A Word on Difficulty

I knew from the start that I wanted Another Star to have multiple difficulty levels. There’s really no valid excuse not to, and it’s not all that hard a thing to implement so long as you begin adding the basic framework for it early on.

However, what is hard is realizing just what constitutes “difficulty”. In the first beta version of Another Star, changing the difficulty setting affected a couple of things, but most noticeable among them was that it raised (or lowered, depending on the setting) the maximum hit points of enemies in battle. This is actually fairly common. Countless games simply increase enemies’ attack and decrease the player’s defense based on the difficulty setting and then call it a day.

Trouble is, this doesn’t necessarily make the game any harder or easier. What it’s practically guaranteed to do is make the game more tedious. Sure, boss battles can actually be more fun and challenging if you stretch them out a bit on the harder difficulty settings (supposing the boss is well-designed and fun to begin with). But when minor cannon fodder enemies start requiring double or triple the hits to take out, even if it does make the game “harder” it makes battles tedious and repetitive.

Because the enemies in Another Star‘s first beta were so seriously unbalanced, it made this problem especially noticeable and I realized that I’d have to deal with it somehow. So I tossed the idea of merely making enemies tougher and focused instead on making them more relentless. Now when you play on the two higher difficulties in Another Star, the enemies will still get their old bonus to the STR (strength) stat—they will hit you harder and deal more overall damage—but they don’t get any real bonus to their RES (resistance) stat, and they don’t get any more hit points (except in very special cases). So long as the player can survive the enemy party’s barrage, they will go down just as fast as they would otherwise. This greatly decreases the tedium of playing on the higher difficulty settings. It also gets the player to focus more time on how to react to and survive enemy actions, rather than just figuring out ways to deal greater and greater damage. On higher difficulty settings, some enemies even get additional abilities at their disposal, and some abilities may have additional effects. Enemies also won’t lose as much HP when using abilities (most all magic and abilities in Another Star require spending HP), meaning that waiting for the enemy to drain their own HP is no longer a valid strategy at these difficulties, and the player will have to react accordingly.

I’ve also done my best to make sure that the easier difficulty settings are actually easier. In the game’s “Beginner” and “Easy” difficulty settings, enemies don’t check their HP before using abilities to make sure they have enough HP to actually use them. This leads to the enemy occasionally wasting turns, allowing the player to be more loose with their own strategies. Several enemy abilities are also somewhat nerfed on the lower difficulties, giving the player further wiggle room to make (and bounce back from) mistakes.

But, most of all, I wanted to make sure that there was no penalty for choosing any difficulty setting. Achievements are one thing, but there’s more than a handful of games that lock off features—or even halt the player’s progress altogether—as a punishment for selecting an easier mode of play. I’m sure there’s game developers out there that will disagree, but I find this practice deplorable. A person shouldn’t be missing out on content that they payed for just because they wanted a more laid-back experience. In Another Star, you won’t have to miss out on anything just because you switched to “Easy” mode for a boss fight that was giving you trouble. Likewise, there’s no extra items or special scenes triggered by you showing off in “Hard” mode. The game’s difficulty setting is about difficulty; nothing more.

Another Star is not a perfect game, and I would never claim that it is. There’s probably a lot with the game’s difficulty system that could have been done better, and admittedly there’s a lot about it that I overlooked (maybe I’ll end up making tweaks to it after release based on player feedback). But it is something that I put a lot of thought into, and I encourage other developers to do the same. Difficulty modes should not be an afterthought!

When the game comes out soon, I hope you’ll be able to enjoy the game on your own terms.

3 thoughts on “A Word on Difficulty

  1. I’m glad that you think this way. I also think players should always be given some way to finish the game. I also agree that it’s actually quite a problem in many games that they reward higher difficulties more.

    So if that one player plays on hard he gets 3000 gold while the player who plays on easy just gets 2000 gold? Doesn’t that automatically make the game harder for people playing on easy and easier for people playing on hard? Wanna go to that bonus dungeon everyone is talking about? Too bad, you only finished the game on easy, first finish the game on hard. Want to see the good ending because the bad ending makes you feel uncomfortable? Well you selected easy, so you’ll have to restart from scratch as the good ending requires to play at least on normal difficulty.

    I’m argueing often with other game designers about it. Most of them disagree with me on this concept. Their reasoning is: Allowing the player to change how the game plays is like admitting the players are better game designers than you.

    But I don’t really see that. Sure I’m a great game designer and can design the exactly right difficulty where I can only barely win battles when using the right strategy an otherwise die, but with no battles being impossible to win ever. But just because that works for me doesn’t mean it works for other players.

    Besides what about if you really enjoy the story and want to see it through to the end, but don’t really enjoy the battle system and just want to breeze through it? Some games made me quite unhappy because I really wanted to continue with the story but just couldn’t beat a certain boss battle.

    As for how difficulty should work, I think – unrelated to Another Star – that generally just increasing monster stats by X is a bad idea. Stuff like this is almost impossible to balance on all difficulties. Instead to really create a good difficulty setting you would need to balance it out for each enemy. Generally higher difficulty shouldn’t mean that you should grind more and otherwise the combat is impossible, but rather that you have less room for mistakes. As in on the highest difficulty you can only win battles if the perfect strategy is allowed. On a lower difficulty you can 1 mistake out of 10 decisions and still win. And then the lower the difficulty, the more mistakes you are allowed to make. The easiest difficulty should probably allow playing through the game without having to game up with a strategy.

    One thing that for me would be a reason to NOT have a difficulty setting would be: Not being able to beat a boss and then just lower the difficulty feels like cheating. I had this in Another Star (old beta) often if you can recall. When I died on normal it didn’t really feel like I missed out on the proper strategy. It just felt like it’s impossible without further grinding. So I just took the easy way out and set it to Easy for that one battle and then back to normal. It wasn’t that I just wanted to breeze through that game, but more that I felt I couldn’t win that battle even with the right strategy. But if that would actually be the case (being able to beat the combat) then no difficulty setting would feel more rewarding in the end when I win the battle anyway.

    I was also thinking about very uncommon ways for difficulty settings, for example I thought of something like when you level up, instead of just saying “STR increases by 2, RES by 1, AGL by 1” it will actually offer a “Custom” difficulty mode that when switched on will prompt you as to how much the stat should be increased by and allow everything between 0 and 9 per level (still displaying the recommended number).
    Also not changing the difficulty at all but just the exp rate seems quite fine in games where you can actually grind (and grinding actually has a meaning).

    Generally I also like an approach you often see in WRPGs: Only have one difficulty setting which is perfectly balanced, but don’t encrypt any of the files used by the game so it can be easily modded and save game edited.

    Lately I’ve been playing Might & Magic X and really like how the whole game basically read from readable CSV files for almost everything and you can simply change those CSV files and completely turn in into a different game if you want.
    That game by the way offers two difficulties: “Adventurer” and “Warrior”
    “Adventurer” is still quite challenging especially early on, but you end up with a lot of gold and a lot of room to create messy builds and still being able to beat the game while “Warrior” requires min-maxing and you will have constant gold problems until the very end. I really liked how that worked out, though it’s a bit too easy towards the end on Adventurer.
    (Hint: The game does NOT allow grinding. Killed monsters do not respawn.)

    1. So if that one player plays on hard he gets 3000 gold while the player who plays on easy just gets 2000 gold? Doesn’t that automatically make the game harder for people playing on easy and easier for people playing on hard?

      This is also why I wanted, from the beginning, to split Another Star‘s “game difficulty” from the “grind rate”. (For those reading who don’t already know, you can set them separately in Another Star.) Having to grind doesn’t necessarily make a game easier or harder; it’s mostly just a different sort of experience. I knew I wanted to enjoy the game by having to explore in order to grow strong enough to progress, while others were going to want to just zip from story point to story point, and still others were going to want a Dragon Quest style experience where they have to walk back and forth for random encounters with the occasionally run back to town for supplies and healing.

      I’m [arguing] often with other game designers about it. Most of them disagree with me on this concept. Their reasoning is: Allowing the player to change how the game plays is like admitting the players are better game designers than you.

      There’s a saying in screenwriting along the lines of, “don’t tell actors how to act.” What it means is to put as little acting instruction in your movie/television/stageplay script as possible, instead focusing on the dialogue and action with only a rare and occasional hint on how a line is to be delivered. In short, get out of the way and let the director and actors do their jobs.

      I think the same basic advice can be carried over into game development. Don’t force players to always play the game your way and only your way. Granted, as I designer I struggle with this too, but sometimes you have to learn to give up a little control to the player here and there.

      Not being able to beat a boss and then just lower the difficulty feels like cheating. I had this in Another Star (old beta) often if you can recall. When I died on normal it didn’t really feel like I missed out on the proper strategy.

      Thanks to a friend who gifted it to me, I’ve been playing a lot of Skyrim lately. It’s an enjoyable game, but the difficulty is all over the place. I mowed through the “final boss” yesterday like it was no big deal, but the random mooks leading up to him were able to kill me in one hit at over half my health, while each of my own hits would only shave off a tiny sliver of their own health bar.

      It got tedious to the point where I finally “rage quit” one night while going through the final dungeon. Sure I could lower the difficulty setting, but like you said, it felt like cheating because it didn’t feel like my strategy in and of itself was at fault. (Honestly, all of the Elder Scrolls games I’ve played feel like cheating because of how dependent they are on saving and reloading to progress through even the most basic of battles. They might as well be called “Save Scumming: The Game”, but that’s another discussion for another day.)

      I gave up on Dragon Age for a similar reason. I got stuck at a point where I couldn’t defeat a certain enemy no matter what I tried and the game wasn’t giving me any feedback on what I was doing wrong other than, “your party is dying.”

      I think one of the big challenges in game development that people gloss over is giving the player some sort of feedback on when their actions aren’t working and they should try something else. If I need to change my strategy or approach to a problem, I should be able to realize it on my own in the game. If I’m doing something “wrong”, the game should let me know somehow (and preferably not in a text pop-up).

      Hopefully Another Star is a lot better about this in the current version, but by my own admission it’s far from a perfect game. If nothing else, it’s given me a lot of insight in how to approach future projects.

      Lately I’ve been playing Might & Magic X and really like how the whole game basically read from readable CSV files for almost everything and you can simply change those CSV files and completely turn in into a different game if you want.

      Sadly, practically everything in Another Star is hard coded, so (beyond editing the plain text save game files) modding is mostly impossible without decompiling the whole thing and trying to make sense of it. This was supposed to save time on my end, but it actually didn’t. Maybe I’ll release the source code eventually for those who want to tinker with it.

      1. The example with the actors is really interesting. It is somewhat admitting that players know better how to play a game than game designers, but it’s put into a perspective that doesn’t make you look like a fool (as game designer).
        I might use that for future discussions. 🙂

        And yes you are right, it’s quite important to give the player feedback on what he does wrong. Though I’d usually say this shouldn’t even require restarting the battle. Because I’ve played games where I could always win a boss battle on second try but never on first. Often because I cast a wrong elemental spell which made him wipe my party or simply because I need fire-resistant equipment but the game doesn’t allow me to change equipment mid-battle. It kind of feels “cheap”. Like the first time you go into the battle it’s already with the knowledge you won’t win it and just learn the boss and then let yourself getting killed.

        A better approach would be to give the players hints even far before the boss battles. As you said not in form of a text popup, but rather intuitively added. Say there is a boss with strong fire attacks and you NEED fire-resistant equips to beat him, why not put an NPC in the town closeby talking about “If you want to defeat the dragon in the mountain you better visit the dwarven blacksmith first, he can provide you with fire-resistant equip” or even more subtle like a sign you find on your way to the boss saying something like “Fire dragon ahead, his firebreath burns like hell.”
        In that case if I die during the battle I won’t blame the game for it but remember the hints I got and will be like “Duh, I should have gotten those equips first”.

        Honestly, in my opinion a game should be fully finishable without dieing even once when played for the first time as long as the player understands all the hints he gets.
        In the end this doesn’t even need a difficulty setting, but rather just good game design.

        Also I think it’s already great enough when the save files are in plain text. Too many people encrypt those. Even if Another Star didn’t have any difficulty setting, I could still go ahead and edit my character stats to make it easier if I’m stuck, so it probably won’t stop me from finishing the game.

        You can do great things with plain save files… the first save files I edited were in The Settlers 2 (I think it was 2) where all the resources you and the AI had were saved in a file and when I had to struggle, I simply checked what the AI started with and gave the AI a similar amount of resources than I had, so it’s all fair.
        (Cheating AI is for another discussion. Doesn’t really fit when talking about JRPGs.)

Comments are closed.