Crawl tourney

[pat]

the setting isn't a huge problem, it's just not really a selling point. The best bit is probably the work that's gone into the gods but other games do that better, ie Incursion in my opinion.

[reaver]

it's a must-play for roguelike fans but I'm glad that it's largely in my past.

I don't understand this. Roguelikes are not literature ("War and peace is a must-read and I'm glad I don't have to read it anymore"). How can one say that this is a must-play while actively avoiding it? If it's such a great game, I would play it. If the game is too difficult/slow, but the end-game content is great, I would try to see the end-game content with dubious means without spending a lifetime slogging through it.

[pat]

I don't understand this. Roguelikes are not literature ("War and peace is a must-read and I'm glad I don't have to read it anymore"). How can one say that this is a must-play while actively avoiding it? If it's such a great game, I would play it. If the game is too difficult/slow, but the end-game content is great, I would try to see the end-game content with dubious means without spending a lifetime slogging through it.

because I finished it a bunch of times and experienced everything it has to offer? That's the only reason I avoid it now. It's a great game that I'm largely finished with.

[reaver]

because I finished it a bunch of times and experienced everything it has to offer? That's the only reason I avoid it now. It's a great game that I'm largely finished with.

Ok, probably misunderstood you. I'm usually sad if I've seen everything a game that I've loved has to offer, not glad -- Glad that it has been part of my past, which is probably what you meant, not glad that it is past.

[Quendus]

(I know people like to say "sci-fi!" but this is, in my opinion at least, a departure from the genre and in any case runs into issues of lacking a familiar bestiary, lacking familiar weapons, lacking magic, etc., etc.

Huh? The average sci-fi story has 10 times more magic than the average fantasy story!

[awake]

I like all the weird class and race choices in DCSS or Nethack for that matter. Playing as a cat or a vampire beats having five different races of elves, coherence be damned.

[jim]

I like all the weird class and race choices in DCSS or Nethack for that matter. Playing as a cat or a vampire beats having five different races of elves, coherence be damned.

Totally. The possibility of unconventional builds is very alluring. For instance, the Octopode has a grapple attack that can be in effect against multiple opponents simultaneously. So: grapple one guy, do damage every round, grapple the next guy, do damage every round. Very cool indeed. However, anything sub-optimal tends to be very dangerous, Octopodes have fairly low HP.

[Vanguard]

35 floors is too much for your taste? Wow. If it's less than 50, I want my money back!

I'm really baffled by a lot of the commentary to this effect around here. So what's the optimal number of floors to the prevailing tastes? 20? That seems really short to me, for a roguelike.

I want to hear your opinion on this.  How long do you think roguelikes should be, and why?

[mushroom patch]

35 floors is too much for your taste? Wow. If it's less than 50, I want my money back!

I'm really baffled by a lot of the commentary to this effect around here. So what's the optimal number of floors to the prevailing tastes? 20? That seems really short to me, for a roguelike.

I want to hear your opinion on this.  How long do you think roguelikes should be, and why?

Okay. Well, first, I think crawl is fine even at its minimum length of ~35 levels + ascension run as you said. The games I've played the most are moria, angband, zangband, and tomenet, in other words moria derivatives. Hence my 50 levels remark. (Obviously there's some question about whether you really have to play 50 or 100 levels in an angband variant the way you do in Crawl -- in fact, you usually play more than that, but only the small piece you need to get what you want from each level you play and you can dive 20 levels without even fighting any monsters.)

Of course the original rogue was itself pretty short and larn even more so, so there's plenty of room for short games in the genre.

All that said, to my taste, longer remains better for a pretty long time. Why? Essentially two reasons:

First, length increases the significance of a play and especially of wins. The character develops, in its roguelike way, more. Longer games increase the stakes in the game and therefore the atmosphere of danger and risk. Related to this is the sense of loss when your character dies. I understand most people would call this "frustration" or otherwise find it stressful -- I've definitely had frustrating roguelike experiences myself -- but I think this is the path to the dark side. Loss and regret increase the emotional depth of the game.

Second, there's just more stuff in a longer game, given acceptable execution on the author's part. When you try to put the same amount of stuff in a shorter game, I don't believe it can really gel properly. The key thing then is just knowing what to do with all the crap everywhere, not actually finding it. This means more room for turning points in the game -- "Then I got this thing so I could do this and go there, where I found this thing, etc." vs. "Then I got this thing, so I ascended." It's not just items, of course, there's more monster types, more contexts, etc.

I know the current thinking is about intensity (or variation, which I don't believe in yet), which sounds fine. I think the way Sil cuts out the part of the game where you beat down varmints for an hour before anything interesting happens is great, for example. (Similarly, "make your own" items reduces the need to scum/farm/whatever for equipment.) I'd have to win a game before I decide whether 20 levels is good or wack.

Really, I'm reacting to a change in thinking away from ambitious, large world roguelikes toward more bite sized offerings. I just don't see the genre realizing its potential by paring down the number of items, monsters, and levels. Why not just go full graphics then? If you have few enough monsters and simple enough items, why not make it an action RPG?

My feeling is that the strength of roguelikes comes from the lack of graphics and procedural content. Given these two advantages, developers should be able to wildly outstrip what's possible in the current commercial environment in terms of world building (and by the way, there's a real lack of original thinking about how this can be accomplished via the current internet culture).

Anyway, this is turning into another unfocused ramble, so I better stop.

[Vanguard]

Good post.

I disagree with the idea that short roguelikes are inherently unambitious.  I play tons of shmups and the better ones fit an enormous amount of depth into 30 minutes of content.  There's no reason why roguelikes can't do the same.

Another issue is that most long games rely on filler content to pad out their length.  If a developer really has 50 floors' worth of unique, worthwhile ideas and wants to make a 50 floor game, I won't tell them not to.  Crawl starts to repeat itself itself well before the ending and Angband is nothing but repetition.  Compare that with Brogue where every enemy stands out in some way, every element brings something to the game.  Brogue is also consistently generates floors that meaningfully change your options in a way that Angband and Crawl can't compete with.  Lava, pits, and traps all bring both opportunities and dangers.  Treasure rooms can alter the course of the rest of the game.  Nothing goes to waste.  Crawl, on the other hand, is perfectly happy to let you o + tab your way through orcs for a good hour and Angband hardly lets you do anything else.

e:

I really ought to proofread my posts before posting.

[LazyCat]

I have to say, I find this thread a little sobering. Here's a bunch of self-professed roguelike fans/developers saying they don't have the patience for one of the standard bearers of the genre.

Told ya.

[jim]

I have to say, I find this thread a little sobering. Here's a bunch of self-professed roguelike fans/developers saying they don't have the patience for one of the standard bearers of the genre.

Told ya.

To be fair, I've kicked plenty of ass in Crawl in my time. My statement was more akin to "man, I can't party like I used to in my 20's" rather than "whiskey tastes yucky!"

[mushroom patch]

Another issue is that most long games rely on filler content to pad out their length.  If a developer really has 50 floors' worth of unique, worthwhile ideas and wants to make a 50 floor game, I won't tell them not to.  Crawl starts to repeat itself itself well before the ending and Angband is nothing but repetition.  Compare that with Brogue where every enemy stands out in some way, every element brings something to the game.  Brogue is also consistently generates floors that meaningfully change your options in a way that Angband and Crawl can't compete with.  Lava, pits, and traps all bring both opportunities and dangers.  Treasure rooms can alter the course of the rest of the game.  Nothing goes to waste.  Crawl, on the other hand, is perfectly happy to let you o + tab your way through orcs for a good hour and Angband hardly lets you do anything else.

e:

I really ought to proofread my posts before posting.

Yeah, what you say here is fair enough, but I think this kind of analysis suffers from too much of a modern game design-y reading of crawl (and maybe angband). So you say, for example, it's sub-optimal that you can/must tab/o-move your way through significant parts of crawl. This is indicative of "filler content," which is bad. The criticism is local: It's about turn to turn game experience. Is the player engaged, are new things happening every minute, etc.? More than that, "filler content" is so bad it's not enough to just offer mechanics to get through it relatively quickly like tab/o-move.

How do you remove this "filler" from crawl? Well, how do you get into a position where you're tabbing through orcs for an hour? I would say you do this by clearing most of the lair first and coming back to the orcs. The player makes a strategic choice that makes one part of the game easier. If tabbing through orcs is so bad though, it must be that it was a mistake to allow the player to do this. What's the solution? Remove the orc cave? Place it after the lair and make it a troll cave instead? It sounds like the solution has to be to straighten out the game so that it's more linear and streamlined.

Streamlined is exactly what you seem to advocate, but more streamlined is going to be more linear (the shortest path between two points is a line -- a fact so robust that it's true in a huge range of situations and not merely as a metaphor). If you're able to do things out of sequence, then it's going to be possible to make a lot of game content trivial, even boring. (This is ignoring the fact that it's fun to play a character you built to be overpowered in some situations.)

I disagree with the idea that short roguelikes are inherently unambitious.  I play tons of shmups and the better ones fit an enormous amount of depth into 30 minutes of content.  There's no reason why roguelikes can't do the same.

First, short roguelikes are shrinking away from the potential of the genre, paring down their ambition (probably in response to widespread belief that ambitious roguelike projects rarely materialize -- a belief with some justice to it). Roguelikes should aspire to be complex and meandering, full of unnecessary odds and ends. If they don't, they're tending toward what other genres do better with graphics, real time mechanics, and so on.

As to shmups vs. roguelikes, I would need clarification of what you mean here to address it. Are you saying that a lot of practice (hence "depth") goes into winning a shmup with 30 minutes of back to back game content without deaths and retries or are you saying the density of a 30 minute session of a shmup vs. a roguelike should be comparable? (In the latter case, I would disagree with the implication that the shmup has the better roguelikes beat on this front, even the more traditional, longer ones.)

[Vanguard]

How do you remove this "filler" from crawl? Well, how do you get into a position where you're tabbing through orcs for an hour? I would say you do this by clearing most of the lair first and coming back to the orcs. The player makes a strategic choice that makes one part of the game easier. If tabbing through orcs is so bad though, it must be that it was a mistake to allow the player to do this. What's the solution? Remove the orc cave? Place it after the lair and make it a troll cave instead? It sounds like the solution has to be to straighten out the game so that it's more linear and streamlined.

That's a good point.  Imbalanced difficulty is inevitable in Crawl's non-linear structure.  I don't really mind that that, though.  Bumpy difficulty curves are more interesting than a smooth ones.  The problem is that Crawl's mid game is overall really easy compared to the early and late parts of the game, and it goes on for a very long time.  As long as you choose your order well, almost none of it will be difficult.

You could always advance ahead to make the game more exciting, but that's a bad solution.  Roguelikes are about using every tactic, every ability, doing anything it takes to win.  The genre's punishing nature encourage risk-averse play, even more so in long roguelikes.  The player should not have to make bad decisions to stay challenged and entertained.  Managing a game's difficulty and pacing are the developer's job.

Another problem is that by about the halfway mark I feel like I'm retreading old ground.  Two hours ago I was fighting ogres.  One hour ago I was fighting hill giants.  Now I'm fighting an ettin.  The location and monster type are different but I'm still pressing tab against an enemy with high health, high attack power, and no special abilities.  Even if it was harder it still wouldn't be interesting after the 30th time.  Crawl's enemy list is large and diverse, but that doesn't always translate into a need for diverse tactics.  Against large groups I find a tight hallway and press tab.  Against ranged enemies I wait around a corner and press tab.

It doesn't have to be that way.  Crawl can be really intense when it wants to be.  Vault 5 drops you into a tough situation where you can't get careless.  The fire orbs in Zot are some of the scariest enemies I've ever seen in a roguelike.  I'd like Crawl much better if it had more moments like those and if I didn't have to cast fireball on yaks for five hours to see them.

What I was saying about shmups is that the better ones manage to pack a huge amount of variety, challenge, and choices into about 30 minutes of content.  I think a winning game of Battle Garegga involves more meaningful decisions than a full game of Crawl does, even though it's less than an hour long.  The reason is because something important is always happening.  Usually several things are happening at once and you can't afford to ignore any of them.  That's what I want to see in roguelikes - for something important and exciting to happen on every floor.

[Vanguard]

First, short roguelikes are shrinking away from the potential of the genre, paring down their ambition (probably in response to widespread belief that ambitious roguelike projects rarely materialize -- a belief with some justice to it). Roguelikes should aspire to be complex and meandering, full of unnecessary odds and ends. If they don't, they're tending toward what other genres do better with graphics, real time mechanics, and so on.

I like roguelikes primarily because they are single player games that are challenging enough that I have to take them seriously, and because the better ones have enough depth and variety to stay interesting in the long term.  For those purposes, short roguelikes are just as good as long ones, and avoid some of the problems and frustrations of long games.  I like "unnecessary odds and ends" too, but they aren't my top priority.

I might be wrong, but it sounds to me that what attracts you to roguelikes are their large possibility spaces.  Most genres only give you a few verbs to interact with the world.  Move.  Jump.  Shoot.  Roguelikes are different.  Roguelikes let you reshape your environment, and wield monster corpses as weapons.  They dedicate specific commands to cleaning your ears and washing your face.  They let monsters level up from killing your NPC allies.

If that's what you're looking for, you'd be better off with Nethack-style roguelikes than the likes of Crawl or Angband.  I recommend Ragnarok.  It's non linear and chock full of oddities and complex interactions.  Even though it's long, it doesn't have much in the way of filler content.  Nearly every area introduces something new or has some unique feature.  Ragnarok's tactical combat isn't anything special, but if you want to get your character drunk and see what happens when they mispronounce the words on a magic scroll, it's as good as it gets.