Randomly chosen spells in Auro is messy

Discussion in 'Auro: A Monster-Bumping Adventure' started by SwiftSpear, Jan 19, 2017.

  1. SwiftSpear

    SwiftSpear Active Member

    Some things responding to the state of being in version 1.3:

    The ELO ranking system implies I should be playing as close to ideally as I possibly can. I'm bad with some of the spells. If I'm exclusively playing ranked games I'd like to just not play with the spells I'm bad at.

    Spells synergize as well, getting two of the same spell is almost always a nutshot. Blast is particularly good with jump and dash. Jump, dash, gale is pretty comparatively. Rotissery, snowball, and floe are all weaker if you don't also have a movement based spell, but the opposite is true if all you can do is move around well and not bump monsters better. Firebomb is nearly an ultimate it's so good.

    The punishment for breaking a win streak is so onerous in Auro, those rounds where I look down at my spell setup and see something I don't like are heartbreaking, and I the player have not made any mistake.

    I was being hyperbolic with my statement that "that's way better!!!" with being able to choose spells. But randomly generated spells definitely conflicts with single player ranked play. I'm on the fence as to the actual best way to fix the problem. Has a lot of this changed in 2.0?
    richy likes this.
  2. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    You're worse with some spells than others, so we should let you choose which spells you get. You're also worse against some monsters than others. So you should be able to choose which monsters you're up against, too. And also certain level geometry you find challenging, so we should maybe allow you to choose what kinds of shapes are allowed.

    The point is that to succeed as an Auro player, you have to succeed with all of the monster combinations, all of the geometry combinations, all of the spell combinations. To allow the player to arbitrarily "cut out" half of this game (which is already up against the low end in content) so that they can play better is a bad idea to put it nicely.

    There is still a version of the game where you can get 2 of the same spell?? I thought that was gone as of like 2013 or something.

    Assuming it's true that some spell combinations are significantly worse than others - I would say this is mostly false, but assuming it's true - the answer is not "throw out the entire concept of random spells". The answer is "don't have spell combinations that are worse than others"... which I think we basically achieved in the latest version of the game (at least roughly).

    Dude, just PM me and I'll send you a key for 2.0. This is the second time tonight I had to defend 2+ years-old versions of the game to someone. :)
    Last edited: Jan 19, 2017
    Toad_Racer and vivafringe like this.
  3. richy

    richy Well-Known Member

    Re. choosing spells (I paraphrase):
    Hmmm, this kind of argument was made yesterday in a different thread. I don't think it works to support an opinion about X (some disputed thing) using the analogy X is like Y and Z (things where your opinion is more clearly right) and then hope agreement about Y and Z will settle the dispute about X. At least not unless the similarity of X to Y and Z is really obviously valid within the context of the dispute. It's also avoiding a potentially useful discussion about X.

    Spells are clearly different to monsters and map layouts. Sure they're all game features but the latter two are aspects of the game world, which the player is battling against. The spells are your powers in opposition to the game world.

    I think generally players feel that their character is "them" in a game world, and it's natural people will want to feel more control over themselves and their abilities than they have over the stuff the world throws at them. That's how it is in real life after all. I'm not saying total control, just some control that works within the design. (I'm not advocating letting the player design their *whole own game* like I sometimes get accused of! Just because I believe the designer shouldn't aspire to 100% control of the experience.) Anyway it seems like a useful area to explore.

    I think theres a definitely a sense in which players feel that the various aspects of the "world", i.e. terrain, monsters, monster movement patterns etc. are what "the game", i.e. the designer, is throwing at them to overcome, and that their character is "their responsibility", to make it so that it can overcome the obstacles. Obviously as designers we want it so that most of that happens via conscientious improvement of skill at the basic game systems so that the player just plays better. But to the player if that's all it is, they will feel to a degree that they're just "following somebody else's script". As in real life there can be something uninspiring about that. People need to feel they have free will, like they're writing their own script. Even just the illusion that's happening is better than nothing.

    Ideas around choosing player character abilities feed into that, and ideas around changing aspects of the rest of the external game world (monsters, map terrain) don't. That is why I don't accept the analogy you made. And why I think the OP is a valid point and it's worth looking for more design variations in addition to "keep it" or "throw it out".
  4. vivafringe

    vivafringe Well-Known Member Staff Member

    It’s a valid point that player expectations matter and should be accommodated whenever it is cheap to do so. In the past I’ve complained at Keith for calling health “barrier”, when health is the standard term according to player expectations, and he agreed he’d just call it health if he were to redo the game.

    In the case of spells, though, I think it’s just way too expensive to accommodate player expectations here. We had pickable spells for a long time, and the result was not pretty. It just guts your content too much, since players will just naturally pick whatever maxes their winrate, even if the difference in power is like 1% of winrate.

    Balance in Auro 2.0 is pretty crazy good. I think if you looked at winrate for all possible spell combos for a player that had stabilized, the lowest winrate would be maybe 40%, and the highest 60%. That’s really amazing given how few knobs there are to turn. But even then, it’d be silly for players to pick the bottom half, or even the bottom 90% of spell combinations.

    In fact, it would probably be preferable for Auro to just have 4 spells rather than letting a player pick 4 out of 9. In practice, the game will be pretty much the same, but at least Keith could spend his time completely on those 4 spells rather than trying to design a bunch of useless content.
  5. SwiftSpear

    SwiftSpear Active Member

    Or selectable character classes would be another way of isolating spell layouts without pushing an intimidating amount of choice on the player at the beginning of every round or throwing out the spells that already exist.
  6. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    Selectable spells is like a crucial lesson in why we have game designers in the first place. Players think that they would like to be able to choose their spells. "I like THESE spells, not those spells", they say. But the thing is that players actually might, despite their belief about what they like, be incorrect about that. They may actually end up having less fun by just using the same spells over and over again, than if the spells were randomly given to them. That is my prediction, and those kinds of predictions are exactly what game designers are for --- hopefully seeing beyond what a player says they want.

    If we aren't going beyond what players say they want, we do not have any need for game designers. We can just have programmers implement what players say they want. (Sadly, this is a lot of what the industry is right now, and a lot of so-called "game designers" are actually more like producers or project managers.)
  7. SwiftSpear

    SwiftSpear Active Member

    "It was the right choice because the game designer made the choice and the game designer should know better than the players" is circular logic. The game designer has executive decision authority for sure, but that doesn't mean they're infallible simply because of their status.

    This thread is supposed to be a discussion about the state of things, I don't intend it in the slightest as a shot at your competence. You don't need to justify your rightness to make the decision in the first place. The old "choose your spell" system I'm sure had more problems than the current system, but that doesn't imply the current system is perfect either. You should use your authority to make decisions about what does and does not go into the game, not to shut down discussion about how things work right now, or could work with tweaks.
  8. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    I am not saying game designers are infallible. I am saying that it is their job to try and look beyond what the player says they like/want. Most of the time they fail at this, myself included.

    I don't think the current system is perfect.

    Suggesting we allow players to choose spells is not a "tweak".

    I have no interest in shutting down conversation on any topic.
  9. richy

    richy Well-Known Member

    I thought we were trying to explore, as fellow (aspiring) designers, ways to get to that "beyond". Here taking as a starting point what demonstrably does motivate people to play games, rather than unfounded assumptions like e.g. all players only want to win.

    In the spirit of brainstorming and to break the negativity, here are some ideas from an hour's thinking about this. I doubt all these were tried, and the conclusion "choosable spells didn't work so no spell ideas will work" surely can't apply to every one! If it does I will bring more till it doesn't. :)

    Initial spell lineup
    • Number of spells available is choosable: 1 x ult + 3 x normal (as currently), or 2 x ult + 1 x normal, or 5 normal.
    • 4 random spells, but cooldowns can be chosen, you start with say 15 "cooldown points" to spread across your 4 spells. Ultimates cost double.
    • Size of starting cooldown bank used in the above as a difficulty adjustment.
    • Limited cooldown customisation: once you see your random spell lineup one cooldown point can be moved from one to another spell.
    • Same as above but more than 1 point can be reallocated, with the number used as a difficulty adjustment.
    • 4 random spells, random cooldowns, with total cooldown points dependent on difficulty. You might get one with no cooldown at all - eek!
    • As now, but 3 random spells and 1 choosable. Or 2 random + 2 choosable. You know where I'm going with this.
    • As now but you get a variable number of spells, not 4. Ranging from 0 to 10 depending on difficulty.
    • As now but you can get multiple of the same spell.
    Changing spell lineup
    • Each spell when used is replaced with a random one.
    • As above but you can see the next one that will be given (like tetris) so you can plan for it. Or the next 2 or more and higher difficulties can see fewer.
    • Each spell when used is either retained, or replaced by another random spell. Some cost is applied which is different either way and you have to choose.
    • Each spell when used becomes more powerful if retained, but it costs something to retain it and the default is that it gets exchanged.
    • Each spell when used becomes weaker if retained, but it costs something to exchange it and the default is to retain it.
    • The board contains candies (effectively a "store" I suppose) which add an extra spell to the lineup at some cost.
    • As above but a candy that lets you exchange spells like scrabble tiles.
    • As above but a candy that lets you swap your cooldowns around.
    • Killing certain streaks of monsters or with certain streaks of spells allows spell lineup reconfiguring somehow. (I am a fan of streak bonuses in general as a source of interesting decisions!)
    Class related spell lineup
    • Player classes with set groups of spells. Some classes obvs harder to do well with. Extra challenge with those ones. (Good players would not choose the easiest class to max their win rate, more likely the hardest! Or just the ones - whisper it - they enjoyed most).
    • Random spells but player classes having extra affinity with one or more spells, e.g. all fire. Shorter cooldown say, or some bonus effect like an extra smokey flavor tile or extra turns of floe persistence.
    • Player classes with some special spell that's only available to their class.
    • Player classes which affect other aspects of spell use, e.g. quicker-cooldown, more spells in initial lineup, more power-tiles, reduced spell lineup change costs, more candies on the board, etc.
    Spell lineup and monsters
    • The monsters have spells instead of/as well as their own attacks.
    • Only the spells you don't have, they can have. Maybe there is a draft at the start of a match.
    • Maybe just one special strong monster has spells and the rest are just as they are.
    • The spell monster counts for 2 kills, or takes your meter straight up to 5 or something.
    • The spell monster is worth no kills, but drops their spells when they die and you can pick over them to refine your lineup. Or maybe they're dropped as candies.
    • Maybe killing the spell monster is the key to the changing lineup above and forget all those other ideas.
    • No spell monster(s) but the monsters at least can pick up candies if they beat you to them and use them on you.
    • Roderick, Kaytin etc. are actually on the board with spells and you can kill them and take their spells.
    • Or, those guys are actually under your control, like coop. So you have your spells and their spells to use from different positions.
    • Quillsh is on the board and if you push him into a vortex all his feathers fly off and recharge your spells.
    Last edited: Jan 20, 2017
    Batlad and SwiftSpear like this.
  10. Batlad

    Batlad Well-Known Member

    These ideas sound especially good.
  11. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    This statement represents a huge misunderstanding. It is not that I "Assume all players only want to win". It is that I am designing systems wherein the objective is to try and win. I think it is impossible to design strategy games if you cannot assume that players will be trying to win.

    Dunno who that's a quote from but that's not my position. Tons of spell ideas will work.

    I'll respond to the ideas Batlad thought were good:

    - Each spell when used is replaced with a random one.

    Not a terrible idea, but Auro already has a problem where each match lacks identity from other matches. This would exacerbate that problem. (Same answer to the next bullet point you liked.)

    >Maybe just one special strong monster has spells and the rest are just as they are.

    What does "has spells" mean? Doesn't the Lich have a spell? Trickster? Even the Brute?

    >The spell monster counts for 2 kills, or takes your meter straight up to 5 or something.

    Why is this good?
  12. SwiftSpear

    SwiftSpear Active Member

    I see it as a brainstorm, I can dismiss many of those ideas out of hand, many of them have something though.
  13. keithburgun

    keithburgun Administrator, Lead Designer Staff Member

    A few more responses:

    I kind of like this idea, that you build something up while not-using a spell. However, it may be redundant with other systems. In other words, the behavior we want from this may already be covered by other things, and then whatever bonus we'd be giving is basically just deforming the system arbitrarily.

    Yeah, for future versions of the game we'll do player classes for Chaff Mode.

    The thing with allowing players to choose this kind of stuff is, it's going to be an easy choice. I'm picturing a "shop tile" that appears, you walk on it, and then you spend, I dunno, 1 health to take a candy, or some points or something. That sounds fine. The problem is that it's not going to be a choice. It's going to be 90% of the time utterly obvious which spell to take there. So instead what I would do is (maybe) just have candy (rune - are you playing the mobile version too?! Grrrr) tiles that refresh every 10 turns. When you step on them, they subtract 10 points or something but give you the symbol on them, and then pick a new spell icon and in 10 turns you can grab it again for another 10 points. That's not a bad idea but also may not be worth the rule cost.
  14. richy

    richy Well-Known Member

    No am on PC, I just like "candy" - it's fun! The whole brainstorm was for fun really, too much heaviness going on.

    I agree shops are probably bad. Flow-breaking maybe. Refreshing runes sound neat. Random-contents runes might be another interesting source of hidden information.
  15. Bianary

    Bianary New Member

    Here is the starting point I will be working from: At the beginning of each round -- after seeing the level layout -- players select their skill set. This satisfies the roleplayer in me because it lets me feel like I'm in control of my character, and still leaves monster and level design in the hands of the game creator.

    Given that, if a player always selects the same skill load out, it implies to me that the game design has failed to create meaningful differences in monsters and levels. Skill power should vary not simply based on the other skills the player can use, but on the monsters it is being used against or terrain it is being used on.

    So I look at ten levels. If I can say, "My standard loadout gives me the best chances of winning", the game design has failed. If I instead say, "Each of these has a slightly different scenario and I have 9 choose 4 skill combinations (126), I see a different best set for each level." then the game design has succeeded (Level of success could be measured by how evenly the 9 skills are selected at high level play).

    This actually adds a new level of skill: Not only being able to utilize abilities in different combinations, but being able to see -- before actually getting to play the level, but only looking at the enemies and layout -- the most effective combination that will carry you through to victory on it. It loses "being able to make the best of a random set of skills", but with enemies and layout already available for randomization I don't think losing a third is a poor tradeoff.

Share This Page