Halls of Mist is coming soon

[getter77]

Hmm, given the timing, I humbly, yet strongly, suggest you strive to release this for ARRP 2012 for even better visibility than usual AND probably a guarantee of some reviews, feedback, and impressions moreso than otherwise due to folks covering the entrants.

http://roguebasin.roguelikedevelopment.org/index.php/2012_ARRP

[Mikko Lehtinen]

Hmm, given the timing, I humbly, yet strongly, suggest you strive to release this for ARRP 2012 for even better visibility than usual AND probably a guarantee of some reviews, feedback, and impressions moreso than otherwise due to folks covering the entrants.

http://roguebasin.roguelikedevelopme....php/2012_ARRP

Wow. The timing is perfect indeed. Thanks!

[Mikko Lehtinen]

Sounds good, although isn't that what the ladder would do anyway? I guess a competition adds a time limit.

Yes, that's right.

Another possibility I thought of would be to limit the number of tries ("lives") to three or so. A generous time limit and only three tries per player would result in a very different kind of competition.

Overall I'm excited to see how Halls of Mist plays, and very pleased that you're making a thread to provide documentation. I tried Fay at one point after reading your comments about design philosophy, but lacking documentation I didn't find the mechanics transparent enough for me to engage with (and I guess I wasn't prepared to put the time in to learn them by trial and error).

The base mechanics, especially the skills, are more transparent this time around. [I]nspecting items gives a lot of information about new item uses, success chances, etc.

On the other hand, there are some new mechanics that may not be so intuitive (yet).

I hate the feeling that I could be increasing my chances of winning if only I did this thing I'm not going to bother with because it's dull. I enjoy working out how to do well at games, and trying to optimise my play, but I play games to have fun. So having optimal play be fun is an obvious requirement for me to want to engage with something as a game (at least in this sense of game; open ended exploration can be fun too, but feels less gamelike). It doesn't need optimal play to be the most fun way, but that's a good thing to aim for.

The designers of Magic the Gathering are firm believers in this design theory.

Magic is at most fun when both players have creatures that fight each other. But in the past creature cards weren't very good. Most kitchen table decks were full of creatures, and people were having fun. But the tournament scene used to be dominated by Necropotence decks and combo decks without any creatures. Lots of competitive people left the game at that point.

After that experience, the designers of MtG have spent lots of time trying to align the fun with optimal. They claim that's one of the biggest reasons to why the game is selling so well at the moment.

I guess roguelike players can be similarly divided into "kitchen table players" and "tournament players".

I think that the strategic choices come from not making it obvious what the optimal play in lots of situations is. Trying to increase the number of circumstances in which there's more than one viable alternative is how this kind of gameplay is improved.

Absolutely. I find Sangband and Oangband to be especially good at this. Leon Marrick succeeded in making lots of different ways to kill monsters viable.

I'm making an effort to offer choices, too. For example, I've tried to make all abilities important for all classes, and invented alternative uses for many items.

On the other hand I'm not convinced by your anti-(summon-scumming) rule (though of course I agree that you need some rule there). It's complicated in an artificial way -- as opposed to complexity which comes from following flavour, which is easier for players to understand. And it must really be understood for optimal play, which is likely to involve farming the summoners somewhat, but not beyond the bounds of safety.

Thanks for pointing this out. I suspect that LostTemplar found the mechanic unintuitive also, and some of his criticism arose from that.

I love it when people to point out unintuitive mechanics to me. Or strange pieces of flavour, for that matter. When I design games, I always get lots and lots of ideas, and I try many of them in practise. Only after many iterations I'm able to find mechanics or pieces of flavour that I'm comfortable with.

Sometimes after I've been testing a mechanic for months I realize that it's actually too complicated. I'm happy to kill my darlings if it's necessary to make the game more elegant.

How about a really simple rule like giving no experience or drops for summoned monsters? Either by setting a flag on the monster, or by duplicating monster entries to things like "illusionary grue", which is identical to a regular grue except that it has no drops, gives no experience, and doesn't regularly appear in the dungeon, but sources of summoning always create only illusionary monsters.

Sounds good. I've spent the day picking berries from currant bushes, and I've had lots of time to think. This is one of the alternative mechanics I thought of.

[Scatha]

Glad to hear you are also pursuing elegance! Good analogy with Magic the Gathering too. As well as making creature-based strategies better because they are fun, they try to make it not obvious which creature-based strategies are the strongest.

Sounds good. I've spent the day picking berries from currant bushes, and I've had lots of time to think. This is one of the alternative mechanics I thought of.

The fact that we independently arrived at the same mechanic is definitely a sign that it's decently straightforward. I actually just came back to this thread to alter my earlier post to say that of the two slight variations I suggested, having differently named monsters (with "Illusionary" or "Phantasmal" or some such prefix) is almost certainly superior, since it is so much more transparent to the player.

Edit:

The base mechanics, especially the skills, are more transparent this time around.
[...]
I'm making an effort to offer choices, too. For example, I've tried to make all abilities important for all classes

Well, you asked for questions, so: what are the skills and how do they work? Are the abilities something else?

(Feel free to ignore if you have more of a planned order of explanation, but that might give you something to run with if you're not sure.)

[ekolis]

Not sure if "illusionary" is actually a word, but whether or not it is, for some reason it grates on me a bit and I'd prefer "illusory"!

[Mikko Lehtinen]

Well, you asked for questions, so: what are the skills and how do they work? Are the abilities something else?

Attributes and skills are defined just like in Angband. Attributes affect skills based on a chart, in a classic 80s RPG style. I'm just trying to improve on the classic design. I want to offer more character options and make everything more transparent. Tomorrow I'm planning to write extensively about attributes and skills.

BTW, I think I like Phantasmal. But I'm not sure whether illusionary flavour fits my world. Some of Mist's most iconic summoners are different kind of Hags. They summon faeries, for example bogles and redcaps. Opinions on this?

[LostTemplar]

Originally angband, and most other 'classical' RPG's are 'scummy' by design. Why it is so?
1) It have fixed 'danger level' associated with dungeon level, so player can control, how strong monsters to 'farm'
2) It have boss, that is by far stronger, then anything else, so it creates boring 'farming preiod' when anything but boss is a pushover, but boss is still too powerfull.
3) Reward for kills is not linked with difficulty, or is linked very loosely, so it is possible to slowly gain power without risk.
4) Power gain is permanent, no matter what character do, it becomes stronger as time pass.

while fixing 2 and 3 may be too radical, 1 and 4 may be easily fixed

E.g. if you make monster's spawn as MAX(c_lvl, d_lvl) it will immediately fix 'stay at low depth' problem.

Also adding some permanent injures (as in dwarf fortress), permanent stat and level
drain, overall 'monsters become stronger with time' so character become relatively weaker
etc. may reduce intention to farm.

Well, time limit also works, but it seems somewhat nobrainer, and also IMHO, optimal play style should be cautious play, not fast and reckless.

Actually some of your 'anti scumming' changes have an opposite effect, they increase intention to scum as much as allowed. An example with summon monster scrolls was a proof. Same thing with SIL time limit: it forces to go as slow as allowed, instead of as fast as player can (it would be the case if e.g. monster power will grow with time).

[Mikko Lehtinen]

Actually some of your 'anti scumming' changes have an opposite effect, they increase intention to scum as much as allowed. An example with summon monster scrolls was a proof. Same thing with SIL time limit: it forces to go as slow as allowed, instead of as fast as player can (it would be the case if e.g. monster power will grow with time).

Yeah, Summon scrolls are just as abusable in Mist as in all other ironman-style *bands where they exist.

I get the impression that you still prefer ironman-style play to Vanilla Angband? Even though you perceive that even ironman-style *bands have serious design problems?

EDIT: I think in Sil you should dive as fast you think you can handle, so that you can spend more time deeper in the dungeons, where the rewards are bigger? Halls of Mist works quite similarly, except that you clear complete levels.

Both games use the concept of Minimum Depth which increases as time passes (Sil) or when you navigate deeper in the dungeons, choosing how many levels to dive at a time (Mist).

[ekolis]

In many games of Halls of Mist is probably unwinnable unless you get lucky. Random content + limited dungeon levels tends to do this. Killing breeders or summoned monsters, if it was profitable, would make these games winnable. Just like you said: "It is definitely bad if some boring actions are required to win."

Wait, did you just say the game frequently generates scenarios that are unwinnable from the point of character generation? That sounds more than a bit cruel!

[Mikko Lehtinen]

Wait, did you just say the game frequently generates scenarios that are unwinnable from the point of character generation? That sounds more than a bit cruel!

Yes, I think so, although I'm not sure how frequent it is. As much depends on luck as on your play skill. There's not much I can do about it.

On the other hand, I'm pretty sure you can't win Mist without being skilled. Mist is not designed for Borgs.

EDIT. Hell, I really don't know how common it is. It's a hard question. Sometimes the game generates levels where escaping alive is pretty hard. If you succeed in Mapping skill check, you can always escape back to town. Otherwise you need to be somewhat lucky to escape alive from tough situations.

[LostTemplar]

Scumming is not a problem for me, personally I prefer to fight strongest monsters, my character can handle, it is faster, and gives much less annoying feeling after character loss.
I just say, that many common 'anti scum' solutions actually make scumming more appealing, or more unnatural ways of scumming are being used.

E.g. farming monsters, summoned by enemy is more natural, and better for gameplay, then farming monsters, summoned by self.

It is also hard to make a formal definition of 'bad scumming habit' for me.
Character needs to kill e.g. 1000 monsters to become strong enough to kill big boss, is this fact by itself OK for you ?
If it is not then game should be more like DOOM then like angband (no level ups, etc.).
If it is OK, then what is wrong, if character kills all thees monsters on one dungeon level, or few levels ?
Also you say, you want every level to be cleared, but why ?
And if so, why not make stairs only appear, then last monster is killed ?

I think in Sil you should dive as fast you think you can handle

and then scum dlvl 18-19. (Why is this behavior considered good, while scum dlvl 1-2 an then dive is bad ?)

[Mikko Lehtinen]

Scumming is not a problem for me, personally I prefer to fight strongest monsters, my character can handle, it is faster, and gives much less annoying feeling after character loss.

Me too. Mist is designed for this kind of play.

I'm not forcing anyone to clear complete levels. It's just the most effective strategy in Mist because the content is limited, and I'm being honest about it. Similarly in most CRPGs you will usually want to take all side missions before moving on in the main quest.

Keep in mind that dungeon levels in Mist are quite small, and gameplay is faster than in Angband. Clearing a level doesn't take too much time.

After you have cleared a level and perhaps visited the town, you have the freedom dive either one or two levels. If you always dive as fast as possible, you will probably find yourself in dangerous situations all the time.

I always dive too fast and usually die around DL 22.

In Halls of Mist you get a big score multiplier if you dive to the level 48 as fast as you can (clearing 24 levels) and kill the big boss right away. If you keep scumming level 47 again and again, you will get less points. This is all just theory, BTW, as no one has killed The Thin White Duke yet.

[Mikko Lehtinen]

and then scum dlvl 18-19. (Why is this behavior considered good, while scum dlvl 1-2 an then dive is bad ?)

Good question. Do you get bonus points in Sil if you win the game without maximizing Minimum Depth?

[Scatha]

The scoring in Sil rewards winning in as quick a time as possible, but the time limit means that winning at all (or making it to any particular depth) is a real achievement in itself, and gives a canonical challenge level for people to aim for. I feel that part of the game designer's job is to set the difficulty for those who want that. Letting players choose their own difficulty is fine for some, but it becomes harder to feel satisfied at beating the game.

Given that there is a time limit, turns are a resource for the player to spend. Different characters may want to spend them in different ways. To maximise chance of winning, or particularly to maximise chance of spectacular win, diving at a fast but safe speed to get more turns at the deepest floors with the best stuff can be valuable. But there are enough dangerous foes at those depths that a good number of characters seem to die while trying to up their power level (and I suspect many of them could have escaped with a Silmaril if they'd gone straight down). On the other hand, some builds might need a bit to get going, and want to spend more turns right at the start of game. Characters with a high initial investment in Smithing can certainly qualify here.

Making monsters gradually get tougher would be quite like the forced descent in Sil but without the option to dive. I'm happy that reducing options isn't always bad, but I'm not sure why that should be the case here. I think the biggest point in its favour is that the player wouldn't need to understand the mechanic to work out how best to play, as that element would be out of their control.

I just say, that many common 'anti scum' solutions actually make scumming more appealing, or more unnatural ways of scumming are being used.

OK, but that's an effect of players now seeing the game as something to try to win, rather than just something to explore. And it can be fine if the "scumming" they turn to is not boring (and not otherwise problematic). And if it is, you can find ways to fix that, too. This isn't a bottomless procedure.

Yes, I think so, although I'm not sure how frequent it is. As much depends on luck as on your play skill. There's not much I can do about it.

On the other hand, I'm pretty sure you can't win Mist without being skilled. Mist is not designed for Borgs.

EDIT. Hell, I really don't know how common it is. It's a hard question. Sometimes the game generates levels where escaping alive is pretty hard. If you succeed in Mapping skill check, you can always escape back to town. Otherwise you need to be somewhat lucky to escape alive from tough situations.

It seems like it should be undesirable for this to happen more than occasionally (although randomly getting something which is hard to escape is fine). And you should be able to do something about it if it does occur too frequently, by working out what is causing the overly lethal situations. This might be a good thing to leave until after release, though, as you're likely to get more data.

I guess this can be awkward to identify, too, since as a player you can't always distinguish a death you couldn't do anything about and a death you could have avoided if you'd played a bit better.

[ekolis]

If you succeed in Mapping skill check, you can always escape back to town.

Oh, there's a town? So it's not completely ironman? Interesting...