They Just Won’t Leave! Persistent Heroes for Okhlos

Another week passes by in our endless Okhlos development cycle, and that means another devblog update with a few insights into our process. Mostly with examples about how NOT to do something.

This week, we bring you a subject that is still changing a lot. That is Persistent Heroes.

Types of persistent heroes

In Okhlos, you have two types of persistent heroes: Philosopher heroes and Basic persistent heroes. The basic persistent heroes will behave as regular heroes (that means that if they die, they will revive at the next world), except that, once you’ve unlocked them, you can choose to start each session with them. The philosopher heroes are the ones you control directly, and they replace your current mob leader.

Why we put them?

Okhlos is a roguelike-like game, so each time you lose, you have to start over from the first world. We liked that feeling, but we wanted also to have a feeling of progress, besides how good you get with any particular mechanics. In this fashion, we wanted to have some bonuses you can get at the start of each run. Persistent heroes accomplish this, but they are also proving to be very difficult to balance.

Also, we wanted some type of mob management. We always felt that adding too much management to the mob would be counterintuitive, because the mob should be more of an organic thing. The good thing about this, is that you only have to do it when you start the game, not in-game.

This is the main reason we created the Agora, which is a hub where you select these heroes (and also where you can do a few more things ;P).

The Agora always felt right as a pre-game phase. The camera is zoomed in during this part because we wanted to make a clear distinction between the action phase and this section. (Also, I wanted to show off the character sprites). Additionally, we had to remove some hud elements that had no use at that moment.

The Agora is a very small section, and that’s intentional, because we felt the movement between options should be as quick as possible!

Why add persistent philosophers?

That’s a good question! Thanks!

Originally, we had lots of cool philosophers in the game. That was always the idea. You start playing, and suddenly find Aristotle. So, Aristotle will join your mob, and he will give you a particular bonus. If the philosopher you were using is killed, you will start using Aristotle. Because of that, we had to make tons of new animations for all of this special philosopher heroes.

Each unit has a common set of animations, but the units you control, have a lot more, like running AND commanding to attack. Or commanding to disperse. So each philosopher we add to the game, turns out to be very time consuming.

Lets continue with the previous example. Suppose you are still controlling Aristotle, and you find/buy Plato. Now you have two philosopher heroes in your mob, each one with their particular bonus. And if Aristotle dies, you will start controlling Plato, but Aristotle will revive in the next world. So you can have a lot of philosopher heroes in your mob.

We changed all of that, and we made that you can only have one philosopher hero on your mob. And you select it at the start. I particularly liked the idea of choosing your avatar in the game (a decision that can last seconds). It made sense that you could choose which type of bonuses (sometimes with a trade off) you’d want to start with.

We are still tweaking lots of things in order to achieve that each philosopher encourages a different play style. We currently have 20 philosophers in game.

How many can you take to a new run?

For starters, we wanted to have a lot of persistent heroes, so we set up like fifteen of them. The cool thing about having a bunch of them is that it adds lots of replayability to the game. Each hero has a unique unlock condition, so unlocking them all is a huge achievement.

The problem that comes with that is that if you’ve unlocked every persistent hero, and you take all of them to a new run, the game will be stupidly easy. Also, because of the management side of things, why wouldn’t you grab all the heroes you have at your disposal? If you’ve unlocked ten, you would want to take the ten of them to the game.

So we designed a few rules for the persistent heroes window. When you start the game, you will have no persistent heroes. Once you start unlocking them you will be able to take up to three to a new run.

The more persistent heroes you unlocked, the amount of heroes you will be able to take to a new run will increase. Currently, we have seventeen heroes and if you unlocked them all, you will be able to take up to five to a new run.

Unlock conditions

We set up the unlock conditions so that the player would spend some time unlocking the characters. We think that we might’ve put things a little to easy. We find that we unlock a lot of them in the testing runs, so we might tweak the numbers a little more.

We have weird conditions, like finishing the game without Socrates dying once (you may guess what are you unlocking with that) and more traditional unlocks, like defeating a certain amount of gods.

We found that it was difficult to communicate the player that they had unlocked something. Should we interrupt the gameplay session to let them know? Wait till the end of the session? We decided to communicate it via an unlock window that shows up after you died or won the game. It would not interrupt the session, but it had some problems.

Suppose you meet the unlock criteria, but you quit the game before losing or wining? When will you be notified of the unlock? If you continue your run, no problem, because it will show up when you lose or win, but if you select New Game, you will lose your last session’s progress.

We decided that we would just unlock the heroes when the unlock condition is met, and we would notify the unlock next time the user loses. If players do something weird like quitting the current game and starting a new one, they will miss the screen, but they will have the heroes unlocked nonetheless.

Also, we decided that instead of placing the unlocks in the win/lose screen, we would just place a new window on top of that.

Balance

We are still working the balance issues. Like getting too powerful too soon, or some of the unlock conditions being too easy. We ran a test with a company that works with Devolver for playtesting, and we were able to secure some ideas, but it’s clear that we still have a lot of work to do. However, all of these issues might be connected to some overall and ability changes we are planning for all heroes, not just persistent ones, so it might be more complicated than just adjusting the unlock conditions.

Interesting times…

kripto sporidium

I’ve always been strongly against meta power gain in roguelikes. Unlocking new shinies to make me play better has always (to me at least) been tremendously less satisfying than the knowledge that when I eventually beat the game, it’s because I got better, not because the game got easier. And I know ignoring them is an option (and I always do), but their mere presence makes me think less of the game, and gives the appearance of the game patronizing me. “What’s the matter, son, can’t quite cut it? That’s okay, I’ll make it easier on you. ”

Of course, my own opinions and neuroses shouldn’t be used as game design direction or advice. Your decisions are not illegitimate. Game still looks really fun, it’s just an element that’s really against what I personally feel is one of the greatest strengths of RLs over other games.
- Sebastián Gioseffi
  
  You make a very strong point, and more than valid. Adding persistency elements was certainly not a decision we made lightly. As, you say, even when you have the option to ignore them, I agree that they can feel as a crutch. So we pondered a lot about this, weighing pros and cons, before ultimately deciding to add them.
  
  Personally, I am not against meta power gains in roguelikes per se. One of the first memories I have of being aware of a game doing this, was in the ancient cave in Lufia II. I loved that feeling when I would find a blue chest and unlock a new item to take to the next run. It was whole new way to improve the characters, and it didn’t feel like cheating in any way. The game had been teaching me that the characters had to increase their stats and improve their abilities in order to defeat stronger enemies, and this was just another way of doing that.
  
  On the hand, I also loved games like Spelunky, that take a completely different approach. It relies solely on player skill. It took me over 100+ hours to beat hell but all the time my skill as a player was slowly improving, getting better at timing each jump, at anticipating the enemies’ behavior, knowing when it was worth to risk going for an item or not, and so on. By the end, when I finally managed to beat the game, the character has exactly the same stats, the same abilities, but I had become a much better player and it did feel very satisfying.
  
  Each game took a different approach, one relying more on character skill and the other on player skill, but they were both great and I thoroughly enjoyed them. I don’t think any of them would’ve been better changing that. Each approach has its pitfalls though. Grinding in character development focused games is probably the most common and dangerous of them all. Similarly, in pure player skill games you can spend dozens of hours learning skills that will be completely useless the moment you turn off the game. Is it worth it to spend 60+ hours doing the same thing over and over just to enjoy a moment of satisfaction? Wouldn’t it better if it were 20 hours? Or 8? Or why not make it 200? Or 1000? Some people may be appealed by that but many (and I include myself in this group) may not be so eager to jump in (specially when there are so many amazing games to play at the moment).
  
  So in Okhlos we aimed to have a balance of player skill and character skill (or mob skill in this case) and give the players the choice. If you want, you can use the persistent heroes to improve the mob or ignore them play until you beat it the hard way. But in addition to this, they also work as a means to reinforce the idea of replayability (it is not the only thing that can do it but it helps) and they diminish the effects of luck in later runs (you are guaranteed certain heroes/bonuses instead of relying on good dice roll) which is also a tricky thing to balance in roguelikes. Another benefit we found was that it helped showcase the different mob combinations you can create in the game, with the right choice of heroes you can create very unique mobs that feel like starting templates from which to develop. This is particularly true when it comes to heroes that give you a bonus or ability but they also have a trade off (and I think we should’ve mentioned before that many of the persistent heroes are like this). For example, with Darth Clonus, you will get a considerably smaller mob (he reduces your mob capacity) but the mob will replenish much faster, which doesn’t mean that your mob will be necessarily stronger but it will play somewhat differently. And like that there are several others. So, when we started adding all these things, we finally said yes, let’s go this way.
  
  Anyway, I think I made this waaaay too long (and I should get back to fixing bugs and stuff), but I mainly just wanted to say that we really thought about this, I can see your point, it a very, very interesting one and we definitely appreciate the feedback!
  - kripto sporidium
    
    I appreciate the long and detailed reply, and the insight into your thought process really put my worries to rest. My biggest problem is that there is a fair amount of games where the persistency elements are basically necessary, because the game is close to impossible otherwise (I’m looking at you, Rogue Legacy), but if I’m picking up what you’re saying, you’ll be doing it the right way. While I can’t still fully support it as an element in there, I’ll certainly be happier knowing that it’s been implemented with care and isn’t just being treated as a cheap ploy. Keep up the good work.