mimics

[galehar]

I've almost finished reimplementing mimics. Here is the new design. Feature mimics are mostly finished, I'm working on item mimics now. So if you have comments about the new system now would be the time. Of course, it's a bit late to suggest a completely different design.

[dpeg]

Item mimics reduce loot. (Think of the guardian serpent vaults.) The idea that a stair you want to take chases you is considered funny by some. (This assumes that you cannot throw stones at things to detect their mimic-ness.) Mimics create a unique form of suspense/surprise.

[mumra]

Might it make mimics more interesting if they actually had a chance to turn back into the thing they were mimicking once they die?

Less useful for some types (e.g. door mimics, unless we get a door god) - but say for a stair mimic, you could lead it to where you wanted to conveniently place a staircase to use in your Orb Run. Items are obviously valuable (and the mimic would need to be correspondingly tough for some of the better ones).

[galehar]

They are functionally almost identical to trapdoor spiders. Anything wrong with it? Do you check every floor cell for a trapdoor spider? Does that even mean anything? I was considering going even closer by making them attack you if you step next to them instead of on them (other means of revealing like ranged attacks don't work anymore).
I think if they are rare enough, they can surprise you.

[njvack]

The features they imitate are important enough that it's important to check them, and uncommon enough that it's practical to do so.

According to the mantis wiki page, it's no longer possible to check them -- up to and including letting a monster go through a door mimic. I'm guessing monsters will be able to equip mimic gear, which would reveal when you tried to move on to it, too? Theoretically, that'd let you have more than one mimc on a square -- but if you only make one of them reveal at a time, it's not a problem.

I'm not big on mimics in general, but I'm pretty hopeful about these changes.

[galehar]

The features they imitate are important enough that it's important to check them, and uncommon enough that it's practical to do so.

According to the mantis wiki page, it's no longer possible to check them -- up to and including letting a monster go through a door mimic. I'm guessing monsters will be able to equip mimic gear, which would reveal when you tried to move on to it, too? Theoretically, that'd let you have more than one mimc on a square -- but if you only make one of them reveal at a time, it's not a problem.

I'm not big on mimics in general, but I'm pretty hopeful about these changes.

Monsters will be able to step on mimics but not pick them up.

[dpeg]

The idea that an item mimic could turn back into the representing item is very cute... after galehar's change they don't switch appearance anymore, right?

I guess the real reason for mimics is heritage: D&D, Nethack, Crawl. Of course, that's a pretty good reason to abandon them. On the other hand, players told us how they liked Crawl's item mimics (being _the_ item) and I personally like the idea of being chased by a portal vault entrance.

[galehar]

They are functionally almost identical to trapdoor spiders. Anything wrong with it?

Seems like a lot of wasted effort and code if a monster is almost identical to another monster.

Only functionally. I guess the flavour is a bit different.

Item mimics reduce loot.

I think it's negligible. I initially suggested a chance of 1/1000 for items. I think it may be too low, so I'll start testing with 1/500. Anyway still a negligible impact on total loot size.

[Galefury]

If they are created from actual features and items making them turn back on death could work nicely. I know I would feel cheated by portal and randart mimics otherwise (I already do if the mimic is well-disguised), and that's not very fun. If stats are based on the item/feature and depth this could also add an interesting challenge by making the player fight for something he wants.

About expiring portal mimics: They could turn into stone arch mimics.

[dpeg]

galehar: I was referring to vaults using mimics. It started when I made mock-up versions of the guardian naga (now serpent) vault, complete with naga and with either full or partial replacement of the items by mimics. It's a joke, but a rare one and players seemed to like it.

[bobross419]

I read through the wiki page, but I was unsure of one thing. Will these changes involve item mimics chasing player characters? Or will the item mimics remain stationary?

[dpeg]

They will follow the player, but be of slower speed (than humans etc.)

[KoboldLord]

One monster out of several hundred that is there for humor rather than any specific tactical or strategic merit is fine. Mimics are funny. If we wanted to be Stop Having Fun Guys I can think of several other candidates for removal, but in the end removing the funny monsters isn't going to leave us with a better game.

[Lantier]

I like mimics and enjoy having them as part of the game. They are particularly funny when your friend hits one (or better yet, several... in a row...) on WebTiles.

[galehar]

They are in trunk.

[galehar]

what's the problem of checking all the stairs and doors?

[galehar]

But what's the problem? Why do you need to do it?

[galehar]

Because stairs, hatches and doors are important. If monsters are chasing you, you will often want to go up or down stairs. If the stairs aren't actually stairs, then running to them is a very bad thing. However, usually the stairs you run to are already in explored territory. Which means that once you've cleared out any monsters near stairs, you should move onto the tile with the stairs, to ensure that they are actually stairs; this way, if you actually need to use the stairs later when running from monsters in some other part of the level, you know that they are stairs and that you can use them. The same goes for doors, of course, and to a much lesser extent for other less important features.

You do that for stairs anyway, since you probably also want to take them and check the other side.

A smaller problem: since mimics provide XP, you might want to move next to every statue/fountain/item to ensure you get that XP.

This is too minor to worry about. Item mimics has always been there and nobody has ever cared checking every single piece of junk in case it's a mimic.

[BlackSheep]

You do that for stairs anyway, since you probably also want to take them and check the other side.

Players may walk up to stairs and hatches, but they don't throw things at them. Having feature mimics makes for early edition D&D-esque trap searching, where you don't step anywhere unless you've rolled something across it first.

What about having the Traps & Doors skill give an increasing chance to notice that a mimic is present?

[Kate]

You do that for stairs anyway, since you probably also want to take them and check the other side.

Players may walk up to stairs and hatches, but they don't throw things at them. Having feature mimics makes for early edition D&D-esque trap searching, where you don't step anywhere unless you've rolled something across it first.

Crawl feature mimics don't fall for that.

[BlackSheep]

Then I have to register my disapproval of the concept. The variable timing thing is irritating enough, but being unable to rely upon staircases as real just isn't fun. All that will do is make me traverse every downstair on a level I've cleared before exploring the next level.

[ebarrett]

Who the hell checks stairs and doors for mimics anyway? If you suddenly need to run to a door or staircase and it turns out to be a mimic it means you were probably retreating into unexplored territory, which is a Thing You Shouldn't Do.

also obligatory boo optimal playing

edit: oh by the way galehar links in first post are kaput

[galehar]

Indeed, checking for mimics is something of a part of exploration now, except autoexplore doesn't handle it.

Well, it should be quite easy to change autoexplore to make it check mimicable features.

[greepish]

The point is that you check explored territory. Indeed, checking for mimics is something of a part of exploration now, except autoexplore doesn't handle it. Of course, most people don't bother with it, yeah.

I've gotta agree with minmay here, I can see shop mimics being just funny, but stair mimics really add too much negative to the game. You really would have to check every darned stair first to see if it's a mimic since you they're your greatest means of escape. It's not at all like a trapdoor spider, since it's not the hidden danger that's bad, it's the unexpected lack of escape. If you want stair mimics, keep it a Xom-Only thing or something.

[KoboldLord]

Isn't it already routine practice to check every up-staircase when you get to it? I don't like having stash runs cause twice as much piety decay as normal because my character's taking the long way around because I've only explored one set of stairs.

[greepish]

Routine for some, but hardly all. I tend not to unless I think I've missed a branch, since I tend to clear most levels. I honestly just don't care about the 1-2 piety loss :/

[rchandra]

not having played the new mimics in trunk, I can only add a comment on the old ones:
It would be very good for autoexplore/travel to avoid known mimics and/or to be able to set a 3x3 exclusion instead of needing to exclude each adjacent square.

[Bim]

I've just been playing trunk, and I've found SOOOO many mimics. I'm glad they're a bit more common and not as crazy strong/scaled more, but I've found far too many in my opinion, to the point it's getting annoying
Possibly they could get toned down a little bit?

Also, possibly door mimics could move through doors (and making them rarer/further down) would be more interesting?

[BlackSheep]

not having played the new mimics in trunk, I can only add a comment on the old ones:
It would be very good for autoexplore/travel to avoid known mimics and/or to be able to set a 3x3 exclusion instead of needing to exclude each adjacent square.

Galehar added an auto-exclude for known mimics late last week.

[galehar]

What version of trunk exactly. There has been a number of changes, not all at the same time:
stats scaling with depth (HD, damage, poisonous and constricting attacks starting at certain depth).
Auto-exclude known mimics (detected with Ash or antennae).
Removed depth restriction (a bit later, D:1 is restricted again).
Reduced the frequency of door and stairs mimics.
New mimic type (inept) for the early game. Inept mimics (up to D:11) get only one attack and no resist. Normal mimics (start at D:7) get poisonous and constriction attacks. HD and damage are still scaled by depth.

So in latest trunk, they shouldn't be more common than they used to be. Doors and stairs mimics should even be rarer. Only they can now generate from D:2. There's also a few new mimic vaults.

[njvack]

Removed depth restriction (a bit later, D:1 is restricted again).
...
So in latest trunk, they shouldn't be more common than they used to be.

Now that mimics spawn in the early game, you'll see a bunch more of them, won't you?

I haven't played the new mimics yet, but these sound like some welcome improvements -- a good way to get introduced to mimics in general. Thanks!

[galehar]

Now that mimics spawn in the early game, you'll see a bunch more of them, won't you?

If you usually die before D:7, yes

[Kate]

The fact that they spawn absolutely everywhere throughout the entire game makes them pretty repetitive and unsurprising, so I'd also agree that there are too many of them (and have been for a while).

[Blade]

It seems to me that the best way would be to have them be practically vault-exclusive. They can work very well in things like the guardian serpent vault, or places where they are very specifically set to be dangerous, but they're generally boring irritations in the dungeon. It seems like the ideal balance would be "often in vaults, <5 random ones in a typical 3-runer"

Maybe that's a little too drastic, but it seems like the direction things should go.

[galehar]

The fact that they spawn absolutely everywhere throughout the entire game makes them pretty repetitive and unsurprising, so I'd also agree that there are too many of them (and have been for a while).

Which ones? Item mimics are already rarer then feature ones. Although this is balanced by the fact that there are more items than features. Also, as I've just said, door and stair mimics frequency has been reduced by 4 (since 2 days ago, and this include hatches).
Should they be cut from certain branches (only vestibule, temple and slime currently)? Maybe from Pan since they aren't demonic.

It seems to me that the best way would be to have them be practically vault-exclusive.

Why? Vault mimics are extreme spoiler material. Unless the vault is well designed to be unrecognisable. But then the vault doesn't serve any purpose as random mimics do already a great job at being indistinguishable from normal items/features.

For reference, the current chance for a feature to be turned into a mimic is 1 in 100 (and now 1 in 400 for doors/stairs) and for items it is 1 in 500.

[XuaXua]

I am encountering a ton of mimics in trunk (uhh, pre-trunk now: build ends in 1142) in the early game. At least 3 before lair every game.

[galehar]

I am encountering a ton of mimics in trunk (uhh, pre-trunk now: build ends in 1142) in the early game. At least 3 before lair every game.

0.11-a0-1142-ge65fe5e is after the removal of depth restriction and before the reduction of stairs/door mimics.

Again, specifying the type of mimic which you find too common would be more helpful. Thanks.

[Kate]

Items seem like the most common to me, but I think what makes it feel so repetitive is more the fact that they just spawn all game (and still aren't very threatening the majority of the time). Having a Pan exit turn out to be a mimic that I could kill in a couple of turns wasn't much of a good surprise when I'd already killed off so many mimics in assorted loot vaults throughout the game.

[XuaXua]

I am encountering a ton of mimics in trunk (uhh, pre-trunk now: build ends in 1142) in the early game. At least 3 before lair every game.

0.11-a0-1142-ge65fe5e is after the removal of depth restriction and before the reduction of stairs/door mimics.

Again, specifying the type of mimic which you find too common would be more helpful. Thanks.

The kind that disguise themselves. I had no idea I needed to keep track of type. There are too many "in general" / "in total". It just doesn't feel like the neat surprise that it should; mimics should be as rare as, say lindwurms, otherwise they become humdrum.