Design How to convey information to the player when using moderately complex formulas?

[Setteduetto]

Yo. So I'm developing a turn based RPG-styled roguelike and I've come to the point where I need to start conveying information to the player better, in order to get better playtests.

Basically, each character can perform a basic attack that deals damage equal to their AP (Attack Power) with minor variance that causes it to deal between 85% and 115% damage. In addition, each character has two skills, which tend to have slightly more complex damage formulas. As an example, one skill, "Vocal Flare," hits for (10+MP(Magic Points))/4 and cuts the target's AP in half for a turn. This is about the median for skill complexity.

My problem lies in communicating this to the player. If I reveal the damage formula to the player, they'd be able to make more informed decisions, but seeing (1-(TargetHP/TotalEnemyHP))*AP is going to put a lot of people off. Most people want to relax when playing games, not do algebra!

The alternative would seem to be saying things like "Deal a lot of damage" or "Deal a little damage to all enemies" but that isn't without its own problems. It takes away some of the player's agency to precisely plan their actions, and it means they might make false judgements about a skill's usefulness. There's also the problem that I as a developer, knowing the damage formulas, can't turn that knowledge off when I'm playtesting, so I'd be getting somewhat skewed results from that.

I'm not sure what I'm going to do with this, so any thoughts are appreciated. The game is pretty far along so forgive me if I'm not too open to design tweaks that, while maybe better, could set the project back or take a lot of work. (But don't let that discourage you, any advice is very much welcome.)

 

[IndianaBones]

For formulas that have a single result, just show that value to the player. If the formula has a range of possible values, show the player the min and max amount.

Desktop Dungeons would display the enemy Heath Bar, and when you hovered your mouse over an enemy, it would show a faded version of the health bar indicating how much health would remain if you attacked.

You could still have the formula tucked away in a pop up box or menu somewhere if the player wanted to see it, but in general, min max values and graphical depictions of player actions would be my suggestion.

 

[Setteduetto]

For formulas that have a single result, just show that value to the player. If the formula has a range of possible values, show the player the min and max amount.

Desktop Dungeons would display the enemy Heath Bar, and when you hovered your mouse over an enemy, it would show a faded version of the health bar indicating how much health would remain if you attacked.

You could still have the formula tucked away in a pop up box or menu somewhere if the player wanted to see it, but in general, min max values and graphical depictions of player actions would be my suggestion.

This is an interesting approach and I may go with it, thank you. However there's still the problem of how to concisely convey to the player what stats/mechanics influence that value.

Something like "Deal 10-13 damage. Damage is based on your AP." could work, but I wonder if there's a more elegant solution.

 

[NightFrost]

Most strategy RPGs I've played deal with this problem by not showing the formulas at all. They might give you damage ranges or list attributes that affect damage and mention some modifiers that go into the calculations. For example look at how obtuse Disgaea games can be about the damage you will be dealing. They rely on the idea that bigger numbers are always better (but also let you level up to 9999 where it usually is unnecessary to go past 100 to finish the story campaign, the rest is for the fun of running proc-genned dungeons or fighting hidden stuff.)

 

[Yal]

Most recent Etrian Odyssey games has little icons in a corner of the skill description that show which stat they scale with and which body part it's performed with (head, arms, legs - these can all be disabled by status ailments, making skills using that body part unusable). I find this solution more elegant than the "describe it in the flavor text" solution since it takes up less room AND reduces the amount of redundant flavor text, both noble goals. You could easily expand the system even more, like adding icons for ailments/debuffs the skill causes, and icons for whether it targets enemies, allies or both, and adding little arrows or stars to the stat icon to signify the potency of scaling. This frees up flavor text for completely unique effects (and of course, flavor), like Vocal Flare's "Forbidden art that turns an enemy's mana against them. Deals more damage the more MP the target has."

Showing damage range is pretty standard these days. You could signify scaling by showing damage as two numbers base + scaling; to show the randomness, you could display scaling as two numbers with a dash or tilde between them (e.g. 7 + 3~6), where the base number is actually 0.85x the true base damage, the lower scaling number is 0.85x the scaling bonus at the lowest possible roll, and the higher scaling number is 0.30x the base damage + 1.15x the scaling bonus at the highest possible roll. (So basically you bundle together the randomness into the scaling when displaying the data so you don't need the player to do maths with FOUR different numbers when picking attacks).

Though tbh, I'd say the lowestpossible~highestpossible display is easier to parse... you could put the details hidden on a submenu somewhere (Base: 7 STR Scaling: 3-6) if you want to give the player all the information.

 

[Setteduetto]

@Yal Thanks for the response! I really like your idea of using icons to show which stat goes into something. I could probably use a very simple system of just showing whether or not something is a physical or magical attack and get away with that pretty easily.

I think for now I'm going to start by doing the bare minimum, and change it if there's a problem from there. That probably means an icon to show which stat the skill uses, and a flavorful description to tell the player about the more uncommon attributes.

 

[dannyjenn]

You might not need any explanation whatsoever. Because if an attack has an amazing name, and if it uses a high amount of MP, then the player can readily guess that it's an awesome attack. Plus, once he uses each of his attacks about 10 or 15 times, he's going to begin to learn what each attack's effects are, and he's going to begin to understand which attacks work best against which kinds of enemies. This approach might not work in every game, but it works in pretty well in games that have a lot of grinding. Like the gen 1 Pokemon games... those games didn't tell you what the attacks did ahead of time. For that you either needed a strategy guide or else just trial and error. I think FF1 did this too. And 8-bit RPGs in general, simply because there wasn't a whole lot of room on the cartridge to include lengthy textual explanations for something that didn't need explaining, nor was there much space on the screen to display text in those days. Even though it was due to limitations such as these, the gameplay did not suffer. Players just didn't care that there was no in-depth in-game explanation. (Maybe players these days do care, but only because they've come to expect it...)

(But there's an idea. You could bury the actual formulas somewhere in the game or in the help file for people who want to dig it up, but not some place where it'll be in peoples' faces.)

 

[Yal]

Plus, once he uses each of his attacks about 10 or 15 times, he's going to begin to learn what each attack's effects are, and he's going to begin to understand which attacks work best against which kinds of enemies

Or the player uses one attack 10-15 times, figures it out, and then never tries anything else out as long as their previous attack does the job. In D&D there's a well-known problem that's a type of inverted decision paralysis; players don't even know what their options are because there's so many things you can do in any given situation... and tend to sort of fall back into doing the same thing over and over again if it works well enough. If there's high risks involved (or resource costs), people tend to go for safe bets over something unfamiliar, even if they're not optimal (because usually they don't KNOW the other option is objectively better).

 

[dannyjenn]

Or the player uses one attack 10-15 times, figures it out, and then never tries anything else out as long as their previous attack does the job. In D&D there's a well-known problem that's a type of inverted decision paralysis; players don't even know what their options are because there's so many things you can do in any given situation... and tend to sort of fall back into doing the same thing over and over again if it works well enough. If there's high risks involved (or resource costs), people tend to go for safe bets over something unfamiliar, even if they're not optimal (because usually they don't KNOW the other option is objectively better).

Yeah, I guess that's a downside. But good game design can alleviate the problem to some extent. If from the very beginning each boss requires the player to change his strategy, he'll be more inclined to try out different attacks throughout the game. But you don't want to put the player in the position where he can easily get through 99% of the game with his fallback attack, but his fallback attack doesn't work against the final boss. That would be incredibly frustrating. (Especially if there are no effective attacks available to him by that point, and he'd then need to do a ton of backtracking just to get the game back into a winnable state. Game designers need to make sure this sort of thing doesn't happen.)

 

[Yal]

Yeah, I guess that's a downside. But good game design can alleviate the problem to some extent. If from the very beginning each boss requires the player to change his strategy, he'll be more inclined to try out different attacks throughout the game. But you don't want to put the player in the position where he can easily get through 99% of the game with his fallback attack, but his fallback attack doesn't work against the final boss. That would be incredibly frustrating. (Especially if there are no effective attacks available to him by that point, and he'd then need to do a ton of backtracking just to get the game back into a winnable state. Game designers need to make sure this sort of thing doesn't happen.)

Dark Souls 3 is a pretty interesting case study on this exact effect. The devs experiment a bit with enemies that have varying rhythms to their attack animations, so you need to time your dodges to their actual attack animations and not just do it on reflex X seconds after they start winding up for an attack. The two biggest roadblocks in the game, Pointiff Sullivan and the Dancer of the Boreal Valley, both have very long startup delays on most of their attacks, so if you just get used to the "normal" rhythm of the game it's almost impossible not to dodge too soon and then get hit before you can re-dodge. Technically, the longer windup makes their attacks easier to dodge, but the game conditions you to dodge based on a specific timing instead of actually paying attention to their attacks.

Sullivan (to some extent) and the final boss (moreso) also are notable for having REALLY long attack combos... during most of the game, you don't need to worry too much about running out of stamina, because it recharges really quickly... but the base stamina isn't enough to get past those guys. It's perfectly possible to get into an almost unwinnable situation if you never level up your stamina before the final boss, because you won't have enough stamina to dodge his entire combo AND attack him... and at that point, you're so many points behind in stamina and levelups are so prohibitively expensive, there's no easy way to grind for more. This hit me really hard on my first playthrough, and I was stuck on the fight longer than on any prior obstacle.

 

[woods]

hovering over the weapon gives min-max damage tooltip
damage = 25~37
chance to hit = 75~90

Rclick opens mini window with full attack explanation for those that want to see the math behind everything
flavor fluff text
damage= (10(str*1.5)+(dex_modifier)+(3~7))
chance to hit= (20+(dex+1.5)+(int*2)+(5~10)

====
there is something to be said for giving the player the option to see the meat and taders of what is behind the scenes