"Citing games like Day Z, FTL and XCOM, Jim’s argument made one small mistake: it was all wrong."

October 24, 2012 7:39 AM

A point-counterpoint on the nature of fun in video games:
Games Are Best When Things Go Wrong
Games Aren't Best When Things Go Wrong
posted by griphus (70 comments total)

I'm with go wrong camp personally - or at least with games that feature recoverable unplanned crisis that encourage you to keep going and see what happens rather than throw up your hands and go back to the last save.
posted by Artw at 7:57 AM on October 24, 2012 [1 favorite]

Sometimes grabbing the Amulet of Yendor is all the more exciting because your illiterate pacifist archeologist has died 15 times straight and sometimes you want to jump across the mushroom kingdom knowing that you got the hang of every single jump. Both approaches can be good.
posted by ersatz at 8:00 AM on October 24, 2012 [2 favorites]

Damnit, I was going to post this.

I think both of them are right, and the best games have both sections of flow and chaos and panic. Though either can be fine.

Actually, I'm incredibly addicted to super hexagon now which I don't think really fits into either category.
posted by empath at 8:00 AM on October 24, 2012

I think it's interesting that games that go for concrete narative realism rather than abstract simulation have to rely on the most abstract mechanic of save/reload to make it playable.

Fable 2 is the only narrative game I've played that did away with it, but they did it by removing failure, which I think made the game dull.
posted by empath at 8:03 AM on October 24, 2012 [1 favorite]

I think that games like Super Hexagon (and VVVVV and Super Meat Boy and the Bit Trip games) fit into the "going wrong" category. I mean, you spend most of those games making mistakes until you don't. They don't really allow for careful deliberation on tactics and strategy and you have to make the right decision RIGHT NOW and you probably won't.
posted by griphus at 8:05 AM on October 24, 2012

Actually, I think SuperHexagon falls squarely in the "games are best when FUCK YOU STUPID GAME THAT I'M NEVER GOING TO GET ANY BETTER AT BUT CAN'T STOP TRYING" category...

So I guess it falls in the special "always going wrong" category. At least for me.
posted by MCMikeNamara at 8:11 AM on October 24, 2012

SuperHexagon isn't technically a game so much as an evil relic brought up from the netherworld by malicious forces bent on destruction.
posted by shakespeherian at 8:15 AM on October 24, 2012 [1 favorite]

Oooooh. Is it a roguelike?
posted by Artw at 8:16 AM on October 24, 2012 [1 favorite]

However, the "disaster fantasies" that do it really well aren't, to me, actually stressful. In Dwarf Fortress, FTL, and suchlike, I think there's a distinct feeling that the game is laughing with you, rather than at you. The games that stress or annoy me are the ones that treat failure with a sort of "Looks like you'll just have to get better, loser. Because you lose."

My first thought is that both of those don't just send you back to the beginning when you screw something up. Sometimes you can struggle along gimped and sometimes you die and can try again doing something completely different, unlike in a lot of unforgiving games where if you can't do one specific thing the way the game wants, you're screwed.
posted by Copronymus at 8:19 AM on October 24, 2012 [1 favorite]

Oooooh. Is it a roguelike?

SuperHexagon? No, it's what Keith Burgun would call a puzzle, not even a game. There are no ambiguous decisions (if my understanding of it is correct), just a pattern to learn. Not so?
posted by adamdschneider at 8:21 AM on October 24, 2012

This is SuperHexagon. Just keep your guy from getting hit by walls, basically reflex gaming.
posted by shakespeherian at 8:26 AM on October 24, 2012

Procedurally-generated complex games like Civ, XCom and FTL work well with peril because you have a different experience each time. Death is part of the experience. Linear "N-hours-of-gameplay" games get frustrating because you have to relive the same moments over and over.

I think there should be multiple levels of peril. In FTL you can unlock ships by completing achievements, so that's not complete permadeath. But sometimes it's frustrating when the peril is too sudden (like when you're boarded and suddenly everything's on fire and you die).
posted by RobotVoodooPower at 8:28 AM on October 24, 2012 [1 favorite]

This seems like a good place to mention the perfect deck-stacked-against-you game. It isn't a roguelike. It's an arcade game.

Get your Sega Dreamcast out of storage, and rummage around until you find your copy of Powerstone 2. First, under options, max out the difficulty level at 8. Then, in Original Mode, set up a game on the space elevator level, with yourself as Galuda on the red team, against three Chefs, all on the blue team.

You're unlikely to win at all in your first hundred tries. Keep at it. After five hundred plays, you may win a few.

You have to play a perfect game to win. You have to attack and retreat at the same time. Sometimes you're hit just once and you never recover. Sometimes you have a full health bar up until sudden death mode kicks in and then a damn space bee stings you kaput. Continue? YES!

It's great peril and greater fun, and when you finally win you will whoop and shout and jump for joy, guaranteed. Try it, you'll see. Perfect game.
posted by Ice Cream Socialist at 8:45 AM on October 24, 2012 [1 favorite]

I've been playing Borderlands 2 piecemeal, mainly in that sweet spot that lasts from the last of my evening chores until I need to stop playing games lest my insomnia get triggered. So it's one or two missions a night for me.

I like it when things go wrong based on my choices. So if I gamble that I can take that last Hyperion bot before needing to go rearm myself only to find it had a gaggle of WARloaders waiting near by so suddenly I'm pinned down behind some dumpster, out of ammo for my corrosive weapon of choice, and need to fight my way to safety, that's fun.

I don't like it when things go wrong based on things outside of my control. The fun escape above becomes a tedious chore if for some reason instead of picking up the shieldbooster I need to stay vertical, the controls make me reload my gun instead, leaving me armed only with bad language.

The Minecraft example is not a great one - it's really easy to protect your awesome house from Creepers and to install Plan Bs just in case the green guys show up (keep things well-lit, build second exits, moats, etc). A better one, I think, would be the Arkham games where you are Batman and should be able to dispatch a bunch of mooks easily, but for whatever reason you as a player cannot, thus forcing you to replay the scenario over and over. With the limited time I have to play games, I can't afford games where the punishment for things going wrong is a loss of time. I'm not going to give up my 45 minute playing window to punch some dude with an Uzi over and over again when what I really want to do is punch the Penguin.
posted by robocop is bleeding at 8:46 AM on October 24, 2012 [1 favorite]

My only problem with FTL is how hopeless everything seems after you make one of those thrilling recoveries with three hull left and one crewman scurrying around fixing all the systems. Surviving the battle is one of the best parts of the game, as the first article points out. But the random events aren't going to be nice enough to let you recover and build your arsenal after that point. You are just biding time before death.

I could restart after that point. But what I really want is an option to warp my ship off to a nice tropical planet, record my score as a deserter and hope the rebels destroy this place last.
posted by Gary at 8:46 AM on October 24, 2012 [6 favorites]

"My bad, Mario. I apologize." (YT, NSFW audio)
posted by kurumi at 8:48 AM on October 24, 2012 [3 favorites]

From the related videos for Shakespeherian's link:

WHAT THE HELL TERRY CAVANAGH
posted by Drexen at 8:58 AM on October 24, 2012 [1 favorite]

I am much more in the second camp myself. I just played the heck out of XCOM and I did it by reloading a saved game any time the baddies got the upper hand. I play to win, because I hate hate hate feeling nervous.
posted by rebent at 9:02 AM on October 24, 2012 [2 favorites]

Nice post. Before I read these, I'd already been thinking about this a bit in my role as designer and chief player of my personal Civ4 mod. I think the most un-fun aspect of Civ4 is the fact that so much about the rest of the game depends, Jared-Diamond-like, on your starting position and the first few huts and barb encounters. It was also the case in 2, but more so in 4. (I never played 3 enough to make a judgment, but I assume it's similar.) I find I abandon about ten to fifteen Civ games to every one that I play to completion, largely because I either miss a key wonder I can't go on without, or for reasons of geographical aesthetics.

The fun, on the other hand, can arise from infinite combinations of circumstances, which is probably why it's my favorite game of all time and the only one I consider to have limitless replayability.
posted by AugieAugustus at 9:14 AM on October 24, 2012

All my favorite Civ and SMAC games are the ones where I lost. More specifically, ones that I lost while there were still other multiple computer players in the game.

If it ever happened, my favorite game of Civ or SMAC would be one where things looked bleak halfway through, but then I came roaring back to win. That has never happened, though. I think there's a game design lesson in there someplace: You have to make it so that an early lead is not a guaranteed win without feeling like everyone behind you is getting a speed boost (to mix my metaphor with some Mario Cart).
posted by BeeDo at 9:14 AM on October 24, 2012

I think both of them are right

That's me too. I sometimes want a game that has me in constant crisis, trying to juggle a multitude of factors before things crash; sometimes I want a game where I am powerful and progress is a more straightforward thing.

Understanding that about myself has helped me avoid frustration, because I know now which games to choose based on which mood I am in.
posted by never used baby shoes at 9:19 AM on October 24, 2012 [1 favorite]

I tend to prefer lower-stress games. However, the "disaster fantasies" that do it really well aren't, to me, actually stressful. In Dwarf Fortress, FTL, and suchlike, I think there's a distinct feeling that the game is laughing with you, rather than at you. The games that stress or annoy me are the ones that treat failure with a sort of "Looks like you'll just have to get better, loser. Because you lose."

I used to feel bad because I could never beat Grinder in Dungeon Crawl unless I was playing a Deep Elf summoner, where I could gang-goon him with metric butt-loads of imps and dogs and stay out of his line of sight. I gave up, and looked up the proper strategy for dealing with him on the Wiki:

"Run. Just run! Come back when you've leveled up a few times."

Then I felt like kind of a bad ass, as I had beaten a boss designed to be unbeatable with a 3rd level character. It's also a relief to know the game is cheating to kill you as quickly as possible - no big deal if you croak.
posted by Slap*Happy at 9:23 AM on October 24, 2012

WHAT THE HELL TERRY CAVANAGH

Yeah I specifically didn't link that one because it gives you the false impression that the game is possible.
posted by shakespeherian at 9:30 AM on October 24, 2012 [2 favorites]

But the random events aren't going to be nice enough to let you recover and build your arsenal after that point. You are just biding time before death.

That's the aspect of roguelikes that I feel FTL gets wrong. In most roguelikes, your death generally comes from some horrible thing happening that you could have survived if you had been smarter about it, like running into an unbeatable monster. In FTL it's much more of a war of attrition, where most of your deaths come from completely unremarkable battles when your ship is too damaged for you to do anything to survive. Part of what makes the permadeath mechanic bearable in a roguelike is the idea that if you do everything right (either by being smart or getting lucky) you'll be able to win.
posted by burnmp3s at 9:38 AM on October 24, 2012 [1 favorite]

I thought this was gonna be about Helix Snake's Skate 3 vids
posted by jcruelty at 9:49 AM on October 24, 2012

Just for the record, hexagon is randomized, even though there are patterns. It's closer to pac man than a puzzle game.
posted by empath at 9:58 AM on October 24, 2012

burnmp3s: In most roguelikes, your death generally comes from some horrible thing happening that you could have survived if you had been smarter about it, like running into an unbeatable monster.

Actually, I find that about 60% of my deaths in a traditional roguelike (aside from the early game deaths nobody cares about) are due to carelessness from the game dragging on way too long. You often get overpowered and bored, but in the late game a single typo can kill you, like firing a magic missile sideways in a hallway and getting pinged 20+ times by it bouncing back and forth. One really critical thing FTL gets right is an appropriate length.

Of the rest, I'd say 20% are due to me being obliterated by some monster that I hadn't encountered before but was 10x harder than anything else I'd ever seen, and 20% are due to failing to build up enough power and being smashed by a boss.
posted by Mitrovarr at 9:59 AM on October 24, 2012

I kind of like the recent pattern in some indie games (thinking specifically of Dyad and Sound Shapes) where you can "complete" the game pretty easily, but then there's a parallel track for further achievement that's really difficult. Sound shapes also has the contrast between the original levels, which are fairly easy, and the community levels, which often have a very Super Meat Boy vibe to them.
posted by selfnoise at 10:02 AM on October 24, 2012

Part of what makes the permadeath mechanic bearable in a roguelike is the idea that if you do everything right (either by being smart or getting lucky) you'll be able to win.

I'm not sure if this is correct, but I suppose it depends on the roguelike. My favorite pure roguelike is Incursion, and you can definitely die due to bad luck. To pick a less traditional example, Binding of Isaac depends greatly on the random pickups you get. Bad luck means that the later stages are essentially unwinnable. For most roguelikes, since randomness is an important element, consequences actually matter, and the average difficulty level for a run is supposed to be quite high, it's almost axiomatic that there will be unwinnable situations.
posted by Edgewise at 10:11 AM on October 24, 2012

I've been playing 100 Rogues and either I am playing wrong, or that game really, really depends on luck of the draw w/r/t equipment. But that's not really a traditional roguelike.
posted by griphus at 10:12 AM on October 24, 2012

Starting equipment or stuff you find lying around?

I'm also finding that equipping a future player of a different class via the locker can help that subsequent run a lot, but that feels a tad cheaty.

(Zombie is the ultimate stuff-getter class)
posted by Artw at 10:49 AM on October 24, 2012

Stuff you find lying around to replace the starting equipment. I have had more than one game where I got to the first boss with the starting sword/robe.
posted by griphus at 10:51 AM on October 24, 2012

Also, as far as I can tell, there's no guarantee the locker will show up before you die from being underpowered.
posted by griphus at 10:52 AM on October 24, 2012

Games can be hard and put you in brutal situations as long as when things do go wrong I can determine what I could have done better and then have a way of actually doing it differently when I repeat my actions. Demon Souls and Dark Souls are good examples of this done right. Borderlands 2 can be brutal at times but adjusting your gear and perhaps redoing your character's specs to better fit a specific challenge get you through anything, even when playing solo.

I don't know if I'm more annoyed at games that are "cheap hard" now than I used to be because I have less time or if have just played enough games that do it right that I have less tolerance than I used to. I've been playing the iPad Carmageddon port and find it frequently infuriating as in sequences you need to perfectly time I get rammed by cops or other racers over and over again. That leads me to give up on completing the race and instead focus on destroying the other competitors, which now that I think about it is probably the point.
posted by Blue Meanie at 11:47 AM on October 24, 2012

I guess I should clarify that I'm mostly comparing FTL to Brogue because that is the roguelike that I like the best and have played the most.

My favorite pure roguelike is Incursion, and you can definitely die due to bad luck.

Dying to bad luck is not really what I'm talking about, to me deaths in FTL are less bad luck and more running out of finite resources that you have no way of getting back up to necessary levels. In a traditional roguelike setting it would be like having a hunger gauge that starts full and steadily declines for the whole game with no steady source of food, which would mean most deaths would come from dying of hunger.

To pick a less traditional example, Binding of Isaac depends greatly on the random pickups you get. Bad luck means that the later stages are essentially unwinnable. For most roguelikes, since randomness is an important element, consequences actually matter, and the average difficulty level for a run is supposed to be quite high, it's almost axiomatic that there will be unwinnable situations.

I would mostly just call that bad game design. Ideally a roguelike should be trivial to beat if you had save states, and most of the difficulty should come from making tough tactical or strategic decisions with limited information. If the game is significantly easier or harder based on random factors beyond the player's control, then that doesn't really make it difficult, it just randomizes the difficulty. Ideally a given run in a roguelike isn't any harder or easier than any previous run, it's just different due to random factors that don't necessarily make the game easier or harder.
posted by burnmp3s at 12:16 PM on October 24, 2012

Just for the record, hexagon is randomized, even though there are patterns. It's closer to pac man than a puzzle game.

Is it made up of randomized combinations of these patterns? Is victory dependent on memorizing these patterns and being able to react quickly enough to recognize when they are there? In any case, it still seems reliant on pure reaction times rather than any "decisions" per se.

to me deaths in FTL are less bad luck and more running out of finite resources that you have no way of getting back up to necessary levels

I don't know, stores are all over the place where you can repair your hull. Past the initial couple of games, most of my deaths have come from getting to the final boss with equipment that it is just impervious to, like a missile launcher that has helped me defeat every normal enemy but can't get through the boss's antimissile drone.
posted by adamdschneider at 12:33 PM on October 24, 2012

Procedurally-generated complex games like Civ, XCom and FTL work well with peril because you have a different experience each time.

I think this is key. Failure is fine if it doesn't force me to do the exact same thing over and over again until I get it right (and forcing me to watch the same cutscene more than once is right out).

A few games have a basic mechanic that is enough fun to keep me engaged while repeating the same thing over and over, trying to perfect it. Trackmania, Dustforce, Nitronic Rush, Vini Vidi Vici from VVVVVV, shmups like Jamestown or Crimson Clover are recent examples I can think of. Bit.Trip.Runner had me for a while until I got to a level that had a long stretch of dull, trivial jumps before I could get to the challenging stuff. I tried several times but kept quitting out of boredom when I got sent back to the dull beginning for the fifth time.

But none of those are as good as, say Spelunky, where death sets you back but gives you a whole new version of basic challenges to face with your continually improving skill.
posted by straight at 12:47 PM on October 24, 2012

If the game is significantly easier or harder based on random factors beyond the player's control, then that doesn't really make it difficult, it just randomizes the difficulty.

That sounds good on paper, but I would make two objections. First, if there is true randomization, and if details in the game have significant consequences, it seems really hard to have random elements that don't have much impact on difficulty. Second, although what you say sounds right, in practice, I don't think players really mind some element of randomized difficulty.

Hell, most roguelikes have virtual die rolls, and it's always possible to lose as a result of that. That's one reason why, as you say, a roguelike with save/load functionality may be trivially easy to beat. For game purists in many domains (perhaps especially board games), randomness is distasteful, but in practice, the "fun" doesn't always come entirely from the experience of a fair test of skill.
posted by Edgewise at 1:05 PM on October 24, 2012

Procedurally-generated complex games like Civ, XCom and FTL work well with peril because you have a different experience each time.

I think this is key. Failure is fine if it doesn't force me to do the exact same thing over and over again until I get it right .

Part of it, though, is maintaining enough familiarity that, given a particular constellation of circumstances, you have a vague notion of how the rest of the game is going to play out by the time you're five minutes into it. CivIV's replayability, to me, is equal parts "hey, early luxury items I can use to power my economy and not get boxed in!", "oh shit, I share a border with Montezuma and he's going to invade in the BC's", and "oh my, the entire western hemisphere I just discovered is a Buddhist love-fest, they hate me for being Hindu, and they have gunpowder armies." It's a balance between your ability to recognize and react to familiar situations in a way you know will work, and the random number gods being able to smite you at a moment's notice with something you've never seen in a thousand previous playthroughs. Almost no one ever gets it right, but games that strike the proper balance are poetry.

I don't think I've ever been more satisfied with a gaming loss than I was when an AI Ghandi unexpectedly rolled an invading force of tanks onto my undefended western flank while I was busy landing a marine force on Boudica on the eastern front. That backstabbing son of a bitch, I still target him first in games five years later
posted by Mayor West at 1:11 PM on October 24, 2012

most of my deaths have come from getting to the final boss with equipment that it is just impervious to, like a missile launcher that has helped me defeat every normal enemy but can't get through the boss's antimissile drone.

Yeah, I don't really get the need for a non-procedural, pre-scripted boss battle at the end of FTL. It's the real cause of the hopelessness I feel after those close battles. I survived, but after the hull repair bill I won't be able to afford the proper upgrades. So even if I play a really tight game and manage to survive the final boss is going to wipe me out.

I really wanted to like 100 Rogues but each game is 3 pretty simple levels just to get killed by the genie because I don't know whatever trick is needed to beat him. Maybe I don't have the right items or something, but if you are going to serve me up non-random levels then you should give me save points.
posted by Gary at 1:21 PM on October 24, 2012

I would mostly just call that bad game design.

Well, a crucial part of binding of isaac's game design is that it is extremely short, a 'casual roguelike' more or less. So while it is perhaps more possible for luck to screw you over than in e.g. nethack, you can get back to the same point without much time investment. (I also think that more things are survivable in binding of isaac than it might initially seem.)

If the game is significantly easier or harder based on random factors beyond the player's control, then that doesn't really make it difficult, it just randomizes the difficulty.

Part of the metagame for most well-designed roguelikes is risk management. Part of this is things like ID minigames (i.e. learning all the techniques in nethack or whatever for ID'ing stuff while minimizing risk to you) but much of it is just preparing based on what the random events could be. I realize that this isn't something everyone enjoys, but randomization doesn't randomize the difficulty from this perspective -- once you have some knowledge of the possibilities, it provides a reasonably constant upper bound on the difficulty that you can prepare for.
posted by advil at 1:26 PM on October 24, 2012

To beat the genie, all you have to do is beat the crap out of him in melee. Generally, I hit that first scarab in front of me with the hammer attack (which kills it,) get up close and just melee the Genie until the lamp breaks. Then you head up one space and keep doing that. For a finishing move, I wait until I have two scarabs around me and hit the area attack that'll do 30 damage.
posted by griphus at 1:27 PM on October 24, 2012

I love games that are tough and failable, but if you make failure a part of the game you need to design around it. Maybe the player can start over instantly, and really only loses 10 seconds or so, like in Super Meat boy (The iOS/Android rayman game that just came out is a great example of this too). Maybe the game gives you a reasonable chance to recover when you just survive, which means that failure comes with a cost, but pyrrihc victories don't feel like the norm. (Because let's face it, it's fun to have to struggle to win, but a death spiral is no fun, especially if you know that starting over comes with a significant cost.)
posted by aspo at 1:28 PM on October 24, 2012

Er, that's with the Crusader, which is the only class I really play with.
posted by griphus at 1:28 PM on October 24, 2012

Things need to get built up, but they also need to get smashed.

Make it fun for me to build things.
Make it fun for me to smash others' things.
Make it fun for me to dance on the edge of destruction, avoiding getting smashed.
Make it fun for me to deal with my things getting smashed.
posted by fleacircus at 1:28 PM on October 24, 2012

My main problem with FTL are as follows:
1) It doesn't tell you there will be a boss battle, so my best game ever I finally make it, then it kicks me in the teeth.

2) You can't go back and build up strength. If you go the the wrong systems and don't find what you need, too bad. I think a critical element of games like Civ, Nethack, X-COM is if you get out of your current problem, almost always have a chance to retreat, lick your wounds, and can come back to it. So I screw up one level in Nethack. I can retreat, go back to my stash, rearm and try again. FTL I've got that red line that you are racing against. Either make me race, or make me build up strength. I can't do both!

Dwarf fortress you can rebuild most of the time, as long as you get some new settlers and can get things going again. But there isn't a time limit there.
posted by Canageek at 2:17 PM on October 24, 2012 [2 favorites]

The Crusader is TBH probably the most gear dependant class - the rest are all about the skill tree. Beating the crap out of the genie is probably the best technique with him, though other classes require more finesse.

Potion management is important, and in a pinch you'll need to judge if potions of energy will make more impact replenishing your skills or as a bomb. At that level it's probably bomb - that said you might want to hold on to them for later.

TBH I pretty much know if I have the gear/skills to be beat the Genie ahead of time now, and though more often than not I do if I don't I'll just ditch that run and start another. Same goes for the next boss.

Satan I've never beaten, he may require a different strategy.
posted by Artw at 3:35 PM on October 24, 2012

Necromeister is my current favourite class, and that one is a shit to play until you've leveled up a couple of times and one of the harder ones to get past the genie.
posted by Artw at 3:47 PM on October 24, 2012

Thanks to MetaFilter, I just discovered the new XCOM is out, and I have exactly zero hardware that can play it. Why do I even read these threads?
posted by vibrotronica at 4:17 PM on October 24, 2012

If it's an consolation 100 Rogues is very cheap.

*iOS device required.
posted by Artw at 4:19 PM on October 24, 2012

I've been saying this for a long time: without consequence and chance of failure, there can be no victory.

There is absolutely no tension in playing X-COM when you know if you make any wrong move, you can just load a save point and redo it.

X-COM however is brilliant when played in Ironman mode, which disables saving. I would argue that this is how the game should have been played - almost like a roguelike, because it's such a short game anyway.

You need to think about building a larger team of soldiers (maybe 15) and do some succession planning, because, simply, some people are going to die in the war. You look at everyone in your team and think, if he dies, what's plan B? You need to bring a balance of veterans and rookies into missions, so you can train the rookies, and you will tweak that balance based on how difficult you think that mission is. If you bring all your veterans on one mission, what happens if you get unlucky and a sectopod wipes them all out?

And the death of your men really really hurts, some of whom have run 10 or 20 missions with you: just how you've come to rely on a particular sniper or medic, and now you have to rely on rookies to fill those roles. There's the temptation to simply bench a sniper when he gets too high level, to keep him in "reserve" and let rookies run missions instead so you can build up some backups. You actually make the decision to risk the life of your assault soldier veteran you've been running missions with for a week with in order to capture a live Muton, so the rest of your squad will eventually get plasma weapons... if he dies, it's for the greater good and all. I've lost some good men that way.

Playing on X-COM with reloading enabled feels like a pale shadow of what the game should be, but I know many people who might disagree with me...
posted by xdvesper at 5:36 PM on October 24, 2012 [3 favorites]

xdvesper, the counterpoint to that is the second article. However, playing games that have not been designed for it in an ironman 'mode' by refusing to load after you die i.e use one save strictly as a bookend, can offer a new experience and prolong the life of a game.

Well, a crucial part of binding of isaac's game design is that it is extremely short, a 'casual roguelike' more or less.

On the other hand, the more you play, the longer and harder it gets. I'd play it in short bursts, but after a while you need to budget a good chunk of time. However, I like that the game adjusts to your initial skill increases.
posted by ersatz at 6:02 PM on October 24, 2012

From DCSS's design document:

The notions of balance, or being imbalanced, are extremely vague. Here is our
definition: Crawl is designed to be a challenging game, and is also renowned for
its randomness. However, this does not mean that wins are an arbitrary matter of
luck: the skill of players will have the largest impact. So, yes, there may be
situations where you are doomed - no action could have saved your life. But
then, from the midgame on, most deaths are not of this type: By this stage,
almost all casualties can be traced back to actual mistakes; if not tactical
ones, then of a strategical type, like wrong skilling (too broad or too narrow),
unwise use of resources (too conservative or too liberal), or wrong decisions
about branch/god/gear.

The possibility of unavoidable deaths is a larger topic in computer games.
Ideally, a game like this would be really challenging and have both random
layout and random course of action, yet still be winnable with perfect play.
This goal seems out of reach. Thus, computer games can be soft in the sense that
optimal play ensures a win. Apart from puzzles, though, this means that the game
is solved from the outset; this is where the lack of a human game-master is
obvious. Alternatively, they can be hard in the sense that unavoidable deaths
can occur. We feel that the latter choice provides much more fun in the long
run.
posted by tychotesla at 9:05 PM on October 24, 2012

Anyone can make a hard game. It is very difficult to make a hard-but-fair game, and it doesn't help that lots of players these days, when they lose, will blame the game first and themselves last.

Disaster games are fun because they are a challenge. They are about doing the best you can against high odds. Old arcade games were all like that. Donkey Kong: How High Can You Go? Pac-Man: have you gotten to the Apple level? How about Melon? Defender: can you even finish a single wave?

For that to remain interesting, you have to have an avenue towards improvement. If you always get to level 4, but never past it, why would you keep playing? Eventually you're going to write it off as hopeless. Also, failure has to feel like it's your fault. If you always die to a certain boss, you are only going to try it so many times before you give up.

Both articles make points, but the first one is the stronger argument, because it presents a kind of experience that only games can provide. Empowerment fantasies? That's like half the movies in theaters these days. You want to experience wonder? Try Tolkien.

But really, challenge itself is just a special case of system exploration. Can you survive level 7? It's the same impulse that leads people to play around in SimCity, a game without real challenge, or establish a Dwarf Fortress in a Sinister biome, which is not what one would call the most viable survival decision. All these cases are the same: you are satisfying your curiosity, in some way, regarding the behavior of the system, whether that's determined by how far you get, or whether your buildings grow, or whether the dwarves can successfully found their fortress. It's not necessarily that the game is hard, it's that you can allow yourself to get caught up in the question of what happens in the game.
posted by JHarris at 9:50 PM on October 24, 2012 [1 favorite]

I really wanted to like 100 Rogues but each game is 3 pretty simple levels just to get killed by the genie because I don't know whatever trick is needed to beat him.

When Keith Burgun was originally designing the game, he told me how excited he was about the bosses he was putting in, how they'd be a solid Super Nintendo-style challenge, which was part of his motivation when he made the game, to make a game in that mode. He's improved a fair bit as a designer since then, and I think he regrets putting those bosses in now.
posted by JHarris at 9:58 PM on October 24, 2012

He's improved a fair bit as a designer since then, and I think he regrets putting those bosses in now.

Indeed, I do, John! Those were a mistake. Also, 100 Rogues is highly random. With all of that said, I do think that the "tricks" to beating those monsters are pretty simple, and in 2.5 years since the game has been out, we haven't really gotten many complaints about this problem. Although I still do think it's this weird "spoiler" thing, and there's too many of those in 100 Rogues and in roguelikes in general. With all of that said, I'm still pretty proud of the game, as I think it's charming and mostly fun to play, if in a sort of dumb way.

My next game is a similar KIND of game, but with literally no output randomness (i.e. no dice rolls for combat, critical hits), and there aren't magical +1 item nonsenses, and there isn't even experience points! It's pretty crazy. Plus, it's hex based. Feel free to check it out, we ran a successful Kickstarter campaign for it: http://www.kickstarter.com/projects/dinofarmgames/auro
posted by keithburgun at 11:23 PM on October 24, 2012 [3 favorites]

Really looking forwards to it.

(And I think you're under-selling Rogues there)

(And I kind of like the way the bosses break things up)
posted by Artw at 11:34 PM on October 24, 2012

Honestly, the worst part is that the iPhone is such a great device for these small roguelikes, but the options are still so limited. There's a few other good ones (Zaga-33 is a 7DRL and great fun, Cardinal Quest didn't really hook me but does work really well on the device) and a few more games that try to bring all of nethack onto a tiny screen with no keyboard. If I had this complaint about say, an endless running game, it wouldn't matter because there would be 5 more out next week.

Also, the desktop dungeons guys need to stop being so delicate with their game and just put it on the iPhone already. They can keep tuning the gameplay after they collect everyone's $2.99.
posted by Gary at 8:53 AM on October 25, 2012

It's a better situation than Android, which has next to nothing.
posted by Artw at 9:00 AM on October 25, 2012 [1 favorite]

Android is a great, big games desert. That's one of the primary reasons I got an iPad, actually.
posted by adamdschneider at 10:31 AM on October 25, 2012

Yeah I can't imagine why DD isn't on iOS, but then again neither is FTL and WHYYYYY
posted by griphus at 10:34 AM on October 25, 2012

I dunno about games desert - I see lots of Android games in other categories, but for roguelikes there's a couple of clunky primitive looking things and that's it.
posted by Artw at 11:26 AM on October 25, 2012

Well, I mostly go in for board game ports on iOS, too, which is not a friendly space on Android.
posted by adamdschneider at 2:05 PM on October 25, 2012 [1 favorite]

Yeah adamdschneider, the board gaming space on Android is much less robust than on iOS, which is a shame. A lot of iOS board games are made by SAGE, who made the apps for Puerto Rico and Le Havre, and who appear to be an iOS-only developer, at least for the moment.
posted by JHarris at 3:16 PM on October 25, 2012

Not only that, although there may be games available for Android, it's very rare that I hear about good games there. I can think of literally one neat game I have ever heard of available only for Android: Red.
posted by adamdschneider at 6:47 PM on October 25, 2012

I don't care if it's only available for Android, although the Android version of Carcasonne I hear has a much worse interface than the iPad version (which I can vouch, has a terrific interface).

I rather like the board game Don Quixote on Android. But it's more a solitaire game than something you play against a human or AI player.
posted by JHarris at 7:55 AM on October 26, 2012

I did buy Don Quixote. Heard about it on Boardgamegeek's iOS Board Games, which is an incredible resource for this sort of thing.
posted by adamdschneider at 10:00 AM on October 26, 2012

Don Quixote takes some practice to get good scores, but even then there's a bit of luck required. My best score so far is 79. I'm not sure how I'm supposed to get the 90 points needed to get the Veto unlock. I still love the idea though, basically Carcassonne in reverse.
posted by JHarris at 11:59 AM on October 26, 2012

Unpredictability and control in turn-based combat: an examination
posted by Artw at 7:10 AM on November 11, 2012

X-COM - a novel
posted by Artw at 7:56 AM on November 11, 2012