they realize how ironic it is for a game titled "Stone Soup" to be closed to outside contributions.
Yeah, they keep on removing mechanics for what appears (to an outsider) to be petty reasons: the developers don’t like map modification, so you can no longer use wands to disintegrate various solid materials. (Which is a bit disappointing, because it’s sometimes fun (in the Dwarf Fortress sense, at least), to take on the angels found in certain temples.) It’s particularly frustrating because recent versions have some nice UX modifications for things like manual skill training and there’s no easy way to get the modern UI with some of the older mechanics.
Ah yeah, the guy that got banned because he couldn't create the team White Lives Matter, rants at large about questionable political matters and shits on any decision the developers make because he wasn't consulted on any change, created an official-looking website for the game at dungeoncrawlstonesoup.com, with crypto donations links directly to him, and a video in which he blames the dev for being mistreated and persecuted.
He's a great Crawl player, but happy to see the back of him on the subreddit.
And whilst he is a great player, that win streak is mostly a record because no-one else wants to play DDAr consecutively forever just for questionable bragging rights.
Can you say when this started or when you were trying to contribute?
I've been following DCSS very loosely for about 10 years but I only noticed that there seems to be drama around I'm not understanding the reasons for in the 2 or 3 years.
Hey bhaak -- my recollection is hazy, I think pubby is referring to the acrimony around Forest branch. I think this would've been 2014 or 2015? One core dev objected to sweeping gameplay changes merged by another dev and things went sideways.
I'm thinking it's a figure of speech as well. I couldn't find said review elsewhere and it wouldn't make sense to leave one on Glassdoor anyway. It's a noncommercial open source project.
I've been playing this game for 15 years now. Somehow, it just doesn't get old.
One great DCSS tradition: whenever a new major patch is released, there's a two week long tournament. You more or less just try to win a lot, in as many ways as you can, for two weeks -- but it's more fun than it sounds, if you can believe that. For example, there's all sorts of "achievement"-like mini goals, such as "make it to level 6 of the Lair without using a potion" (approximately remembered). Between that and some other quirks, even players who aren't in the habit of winning the game can score some points for their team.
Taking a completely single player game and deciding that tournaments for it should be team-based was a minor stroke of brilliance, by the way. It's surprising how much it adds to the experience.
The tournaments kick ass. Agree that challenges make for some fun gameplay. Playing in a tournament with a random team I found really increased my enjoyment of the game.
DCSS is one of the best managed roguelikes out there. The Webtiles client has been getting some new art tiles in the trunk branch as well. You can spectate on any player's game more of less, I have enjoyed peeking at high level play on some of the bigger servers
Dungeon Crawl Stone Soup and Cataclysm: Dark Days Ahead are the best roguelikes I keep coming back to after years, and getting full-on obsessed for a few weeks.
DCSS webtiles is incredible and I've missed something so effective and well designed in any other roguelike. Playing online is so much better, especially when you make use of one of the help bots and have people spectating your game while making the orbrun or during a hairy situation.
This is one of those games I play intensely for a few weeks every year before forcing myself to stop. It has so much depth and is so addicting. The spectator mode on the web version is fun and I learn a lot watching other players.
I just finished the book Dungeon Hacks by David Craddock, which is a fun history of the early days of roguelikes—encompassing Rogue itself, Nethack, Angband, Moria, and the like…roguelikes had never “clicked” for me, though I wanted keenly to “get” DCSS when I’d played it previously.
I think I get it now! I’ve been playing Brogue, which I understand to be a bit more barebones than DCSS (eg no classes) and I’m loving it. I’ve been saving DCSS to savor it once I’ve really grokked roguelikes a bit more, this might be the cosmic sign / push I need to make the transition :)
I recently discovered this game, and am really enjoying it. I've played roguelikes such as nethack, angband and even rogue itself. Although I'm not really very good at them and haven't ever won any of them. But at least right now stone soup is my favorite. It's less tedious and grindy. There are commands to wait for health to recover, and to auto-explore. I don't have to worry about searching for traps and hidden doors. I don't have to worry about running out of food. But it is still plenty challenging.
I know what you mean. I guess there's just enough novelty to keep me going.
That's something that I think that the devs don't quite get. It's all about intelligent novelty. Not really challenge or advancement, tho that matters some. But really the whole exploring thing, and the encountering of weird new stuff. That's the important thing. Repetition is poison.
The levels need to be better designed fortresses, not just random crap.
And there needs to be truly open-ended and smart generators for monsters.
Maybe a good quest generator too. That might be hard.
As a fan of the game, I always wondered why there wasn't an infographic drawing lines between desired species and desired roles in DCSS. Having made one, I found out why. It's not really readable by humans.
It's kinda funny - I played a lot of moria on the Amiga - which had a very nice tile set. Then I spent long, dark hours in angband on Linux, with no tiles (there are and was tiled variants for Linux - I just played the text tiled ones).
Then I tried looking for other rougelikes, and found stone soup... And I find it a bit... Meh.
There's added features/"improved" ux - but I'd prefer to just play moria without tiles.
Might be similar to multi-player angband - a great idea on paper, but it involved a clock/tick - and suddenly the game was completely different, and IMNHO no longer fun.
I call these kinds of games (roguelikes, terraria, minecraft) "learning curve games", because your peak enjoyment of the game is when you have fought your way up the learning curve about halfway. Once you know everything about the game, it becomes boring. And until you get to that halfway point, it's a lot of trial and error (or looking things up in wikis).
Learning curve aside, most of the gameplay loop in this and most roguelikes is movement / auto-explore, ramming into enemies and healing as necessary. Far less of the "strategic" game people make it out to be, though you can spend time tinkering if you want. Brogue tries to rectify this with more tactical play on smaller maps at the outset - it's nicer but fundamentally you end up doing the same thing.
For my part if I'm going to spend a good deal of time grinding, it's generally more fun to slash and shoot something in the face in real time. But those actions games tend to miss out on some golden moments of random chance, tense decision making or interesting possibilities.
I think this is why roguelites have gotten so popular as they try to bridge the gap. Also the likes of Rimworld and DF that let you pause and unpause, advance time at high speed.
Played the original rogue in the 1980s and dabbled in various versions nethack, angband, moria over the years but they always seemed too complex for my simple brain. I prefered the original rogue, because I could usually get to level 18 before the Umber Hulks got me - even fighting them in a corridor. WIll give this a go - I like the mouse capability although my fingers can move me around a dungeon using hjkl almost as fast.
I don't even remember the mechanics of this game, but a few years ago I obsessively grinded out a small personal achievement: a DUHZ run where you only go into dungeon, depths, hell, and zot. That took a _long_ time. I don't even remember why it was difficult, just that it was really hard and I was very pleased with myself after. Would recommend.
Personally I like using the arrow keys to move, so I wrote macros for the game so that I can move diagonally while holding shift and the regular arrow keys.
I was once going to write 20,000 words or so on this question, but haven't gotten to it yet.
Crawl is much newer, much more actively developed, much more prone to radically change from version to version, much less based on learning details through experience, much more helpful to the player in the UI (both in terms of features like autoexplore, autofight, autotravel, and item search, and in terms of proactively prompting you to avoid accidentally doing wasteful or pointless actions).
NetHack is older, funnier, prioritizes and facilitates obscure knowledge on the player's part more (e.g., complicated techniques for identifying items that require memorization), has more complicated interactions between and among items, players, and the environment (e.g. many more distinct ways to use a given item), and is crueler both in terms of unavoidable deaths ("The spikes were poisoned. The poison was deadly.") and in terms of allowing the player to take actions that foreseeablely result in waste or instant death.
Crawl has more replay value in terms of different ways to win [i.e., not having a standardized set of items that you need to collect and use], and even different definitions of what counts as winning [i.e., you can choose to collect anywhere from 3 to 15 runes]. Both games have conduct challenges of various kinds to make the game even harder.
In NetHack, unidentified scrolls are somewhat more dangerous than unidentified potions; in Crawl, unidentified potions are uniformly more dangerous than unidentified scrolls. :-)
In NetHack, obscure game knowledge that you acquired twenty years ago is still extremely valuable. In Crawl, it is probably already long obsolete and likely refers to monsters, items, or dungeon branches that no longer even exist in the modern game.
The two games famously differ in their approach to spoilers. In NetHack, the game almost never reveals any quantitative or concretely practical information about the purpose or effect of something, and even things like weapon hit percentages and relative efficacy traditionally are considered spoilers. In Crawl, the online help system will, in recent versions, even reveal your current exact hit probability against a particular enemy and the enemy's hit probability against you, as well as what kinds of attacks and resistances each of you has (and you can also learn more about why, if you want). In NetHack you even need comparatively rare items to learn with certainty what resistances you have at a particular moment; in Crawl you can pull this up on a convenient screen at any moment.
Religion is dramatically more important (and complicated!) in Crawl, and different gods significantly change your play style. In NetHack there are minor differences related to your god's alignment which you would probably not need to think about in a typical game; there are always exactly 3 gods, of 3 distinct alignments, in a NetHack game.
Item identification (and modification) is much more important in NetHack. Artifacts (special, unique items with abilities different from their base type) are rare and you might find or use something like 1-4 of them in a typical NetHack game. In Crawl, they're extremely widespread and you might find or use several dozen in a game, including ending up equipping 100% artifact weapons, armor, and jewelry by the end.
Your character class is more consequential in Crawl (there are comparatively larger differences in skills and intrinsics that result from it), though in NetHack it does result in a single completely unique per-character-class dungeon branch in each game.
As an example of the UI helpfulness difference: if you try to walk into lava in NetHack:
> You fall into the lava! Your small shield bursts into flame! Your pair of leather gloves bursts into flame! You burn to a crisp...
In Crawl:
> Why would you want to do that?
Anyway, that's less than 600 words, but I hope it's useful!
Edit: I like the sibling comment's observation that Crawl is "not simulation-y". To take one example, NetHack developers have thought a lot about gravity and circumstances where you or some object falls on top of you or some object or some trap and exactly what should happen then. And they've thought about what would, or wouldn't, cause flasks to shatter or eggs to break, and what the consequences of that would be, and so on. Crawl basically doesn't care about this: yes, if you stop flying and thereby fall into an unsurvivable place, you will die. Otherwise, nothing in Crawl ever falls on anything, and nothing ever shatters from impact. The extreme number of physical interactions that are simulated in Dwarf Fortress, and the resulting chains of causality, were reportedly inspired by NetHack. Dwarf Fortress has taken this much further and made the number of interactions radically larger, but NetHack does have this kind of flavor.
In Crawl, if you're telling someone why you died, it's generally "a monster hit me and I died from the resulting damage" (maybe with a surrounding tactical context about why you misjudged the fight). In NetHack, it could very often be something like "I did X, which caused Y to happen, which caused Z to happen, which then killed me instantly". (Crawl requires you to have 0 or fewer hit points in order to die, except from starvation in older versions that tracked hunger and satiety; in NetHack there are probably 100 different ways to die other than by taking hit-point damage. Maybe 6 or 7 just related to drawbridges...)
I try to explain Nethack to people by talking about cockatrices. They turn you to stone. So you could pick up a cockatrice corpse, and hit monsters with it, and turn them to stone! Well, only if you're wearing gloves, of course.
And then you meet a succubus who seduces you and takes off your gloves so you turn to stone, or you are burdened, try to go down stairs and fall down them instead, so you touch your cockatrice corpse and turn to stone, or...
All of this is of course less fatal if you happened to be polymorphed into a stone golem at the time so you was already stone, but who knows how long that will last.
>yes, if you stop flying and thereby fall into an unsurvivable place, you will die.
In recent versions, you just suffer severe draining (temporary max HP reduction, removed by gaining XP) for every turn you're on an unwalkable tile while not flying (which are only accessible by flying and then stopping flying while above the tile). Which I think is a good further example of the different philosophies - rather than instakill you for what's almost certain to be an oversight from a new player, the game just punishes you so you don't do it again.
[1] https://rhizzone.net/articles/crawl/