Microgue (Updated 9/7)

[Paul Jeffries]

Nice title screen!

At the moment it feels slightly unbalanced, since everything of interest is in the top half of the screen, but I'm guessing that you'll be filling the dark space with a menu of some kind which will probably counteract that.

[jasonpickering]

yeah there will be a three big buttons on the bottom, Treasure, Options, and Adventure.

working on the after win now currently there is a treasure chest. I was thinking of doing artifacts for each level, but that would require me to draw like 20 different artifacts, but it would be cool because I could show how many heroes died to get this treasure. SO the player will know that they died 20 times between this treasure and this treasure

So that got me thinking about maybe doing just treasure. Open up the treasure chest and BLAM!!! money. So when the player clicks treasure it will just show them a gold pile that will get increasingly larger as they go. They wont know how many heroes died on eich unlock tier though. I could also make the money be an actual value. Like as a player goes through maybe the faster they do the more money they will get, but I think that is going more towards score and I dont know if fastest way through is the best way to do score.

I was also given the idea of instead of treasures doing "evil lords" their portraits would be easier to draw. and the player kills a new one each time, which explains them strengthening their defenses.

[jasonpickering]

Hey guys. I was just wondering. How important do you think it is to know the actual turn order. Or at least be able to have a good idea of the turn order in Microgue. I need some input.

[Erik Temple]

I think it would be nice if the turn order were fixed, with each monster type moving in a set order (for example, if the eyes always swapped first or last). But I don't think that the order needs to be explicit or documented--let it be a bit of knowledge that experienced players can exploit.

[jasonpickering]

Was that really obvious? Because I kept trying to think of an idea on logic sorting and my wife suggested the same thing I don't know how I missed it. I am going to implement that today. Also if anyone would like to test the game out I have a brand new build with a lot more stuff. I just need the ending system. On the fence between treasures or boss battles to end the game.

[Erik Temple]

By the way, I'm very happy to see that you're picking up Microgue again. To reply to some of the other concerns you've raised in this thread (I have been playing the build(s) posted at https://dl.dropboxusercontent.com/u/21825227/Prototypes/Microgue_Newest.swf for the last few days):

Difficulty
Obviously, perceived difficulty varies for different people. My perception is that the game is too easy for too long at the beginning. To get to a point where the game is interesting, I have to play at a pretty low level for a long while. Even up to the dragon stage, where I am now (haven't reset my data for new builds, since it isn't obvious when a new build appears), things don't really get interesting until about level 4 (of 7). It would be nice to have a different way of handling difficulty so that players who are doing well don't have to play so many rounds that don't challenge them.

Gauging difficulty
It seems as though the main heuristic you are using for difficulty is the single enemy taken in isolation, with perhaps the number of enemies being a secondary factor. You might consider whether combinations of enemies might be a more useful heuristic; i.e., are some combinations of enemies more difficult than others? There is also an issue regarding rooms: some enemies are much less capable in a room where there are passages of only 1 tile in width. I'm thinking especially of skeletons and demons, which can't even do their diagonal move in such a situation.

Unfair difficulty
I have played many, many games over the last few days, and only encountered one situation where the room setup was such that I could not have survived. So I think this is not something I would worry too much about, at least up to the level of difficulty that I have reached.

However, I do think that there is a problem with unfairness that comes in with all of the timed effects--the demon moving differently every other turn, the Yeti power, the two types of floor traps. These require memorization, which isn't a skill that I'm particularly interested in exercising when I play a game, and currently a heft percentage of my deaths are coming from not having memorized the state of one of the timed things. Please consider having some kind of indication on each of these timed effects that the trap will trip, the freeze will run out, or the direction the demon will go after the player's next move: maybe the demon's eyes flash red before he goes diagonally, the tips of the spikes show in the spike traps, the color palette for frozen enemies shifts to a lighter blue, etc.

Should the game have an ending?
My first impulse is to answer that it should not. There is only one factor that makes me say otherwise, and that is the fact that the upper difficulty level would be defined by simply unfair setups. That is, if monster numbers and difficulty are increased algorithmically, you will eventually reach a point where most or even all setups are impossible to win. That's a recipe for frustration, and a clear endpoint that comes just before impossibility sets in probably makes sense. The problem is that, if I win and then play again, I really don't want to start at the boring lowest difficulty.

Specific enemies
- I miss enemies that move two tiles. Are they coming back?
- What about slimes? Miss them too! If these two types of enemies don't appear until later (beyond the dragon), I think they should probably appear earlier.
- I haven't faced the dragon many times, but he seems pretty wimpy. Why does he mostly run away from me?
- The Yeti freeze lasts too long. Maybe decrease the duration by 1 turn? Currently, the level size is such that, by the time you have killed a Yeti, you have basically beat the room because you can just waltz right out while everybody is frozen.
- Maybe some enemies should be immune to the Yeti freeze? The dragon immediately comes to mind, since he is--I think--invulnerable?
- Floating eyes logically shouldn't be killed by the opening pit traps.

[jasonpickering]

So its been a long time since I updated that original build. (Testing in Stage3D which doesn't work as a regular swf and needs a HTML wrapper.) I just updated it to the newest build. Take a look at it because some of the changes you mentioned are already in. but I will go through your comments.

Difficulty: the comment of "to easy" has been brought up a bit ( I am at the point where I dont know if its easy becasue its easy or because I made it). I have done a full mix up of the enemy unlocks with some of the beta testers so the difficulty seems a bit better now. I suggest clearing your data by hitting 1 on the keyboard and starting over to see the new unlock order. what do you think could be done to make 1-7 all interesting?

Gauging difficulty: I don't really know how I would increase the combination of enemies and use that as a difficulty setting. I mean I understand that some enemies work well together sometimes, but I don't know how I would translate that into a difficulty curve as it seems very dependent on level and placement. I think its a really good idea, I just don't know how to go about that whole thingy, open to suggestions. I do need to redo some of the levels to give them a more open layout.

Unfair difficulty: I always thought of the game being about memorizing, but it is more about interrupting the rules laid out by all the enemies. the idea of a small note for the player could be a good idea. Similar to how st33d did Turnament. but It would need to be something that the player can notice, but isn't like a large neon sign. I just added in the small tip on the spikes to show they will strike next.

Should the game have an ending? I am thinking about some type of ending, but then perhaps unlocking some sort of other mode. maybe just a really hard like 20 level dungeon. I also think that you can do something similar to people playing Spelunky, where players will play to get their "Win percentage" higher.

Specific enemies:
- the arctic wolf is currently in the game. It takes two turns for each players one, although it is fairly broken and needs a full code rewrite
- Slimes are added in now. they should appear after about 7 wins. I am also going to add a Spider. which will put down webs, that are basically like unmoving slimes. limiting the player if they walk into them. They will last for about 3 turns.
- the dragon may just not have any other moves, and that's why he is moving away.
- yeti freeze has been dropped from 3 turns to 2
- that is extremely easy to do. I didn't want the dragon to be all powerful, but I can add it in to a build for later.
- the opening pit traps are going to be replaced by the moving walls, the trap doors looked cool, but they raised many problems like why eyes couldn't see through them, which was one of their main designs I liked. the main problem on the moving walls, is some good art. that reads well. also some people asked why the raising walls killed the enemies so I will need a "Idea" to show that.
- Also not sure if this was mentioned yet, but clicking and holding on an enemy will bring up a description.

[Erik Temple]

Thanks for cluing me in to the update. I see you have a note asking about whether there is a way to use Tumblr to host .swf files. Is there somewhere else that the latest version is always posted?

I've now reset the difficulty and played to the point where there are six levels in the dungeon. I still think it's too easy up to this point, but a game that started with the sort of difficulty that you see in the last three or so rooms of the six-floor dungeon might be about right. What about having a tutorial dungeon of 5 floors that's about the same difficulty as your current first stage dungeon? After beating that, new players should be ready for something a bit tougher. Players who are used to roguelikes & tactical movement games could pretty easily skip the tutorial and get right in w/something more challenging. (All this assumes you're looking for the widest player base possible.) And another mode that's unlocked by getting to the end of the staged dungeons sounds like a good idea.

I think the warning for the spike traps works really well. Especially for a mobile game--which you often have to put aside for 30 seconds to do something else--memorization seems like a cheap way to increase difficulty, and certainly Microgue can be difficult enough without it. Hope you roll out such subtle cues for all delayed timers, from moving walls to giant slugs.

As far as treating difficulty as the product of combinations of enemies (and possibly rooms), the best way to do that is probably by running simulations and capturing statistics on the win-loss ratio for different combinations. The only real reason to do that in a hobbyist project, though, is if it's something that you enjoy doing (never done that sort of thing myself)!

Some thoughts on enemies:
- One on one, giant slugs currently are like easier rats (or they will be if you give some kind of subtle visual cue the turn before they move). But that isn't bad, because as part of a mix of other enemies, slugs change the rhythm and give the player new patterns to deal with--whether to work around or exploit. One way to make slugs more difficult might be to have them leave behind slime that slows the player down (maybe up to 2 tiles of slime could be onscreen at a time per slug)--wait, I guess you're already planning that with the spider!
- I haven't encountered the arctic wolf yet, but it sounds like his moving twice for each move of the player means that he can actually follow you around a corner in a single turn (e.g., move west and then south in a single turn). For the sake of the sort of rhythmic patterns I mentioned for the giant slug, it might also be good to have something like the old orc/goblin which could move up to two tiles in a single direction.

Am just trying Turnament for the first time...

EDIT: I wanted to say that the change to allowing enemies to attack you on the exit stairs seems like a good one. It makes the last 1/3 of many floors matter more than it did. (Yeti still makes it easier to end the level, but it's definitely improved with a freeze of 2 turns. Maybe the Yeti should be a little more difficult over all?)

[jasonpickering]

Its a separate link but its using Stage3D, and the graphics are extremely broken.

so as you go up the tower the floors have different A. Monster Counts, and B. Monster rarities. So I just need to take the 4th floor and set that as the starting difficulty. That should be easy to do. I am also thinking of doing a tutorial dungeon to just teach the basic mechanics.

the stepping away from a mobile game is an excellent point, that i had not even taken into account. I am going to have the snails eyes closed when it wont move, but i am trying to figure out some thing for the demons. thought about changing the eye color, but not sure if that would work.

as for the treating difficulty, that all seems beyond my feeble scope. and I automatically thought of a dude in like 50s era nasa suit getting a a stack of read outs from a computer bank.

-so I always wanted to make the snails stronger. they are extremely slow, what if because of their shells I make them unkillable by moving into them, same as a dragon. that might increase their difficulty a bit.
- I am also thinking that the yeti needs to be tougher, especially because they are so big, I could have them do the original goblin thing of moving two spaces in a direction, but I wonder if that might be better for a more abundant monster.
- the wolves can turn corners. (although they are really broken at the moment, Need to completely redo their AI)
- I am also thinking of possibly doing "Special floors" a couple examples are:
     1. Dragon Floor. I would really beef up the dragon. It would stay unkillable, and would probably move in all 8 directions. but basically the dragon would be asleep. each turn a countdown would go off above his head, so the player would get like 8 turns to get to the end before the dragon awakes.
     2. Locked room. The stairs would be hidden with a large skull symbol where they should be. the skull would have a number on it and each time a monster is killed that number would tick down. meaning the player would need to kill like 2/3rds of the floor to move on.

[Erik Temple]

I like the idea of special rooms, especially if they are primarily used as treasure chambers (i.e., the final room where the artifact is located).

I've now played to the stage that introduces liches and ninjas, the progression is definitely better than in that earlier build I was playing. Still could use some tweaks though! Some more thoughts on enemies and difficulties:

- Witches bring more interest than they do difficulty, since the randomness of their ability can as help you as often as it hurts. I'm not sure that they deserve their own stage, because the difficulty level kind of stays static or even backs off a little. I think you could stick witches in earlier, in low numbers, w/o making them the primary "star" of their stage.
- Slimes don't necessarily bring much difficulty either at present. Maybe it would be best to introduce them in a stage where there's also an enemy that moves quickly, so that the player's mobility is more at risk? Or else introduce them in the stage immediately previous to such an enemy, so that the player has a chance to see what they do before things get rough? (They needn't be the "star" of a stage either. (Ninjas seem more deserving of that--they currently are not the star of their stage.)
- Liches don't summon skeletons often enough, I don't think (needs a graphic cue the turn before summoning). They should also move around a bit more, maybe? Also, they could look a bit more horrifying. Maybe if the face was lost in blackness?
- I agree that a two-step monster might be better for something less rare than the yeti. A couple of ideas for the yeti: Maybe it can move in all 8 directions. Or leave a trail of ice behind it (max of 2 or 3 tiles before melting). Anyone stepping on one of these tiles slides to the end of the ice, a la Relic Rush. (Or that ice thing could be a trap, too.)
- I think an invulnerable snail could be good, probably with minor tweaks to initial appearance and rarity.
- How about a patroller style enemy that always goes back and forth along the same file? If need be, these could be set up like traps (or just be traps). You could have two variants: one that goes step by step, the other that moves the full distance back and forth every turn, killing the player if he steps in the way at the wrong time.
- How about an enemy that can't be killed by jumping on it but *can* be knocked backward one space (the player doesn't move when knocking)? The trick would be to knock it into a trap or moving wall to kill it; otherwise it would be invulnerable. Maybe that's too different from everything else?
- Giant toads (or however you want to reskin) might be fun. The toad would hop two tiles where possible, and the player would need to try to be sure that he would be on the tile hopped over. In other words, the toad would always kill from two tiles away unless the player is directly against a wall. If the toad is directly next to the player, he will just hop to the other side of the player if not killed immediately.
- A super-evil enemy might be one that moves all other enemies one step toward the player. (Dungeonmaster or something).

Not all of those ideas for new enemies are necessarily going to be useful, and of course I still haven't seen all of the content in the current version of Microgue. But I thought it'd be better to get them out there than forget about them...

And a couple of notes for the monster info that you get on touch and hold:

- The wrong portraits are often displayed.
- Notes (or a graphic) about how the enemy moves would be great, because the first time you encounter them you have no idea what's safe. For example, the ninja should communicate that he moves in a knight's move.
- Witches don't have info at all, they display as yeti or ninja.

[Erik Temple]

A couple more notes about gameplay:

The ninja is the closest thing I've seen yet to a magic bullet for the gameplay. Interest and challenge go up quite a bit on most floors where there is a ninja.

Two liches is also a fun, challenging combination. With two liches on screen, I'm not sure whether they should move around more or not. I also don't understand the rules they follow when summoning at all.

The first 3 floors or so are still less interesting than 4-7. That may say something about the formulae for determining difficulty. That is, maybe easier floors may have fewer but individually harder enemies (so they're not just freebies), while later floors have more enemies, with an increasingly difficult composition as you go along...?

I feel like there is one main gameplay problem that probably won't be solved with relatively simple balancing-out (It's possible that you won't even see this as a problem.) It doesn't affect every floor in my playthroughs, but it does affect most: the rightmost third of the level is usually not nearly as interesting as the lefthand third to half. This is because most of the enemies follow you toward the left, and all the maneuvering you do early on allow you to make a break for it later, leaving behind the enemies that started on the right. The idea about patrollers in the previous post was intended to help with that. Another idea (something of a band-aid) would be that some monsters on a floor might be tasked with defending the exit, and refuse to stray more 2 or 3 tiles from it.

[jasonpickering]

No update as it's pretty broken at the moment.

Working on making the game a bit grader. Trying to make level one feel like level three.

I added a new starter enemy. It's a small flame guy called a cynder. They move like the rats, but with two differences. They are immune to the yetis freeze attack and if a cynder is next to another cynder they will combine into a new monster called an infernus. The infernus will move like the old goblins did. Going two spots at a time in the cardinal directions. Later levels will also have infernus on them. I may make it possible for infernus to become a bigger enemy not sure yet. If I do they will basically be like rooks.

Yetis now move in all 8 directions. It's much better.

Laying the groundwork for special rooms. I might also do rooms with just themes. Like a bunch of yetis, or fire guys. I also want to make a demon room with sigils just popping out demons.

I am working on the right side emptiness problem. Trying to think of some different solutions.

Not sure about adding multiple enemies at once. It shouldn't be to different if the guy is just like hey now I got witches and slimes though.

I might also add some enemies that cause problems on death. Like an enemy that shoots out fire around it. Or explodes so you have to be careful when it's on traps. I am also thinking about adding a trap in that shoots fire in cardinal directions and it rotates. ( some enemies will be fireproof, immune to fire traps and freezing)

[Erik Temple]

All sounds good, like the sound of the infernus!

Having played quite a bit more, I think that the levels 1-3 actually do sometimes produce interesting play, and other times no. The main issue seems to be the monster makeup of the floor. Even at the 7-level stage, it's sometimes possible to have those levels be mostly populated by rats and snails. In that case, it's not very exciting. (Though, with the traps, those might make good levels for lower difficulty stages.) Floor 1 of the 8-level stage can be pretty tough!

[jasonpickering]

yeah, Monster generation is super basic and pretty random. Let me explain how I do it.

Monster Generation.

As the game goes on the list of available monsters grows. So at the beginning the monster looks like this:

["Rat", "Skeleton" Eye"]

so then it looks at each one and then checks the rarity value for each floor. so for the rat the 8 floors look like this:

[ 3, 3, 3, 2, 2, 1, 1, 0]

so the game takes the current floor and then sees how many are listed for the rat. So for the Rat the first floor get 3 into the monster bag. so by the time its done the monster bag looks like this.

["Rat", "Rat", "Rat", "Skeleton", "Skeleton", "Eye"]

then the game just randomly pulls 4 monsters out of the bag.

Its not the most elegant solution but its effective. The one problem is that it can be pretty unpredictable. I am open to suggestions on what might be a better way to do this as I am going to be adding the special floors which would be better weighted to certain enemies.

Also some people have asked for an endless mode, which my current system does not support because of the floor rarity.

[jasonpickering]

So I have been reading through a bunch of Tabletop RPG books about their Random Encounter tables (Their logic works well for me). they have a good idea of basically having a table from 1 to 100 of monsters so 1 - 20 might be rats, 20 - 30 might be skeletons. and then ifwant tougher encounters you jsut add a modifier to what you roll. so I might be able to do something like that.  as I fill my monster bag its filled in order of easy to difficult. I can then just add modifiers to push the selection towards the end of the array as opposed to the beginning. It will take some testing, but might work.