Author Topic: Demon: A monster collection roguelike 3/31!  (Read 33171 times)

jim

  • High Priest
  • ****
  • Posts: 373
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #30 on: January 10, 2015, 09:54:21 PM »
You know, I was ready to give you the obligatory thumbs up for being a hardworking indie dev, trying your best etc., but what I was not prepared for was to be genuinely impressed by how much content already exists within the game, and the kind of insight in game design I've seen so far. Bravo.

...Was pissed that I couldn't get the headless zombie to kill the scientist in time :)

Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #31 on: January 11, 2015, 02:52:46 AM »
Awesome, and thank you. :D I'm glad you liked what you saw, but I definitely still have a fairly long way to go yet. :) Many (not quite all) of the basic systems are in, but usually only with just enough content to get the point across. :D I can't wait until I finish everything else up and go deep diving into content creation.

By the by, Headless does make a pretty good buddy. Maybe next game? :)

koiwai

  • Priest
  • **
  • Posts: 99
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #32 on: January 12, 2015, 12:19:09 PM »
I think, I could get pretty far. I've got to the 10th tower and doing alright.

However, it was not easy to start. I died several times somewhere around 2nd-3rd tower, basically in the beginning. Finding a good starting combination takes time, but eventually I chose a pretty good one: 2 Electricity skills + Wraith touch, and a Malingee as the first demon. At the first few levels, also picked Slime, Gandayah, and Faerie demons.

Both electricity spells are strong enough to kill most of the the enemies, but because they use quite a lot of Stamina (SP), it's convenient to have Gandayah summoned: it shares with you its stamina, so you basically get a lot of extra stamina to cast your spells.

I think, Slime and Malingee are pretty awesome melee demons for the early stages.  Although both have their downsides (Slime is weak to a lot of different damage types (e.g. Pierce damage), and Malingee has rather low HP). However, I used Melingee to train Pierce protection to the Slime, and used the Slime to train the Healing (Reshape) skill to myself and Malingee. So, their weaknesses were maybe not completely removed, but lessened at least.

Faerie and Gandayah were ok with their healing spells / buffs.

So, if somebody else has trouble in the early game, this combination was pretty successful for me and I would recommend it.


I'm really impressed by the depth of the game: Even though you are saying that there is not enough content yet, I'd say there is a lot of fun already! Well, there is probably more than enough for several runs. Dr West and the Headless were mentioned already. Usually, there are at least 1-2 unique monsters per level (those who are super tough, those who summon other monsters, a dude with resistances, etc).

Well, another great feature is the the metagame of choosing your best 8 demons: it is quite tough sometimes, particularly, when you want to take a new demon, but before you see their abilities, you have to discard one of the demons you currently have..  :-\  well, sometimes, it feels like losing your good friend ))

For me, levels 3-6 were the hardest. The first two levels seem to be intentionally simple, but then it becomes much harder, and it seems that the most fun uniques are also found somewhere there.

The game is very addictive, I played the whole night. 90% encounters make you think, and only occasionally 10% are simple nobrainers. Which is quite refreshing, most games get this ratio the other way around  ;D

Is the tileset your original? Some monsters, especially the later ones, are very nice! On the other hand, some of the early game monsters don't look that good, imho. Maybe the tileset can be improved a bit.

One last thing, about the map. The basic map topology is fine: 2-tiles wide corridors make a lot of sense. The dimensions of the rooms are fine. What I mean is that once you get to a fight, there is very little incentive to move around, or use the terrain in any way. Do you have any plans to make the map more elaborate? That is, some features of the map that can be used in combat, adding another tactical dimension to the game. This is not necesarily a good thing to have, but I'm just curious about what's your opinion? What are your plans regarding the map generation?
« Last Edit: January 12, 2015, 12:23:51 PM by koiwai »

Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #33 on: January 13, 2015, 01:09:37 AM »
Getting to Tower:10 is pretty awesome. I don't get many reports of people making it that high. :) Congrats. :D

The early game is a little rough, and it's somewhat by design. One of the thoughts running around in my head when I started on this was how much I didn't like the "long yawn to Lair" I felt like the start of DCSS had turned into for me. Barring very very early badness (like player ghosts on D:3 with Lightning Bolt, Ijyb w/ a Wand of something terrible, Sigmund w/ a Haste Potion, etc.), the first several floors of DCSS seemed trivial to me unless I was playing a challenge character of some kind. This bothered me because, in a permadeath game (particularly one with DCSS' strong variety in possible builds), you're gonna see that beginning often. If it's easy enough to become boring once you get the hang of it, that's a lot of boredom you're going to be asking people to repeat. So for that reason, I've tried to make even the early game a little nasty-tempered at times. I'll acknowledge there's a risk it could turn off new players who get impatient with it, but I'm hoping that like you they'll view it as a challenge and work to find a way through it. :D

This sort of sentiment is also what brings in the relatively high challenge level intended for the encounters that you noticed. There isn't much point to easy encounters other than giving the player a chance to feel and see how powerful they've become and to let the tension drain off a little so the hard ones still feel exciting. Those things are important, but you don't need many easy encounters to check that box so to speak: the game is weighted to only give a couple of easy fights per floor. The rest are "at level"... or worse!

I'm happy you were able to dig into some of the depth in the early monsters. :D The starter monsters all have a lot to offer the player and each other, depending on what you're going for. As you discovered, Gandayah can be *very* helpful to a caster build like your Electric/Dark one. Not only do they funnel Stamina to you, but the passive they have and eventually teach you raises your maximum SP. With the help of one monster, you go from having just your own 100 SP to having access to 250 SP (its 125 + your 125) This is one of the ways to get spammy with big magic even early on. :D

Re: Discarding demons. The UI/documentation isn't so good on this yet, but should be able to right-click any demon in a list like that to see its character sheet. This also works directly on the game map too. The build I'm working on now will be adding an "examine" hotkey button and hotkey so that this functionality is much more visible (right now it's buried in the manual to the point of probably being invisible. :( )

The tileset is all programmer art done by me, so I'm not surprised it is a little iffy in spots (I had never done any art at all before this.) It sounds like I'm at least getting better at it as I go though, if the higher floors tend to be better. It used to be worse though, lemme show you my favorite before and after. :D

First, the malingee of today: A serviceable, if odd-looking fellow... but then, what is a humanoid that has been partially turned to stone supposed to look like anyway? :)



Now, the malingee of long long ago... ugh it hurts me to even look at this again. :P



The current art for the haietlik is probably the only art that bothers me as much as this guy's original art did, but snakes are a pain to do right, so I haven't gotten to him yet. Were there any others that stood out as terrible? :)

Re: terrain/environment. I'm still thinking about this sort of stuff, but don't have a lot of ideas I'm 100% happy with yet. The closest I come to that are with cloud spells (spells that create persistent effects in an area) and "summoning" spells (the ability to summon temporary minions... for example a Dark spell might allow you to summon a horde of temporary skeletons.) Ideally, cloud spells would work better in halls (not much room to freely move around/escape them) and summon spells would work better in open areas (much more room for the extra bodies to surround targets): if it works out that way, those could give incentive to change the venue depending on which of those spells the player and/or the enemy has. Terrain-level stuff like water, 'difficult terrain' ala D&D, etc. is possible too.. just still on the back-burner mentally for now.

Thanks for the playing and for the in-depth feedback. :) I'm glad you're having fun. :D If you do run out of content soon, don't worry, I've promised everyone the build after this one is a content push. :D

koiwai

  • Priest
  • **
  • Posts: 99
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #34 on: January 14, 2015, 12:35:02 AM »
If you don't mind, I'll comment more on the art. First of all, I like your art work quite a lot. It's very imaginative, and to get even better, what is needed is just practice more, and you will be really great at it.

I especially liked your lizards and snakes, as you said they were hard to draw, but I think they turned out to be pretty good! I don't have screenshots with the snakes, but I have some other good examples:



My main suggestion is about drawing humanoids. There are a few tricks you can use to draw then better:

1. When drawing a human-like creatures (or any other famililar object), many people have some sort of check list: a torso, a hand, anoter one, a leg, etc.. So, instead of drawing how the whole creature would look like, you get a certain list of pieces to put in the tile. The result is often some sort of abstraction or a stick figure (or a fat figure, but still very abstracted). I read that teachers suggest to draw not "what you think you see" (a human, an apple, an orc, or any other abstract term), but the colors you see (a big green circle, with a smaller circle on top, with a grey rectangle across it..). Maybe this is why your lizards and other monsters (where your drawing is more free form rather than the humanoind check-list-type) turns out to be more effective.

2. Unnecessary one-pixel features will be most likely invisible to the player, unless they zoom in. If possible, don't draw them. This is actually related to the previous point about the check-list type of drawing. 1-pixel features add noize without conveying much information to the player. Simplify the drawing.

This general suggestion works almost always for any drawing. For pixel art, there is another thing I should point out. That's the granularity of pixels. It is a constraint and we should use it to our advantage. In general, the goal is usually to mitigate this granularity, so that the player does not see the pixel grid. There is a very good text about it from PixelJoint (see link [1] below).

I took the liberty to edit your tiles a bit to hopefully show what I mean:



3. Also, try use color to emphasize the shape of the obect. Exagerrate features if appropriate.




Links:

[1] Tutorial from PixelJoint (I especially like the chapter "Things to avoid", in particular, the "Jaggies" and "Banding" sections).

[2] Pixel Art Tutorial WIP - very nice and clear.
« Last Edit: January 14, 2015, 12:47:47 AM by koiwai »

Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #35 on: January 14, 2015, 06:25:11 AM »
I don't mind a bit. :D Art's definitely my weak point, even if it is getting better. :D

Yeah, I figured out how to do snakes some time after I did Haietlik: I'm completely happy with Ilomba and Dipsas (not sure if you'd have seen those yet on T:10 unless they decided to put in an early appearance.) You make a good point about something I'd wondered about: even I'd noticed I seemed to have an easier time drawing the monstrous creatures than the humans. And yes, I was, at least loosely, making a checklist in my head for humanoids rather than just drawing them like I did the other critters. That was a really good insight. :D I will try the color and shape based approach you mentioned as I work on my new dungeon art for this build, and next build on the new characters.

Definitely a little guilty of the one-pixel thing, especially for eyes. I feel like it sometimes works, if there's heavy contrast in the color (goblins' yellow eyes on brownish orange skin, for example), but lately when doing things I used to do with one-pixel, I often go a little further instead (see the 'new' malingee in the previous post, with the slight orange halo around its eyes.) I still do this in some cases I probably shouldn't though (I think the player avatar still has white single pixel eyes for all the good that does. :P )

The ogre and goblin you tweaked there are definitely improved. I feel like their faces have a bit more character to them (especially the goblin, which went from a largely blank stare to having a hint of actual malice or cunning in its look... never thought I'd say that sort of thing about a 32x32 character sprite's 10x10 or so head), and I also think the ogre's simplified and smoothed out clothing makes it easier to see its girth. :D

I'll take a look at the tutorials you linked. :) Art was originally something I was doing only because there was no other way to do the weird monsters I wanted to do, and I expected to be terrible at it and not enjoy it. Somewhat to my surprise, it's one of the parts I enjoy most now and I'm starting to get better at it. :D Hopefully with help like this, I'll improve even faster.

Thank you for taking the time to write all this up, I really appreciate this. :)

jim

  • High Priest
  • ****
  • Posts: 373
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #36 on: January 14, 2015, 11:05:05 PM »
 :)


Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #37 on: January 15, 2015, 02:32:51 AM »
Holy crap, a win! Well, a win of all the current content. :D And by someone I didn't know before I started working on the game! Big congrats! :D

Did you get a screenshot of the score by any chance? I'd love to add you to the high score thread in the forum. :D

Holy crap! I'm still dancing around over here, and I'm not even the guy who won. :P

There's probably a hundred things I'd love to ask you but I'm going to try and force myself to be reasonable and stick to just a few. :D

1) Was the game fun? If it had any particular low fun or high fun points, what were they?
2) How did you find the challenge level of the game throughout?
3) How many tries did it take you to win?
4) What was your strategy/approach for winning?
5) Finally, any other suggestions, thoughts, or comments? :D

Ugh... must... not...bury... winners... in questions. I said I'd stop at a few and I meant it!

Congrats again, especially since you didn't have access to the Ferret Tip Line most other players have been able to use. :)

jim

  • High Priest
  • ****
  • Posts: 373
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #38 on: January 15, 2015, 05:21:48 PM »
Well, first of all, thanks for making such a cool game. Of course I enjoyed it; otherwise I would have quit. No screen shot of the score, but I think it was around 66XX.

1) Yes, game = fun. As others have noted, the high points in the game tend to come when fights get large and chaotic. You have a fairly limited ability to control what your monsters do, and I think most players learn quickly that a lot of success versus death has to do with time/turn management. For me, the high points were when I was wedged between a couple different groups of monsters, with my team in the yellow, wondering whether I should cast regenerate on someone or unsummon them. Choosing between limited options to get the best odds are what makes a RL fun.

Also fun: matrices. IMO half the reason for the success of the SMT series is the demon fusion mechanic. People love the grind, and they love slot machines. It's fun to look through the various possible combinations, something that feeds into gamer geekery very well.

The few battles I had with enemy summoners were also surprisingly fun. I like that they'll unsummon and resummon throughout, and if you kill them before whittling down their menagerie, all of their monsters a) become available to potentially recruit, and b) can also potentially overwhelm you because they are no longer bound into groups of 3.

2) I like the challenge level in this game. This is one of those games where you pretty much need to be dominating all the time, much like the SMT series. If things go south, they tend to go south quickly, and once they go south, you may be screwed. The few times I found myself in serious trouble (lost that egyptian jackal unique and my main healer while trapped between two enemy groups, for instance), I realized that unless I turned things around very quickly, even if I won this fight, I'd probably be so depleted that I'd lose the next one.

3) 3 I think. This ain't my first rodeo! This build was Buff + Healing. Previous two attempts were Mind + Dark (<-about as fucking useful as the mentalist in ADOM :P) and Melee + Healing (<- probably a mistake since monsters display huge preference toward attacking the player most of the time, but hey, I had no idea!)

4) Well, there are a few RL staples. For instance, you can stairway dance ala Dungeon Crawl. There should probably be some penalty for doing so. I backed down to the previous level more than once. Controlling the terrain is a bit more difficult when your monsters charge and attack enemies as soon as they see said enemies, but managing the circumstances of a battle are basically what a roguelike is all about. I did this as much as possible, waiting an extra turn to cast the speed buff, for instance, so that a turn with that buff wasn't wasted closing the distance to melee.

When fighting, 90% of what I did was buff with the speed spell and heal with regenerate (eventually peppered War Cry into there as well, and took the recover breath ability that lets you get stamina back, even though I had no breath weapon.) My stats were pretty even except for Cunning - I invested no points into Cunning.

Occasionally I would try to direct monsters with kill orders, but most of the time, fights were in 2 tile corridors, and enemy monster groups are pretty good at the "retreat and recover" tactic. Ultimately, I ended up with two heavy hitters and a healer and I let them do their thing. Tried a couple times to organize a different strategy, but this was just plain powerful and suffered less from stamina attrition, which is a huge problem if you are relying on anything other than melee, especially with the prevalence of the Infection status debuff. I suppose that this is intended to make players desummon/resummon more, but that's two whole turns, dude!

Anyway, I let my allies do their thing while I buffed, healed and unsummoned when they were about to die, occasionally popping in to take a few cheap shots. When high level allies acquired a range of abilities, their actions seemed more and more difficult to predict, so I focused less on what they were doing and more on what I was doing; I was the only sane member of my party.

Speed is a very powerful buff. Once I had a couple guys who could fight, Speed + Regen + Flesheater + Counterattack made them very hard to take down. I hate to say it, but probably OP (though it could be easily undermined by monsters who are immune to physical damage.)

Last, but not least, you can attempt to recruit certain monsters to get a free ally with no penalties, make an enemy go neutral, etc., even if you have no intention of keeping them. This has a huge effect and is begging to be exploited.

5) Haha, I'm happy to be an armchair dev and tell you all the ways you should make your game like the game of my dreams, but just like you're hesitant to ask a jillion questions, I'm hesitant to offer a jillion suggestions or comments. But if I were to keep it brief, I'd say:

a) demon fusion should be more random and the monster hierarchy should be way more expansive (obviously the latter must be a WIP)
b) you should be able to direct your monsters a little better
c) i really enjoy that you've created a wide variety of recruitment tactics - delightful stuff
d) please make the game open world, with persistent towns, shops, procedurally generated quests, procedurally generated monster types, and please ensure that the mathematically sound principles you are using to determine the ebb and flow of the tides on the perfectly scaled globe take into account the fact that there are now two moons in the sky
e) seriously though, this is the kind of game that NEEDS to be expanded!

Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #39 on: January 16, 2015, 03:53:41 AM »
Thanks for writing up this feedback. :D This is a big help. :D BTW: 6600+ would put you in first place for this version so far. :D 2nd was 6100 or so.

I agree the large fights are probably when things get the most interesting. It's been tempting to try and stack the deck a bit more towards those, but I worry about that becoming overly taxing.

I've been pretty pleased with how the enemy summoners turned out. :) It's funny how just wanting them to behave like a human-controlled summoner (i.e.: not letting minions die) ended up leading to the other features you noticed (having to fight and being able to recruit survivors.)

Now wait a minute. :D What's wrong with Mind? :D Well, I don't expect everyone to like every build or anything. :D Mind's actually one of my favorites though: early access to Sleep makes a lot of things easier to deal with (a lot of chase captures become trivial when you can naptime what you're chasing. :D ) Ego Theft is iffy at first, but it shuts down other Mind-users, and once it can power something stronger than Hypnotic Gaze (like Madness or Alluring, or one of the area-effect versions), it gets pretty nice. It isn't flashy though.. it's probably one of the lowest damage starts. If you want to give Mind another chance at some point,I recommend Mind/Elec, Mind/Debuff, or Mind/Heal. :D

Speaking of Melee/Healing, I do have an odd issue with melee balance I'm thinking on at the moment though: players tend to think melee is weak/too risky for themselves, but tend to love it for their pets. A lot of the issue is definitely the fact that the player's death == game over. In theory, the AI doesn't aim for the player more than his allies (it's based largely on a combination of "how scary you are to me" and "how quickly I think I could kill you"), but right now, the player has a hard time getting on the good side of that second half of the equation. This will probably be addressable once I add Demon's equivalent of the "race" choice from DCSS: choosing your relic. The gauntlet characters wear now is basically the "human" option, but there will definitely be relics that favor melee in various ways and hopefully make it more viable. For the other side of it, a lot of people win with melee pets, but I'm not as sure that's a problem yet... keeping an eye on it though. :)

Stair-dancing probably needs more nerfing.. I've been nerfing it in steps, but I don't think it's where I want it yet. :) I've got a few ideas in store, one of the bigger ones being Demon's version of DCSS' timed portals. No timer, but if you leave a floor of the tower by the main tower stairways, any of these portals present close, even if you haven't seen them yet. That *might* be enough to discourage it... might even be too harsh? Dunno, but not quite ready to tackle that anyway so I've got some time to noodle. :D

The Speed buff spell was recently buffed to have more effect but being shorter duration. It is supposed to be pretty good, Haste usually is in any roguelike. :D But, I'll keep an eye on it.

Stamina + ranged/magic is something else I'm still wrestling with. On the one hand, being able to potentially get 4-5 free attacks on a melee is powerful and feels like it needs to have a decently high cost. On the other hand, the practical result is you maybe kill 1 or 2 melees than are out of stamina once the remaining ones get in your face... and some melees deny you the 4-5 hits anyway by charging into combat with you via Pounce or Bull Rush. I haven't taken a stand on this yet, but it's definitely on my mind.

Recruiting as a "diversion" tactic will definitely have a penalty at some point, some ideas I'm tossing around in my head:
* Distrust Spread: Failing a capture attempt horribly (ex: you start a chase capture and don't even follow it, something where it's obvious you just wanted it out of the way and never intended to capture it) may cause Distrust to spread throughout every character who saw it happen, not just your original target.
* Reputation: The negotiation mechanic already uses this in a very, very rough (and sadly nearly invisible) fashion... a high number of recentish failed captures might count against you, making future attempts you really want to succeed more difficult or even impossible.
* Bitter Enemy: Failed capture targets may attempt to escape the level and, if successful, show up later via ambush with nasty (and fully Distrustful) friends. In essence, some captures already have this sort of "dire consequences for failure" clause, for example, Chachapuma, which remain Heroic even after a failed attempt.
* Limits on Capture Attempts: Not as big a fan of this atm, but I've considered this too.

Re: the suggestions. :D

1) 100% agree. :D Expanding the monster list is definitely a high priority. I have to split my time between it and systems for now, but I plan to have a much, much larger roster.
2) I do want to expand the control options a *little* for monsters, but not too much. I definitely don't want it to turn into something where you feel like you have to micro them or need to micro them.
3) Thanks. :D I have more types planned too. :)
4 and 5) :D I understand completely! Don't worry, I don't plan to stop anytime remotely soon. Demon will be much, much bigger before I'm done with it. :)

Thanks again for playing and writing up the feedback. :) The current build is mostly responses to prior feedback (largely in UI areas) and dungeon art improvements, but next build will be a content push. So hopefully there won't be a super crazy long wait for a tower containing some new challenges for you to tackle if you're up for it. :) Cheers!

jim

  • High Priest
  • ****
  • Posts: 373
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #40 on: January 16, 2015, 09:21:25 PM »
How you like me now, Ferret? 2 for 2. :)

Do you have a forum or something? I would like to follow this game more closely... it's been the first RL to really draw me in for quite a while.


Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #41 on: January 17, 2015, 01:58:53 AM »
Alright you. :P Now I'm shifting from purely "awesome, someone loves Demon!" to a hopefully healthy mix of that and "cartoon villain fist shaking and vowing to come back stronger than ever" :P

I have a devblog and a forum. :D http://demon.ferretdev.org and http://forum.ferretdev.org/ respectively.

I'll confess, the forum is a bit.. lightly attended. Whoever said "if you build it, they will come" never made an internet forum. :P I do have about 3 other semi-regular users, but it's one of those things: not many people post there because not many people post there. Would love to have another person using it! Notably, it does have a high score thread, and you'd currently have #1 for this version. Post the screenshot in that thread and take the bragging rights! (The highest score ever was 14152 on the previous version, that was a 3rd cycle death.. you'd be #2 on overall high scores.)

BTW: In terms of content, the current build could be compared to the trip to (not to and through!) the Lair in DCSS. :D Fairly similar to DCSS, the Tower is intended to have 30 floors, plus a hefty amount of 'side' content you investigate as you climb up (you may have seen a teeeny bit of that side content already perhaps if the RNG has favored you, but what's there now is mostly just a "dungeon branches, do they even work?" sort of thing) This is definitely not intended to be as evil/deep as it gets. :D Be ready for future builds to get a bit meaner past the current "end game". :)

jim

  • High Priest
  • ****
  • Posts: 373
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #42 on: January 21, 2015, 08:05:52 PM »


Okay, my third cycle win with this guy. This time went Mind / Elec. Did not find Mind to be very useful except for putting certain enemies to sleep, which is how I recruited Victor. The fact that these Mind abilities interfered with the use of group buffs like War Cry cemented my opinion that Mind is not so great.

There was one point where my group was fighting about 12 other enemies (including the bikini babe unique) in a 2-tile hallway, and truth be told, I almost died. I'd initially tried the old go-to of turning a centaur into an allied hero. He was promptly torn to shreds and I was suddenly alone. Within a couple turns, I was majorly poisoned, stun-locked by the cavemen, and using every rune I had to try and put guys to sleep or charm them so I could get some breathing room to resummon. I think the only reason I lived was because of the Diehard trait.

But that entire scenario came about because I was being completely reckless - I voluntarily unsummoned my entire team in favor of one centaur who was surrounded on all sides. At this point - and I've sunk maybe 15 hours into this game off and on - I'm pretty confident that I could indefinitely extend this win cycle. So I do think that either the difficulty needs to be ramped up, or - and this is my preference - the randomness factor could be ramped up. You spoke about a temporary portal ala DCSS as a way of forcing the player forward. I think that's a great idea. Some kind of reason to take risks and push forward is essential to a good roguelike. Except in the case of TOME, apparently... still have no idea why people like that game. :)

Some other ideas to make the game more brutal: maybe the penalty for failing to recruit an enemy monster is that they pull something from the bag of misfortune, something akin to DCSS miscast effects. Maybe their buddies get stronger or they have a chance of summoning other monsters. Maybe there is even a chance of this happening regardless of whether you're attempting to recruit.

Maybe in addition to items on the ground, there are treasure chests or anomalies or whatever that might drop goodies and might also put you in a world of hurt.

This may be pushing it, but: in SMT, there is something like an initiative roll at the beginning of an encounter. Sometimes enemies are automatically hostile, sometimes they are asleep, sometimes they get the drop on you and automatically attack, and so forth. Maybe some mechanic to this effect? Enemies are "afraid of you" (accuracy down), "enraged by you" (offense up), "curious about you" (easier to recruit), "rushing you in a frenzy" (free pounce-type attack??) Stuff like that. I'd also like it if enemies sometimes initiated conversation, e.g. "You win! Let me live and take this heal stone!"

One thing in particular I've noticed is that auto-hit abilities are definitely the way to go. For some reason, these are often very inexpensive abilities to use, and in many cases, they also have at-will targeting available. Very, very strong. My vote would be: make them a little more powerful and significantly more expensive, or extend their cooldown (thinking of Punish and Agony et al.)

Last, but not least, I think that regenerate is overpowered. I like that Mend applies a status effect, but an auxiliary healer with the player casting regenerate makes for a really, really tanky team. Most of the time, a well-placed buff and a few regen spells, maybe switch out a weakened guy for a stronger one, is really all that it takes to win.

P.S. One thing that I love about SMT is that you really get to know your team. You appreciate their strengths and weaknesses, you know who does what, and you know what to do in a given scenario. I don't feel this yet from Demon. It feels like it moves too fast for that, and your demons are very disposable. In general, my strategy is "strong guys first, and if one is about to die, replace it with the next best guy." If that were my strategy with SMT: Strange Journey, I'd be all kinds of dead.

And, again, of course, thank you for hours of free entertainment :)

Ferret

  • Bishop
  • ***
  • Posts: 122
    • View Profile
    • Demon: A monster collection roguelike
    • Email
Re: Demon: A monster collection roguelike
« Reply #43 on: January 22, 2015, 02:46:44 AM »
Though I agree with you that this may be suggesting things are a bit too easy right now, congrats on Win #3 regardless. :) Easy or not, no one has managed that before as far as I know, even though there have been a few who have beaten the content non-consecutively several times. If posted in the Forum of Near Solitude, you'd have the #1 score of all time. :)

The Mind thing might be coming down to playstyle, at least a little: If you've fallen in love with the AE buffs, Mind's going to start on shaky ground, since it tends to ask you to choose between "cheap, high success, Presence-based" effects and "very expensive, moderate success" effects.

...but, I will say not helping matters is Mind is relatively under-represented in low level Demons at the moment, and that needs to change. I mean, Stone Soup is willing to throw Confuse at you on D:2, not sure why I soft-pedal and wait until Atua on T:7 or 8. :D This is something I hope to address in the build after this one, which will pretty much entirely be content. Fire, Ice, Elec, and Body (not counting physical attacks/passives that happen to inflict Body effects), and Dark suffer from this issue too: it takes a long time to find new, interesting abilities in these elements, often you only get access to basic stuff early on.

I definitely want to add some more ways to place the player in high risk / high reward situations. The suggestions you've made for that (DCSS portals, increased penalties recruitment going wrong, encounter modifiers, treasure/trap objects) are all strong ideas for this. :D

I've been looking at the auto-hit abilities, both from the auto-hit direction and the "maybe normal attacks miss too much" direction. The auto-hits are usually cheap because they have cooldowns + target restrictions... but I may be giving these restrictions too much of a 'refund' and/or undervaluing how nice free target and auto hit are. Most of these are the sort of thing it's hard to mathulus out a correct cost for, just takes iteration and feedback. :D Going to try out some tweaks in either the next build and/or the following content push build.

I did want to ask for more info on this one bit:


P.S. One thing that I love about SMT is that you really get to know your team. You appreciate their strengths and weaknesses, you know who does what, and you know what to do in a given scenario. I don't feel this yet from Demon. It feels like it moves too fast for that, and your demons are very disposable. In general, my strategy is "strong guys first, and if one is about to die, replace it with the next best guy." If that were my strategy with SMT: Strange Journey, I'd be all kinds of dead.


Can you give me any info on how the game feels too fast in this regard?


Thanks again for the awesome feedback. :D I'm pretty jazzed that you've had enough fun to play it 15 hours. It's early yet and I definitely still have a ton of work to do to get Demon where it should be, but it's encouraging to see folks can find a fair bit of fun in it even now. :D

Vanguard

  • Protector of the @
  • *****
  • Posts: 1112
    • View Profile
Re: Demon: A monster collection roguelike
« Reply #44 on: January 22, 2015, 01:29:47 PM »
What is autoattack damage based on?  Just strength?

I feel like every fight in this game is too isolated from everything else.  There's no penalty to winning with a lot of damage, and no particular reward for winning efficiently.  I can avoid any long term loss by sending my demons away before they die, and there's no time limit to stop me from healing to 100% in between encounters.

My favorite thing are all the different weird requirements to befriend the demons.  Racing against a dog monster and feeding an unwanted minion to a reptile monster so they join your cause are both much cooler than a charisma check.