EMPIRE’s Beta has been a strong success so far, with very active participation from a lot of great beta testers. Many of them were also fantastic testers largely responsible for getting AURO into the balanced, tight and interesting little gem that it is. Good testers are pretty much *as* important as a good designer, and we’ve definitely got them for EMPIRE.
So let me fill you in a bit on what we’ve been working on. This is a pretty highly detailed game design journal for the game, so I hope that it’s interesting, and please let me know if you have any thoughts.
The Game Designer’s Favorite Phase
Right now, we’re still in the phase where we’re making rather significant changes rapidly, trying new ideas, re-arranging stuff, and even removing/adding subsystems where needed. For a game designer, it’s quite scary and stressful, particularly for one like me who is always treading new ground. However, it’s also the most fun and interesting work a game designer gets to do.
At the start of a game’s development, it’s definitely fun to “imagine” all these ideas for how the game will work. As a kid, this was always my favorite thing to do, because the game in my head was always way more exciting than reality. However, as you get to be a more experienced game designer, you start to realize that fact a bit more, and that really drains the fun out of the initial “thinkin’ stuff up” phase. So that phase is still fun for me, but it’s a bit diminished by the feeling that I know 70-80% of what I think sounds good now will have huge problems in reality.
Then there’s the implementation phase, where you and the team really start putting stuff together, building a prototype / alpha version. This is where the things that seemed cool and exciting to you at the start begin to reveal themselves as either not interesting or outright broken, and you go into a sort of panic mode of trying to fix stuff so that it at least sort of works, without having to upend everyone’s progress.
Eventually, you get to the phase we’re in now. I’m not sure what to call this phase. It’s something like an equilibrium of that classic design loop: Design – Implement – Get Feedback – Redesign. This is where testers come in super handy. Spending a good few months in this phase at least is crucial for producing an interesting, deep game.
While there are definitely challenges ahead, I’m really confident that EMPIRE is going to make huge leaps and bounds in the next few weeks.
The Difficulty Challenge
“Difficulty” is always something hard to nail down in games, but it’s particularly annoying at this stage. The problem is that we can’t really spend too much time “honing” the difficulty, when the way entire systems work is in flux. So what we’re sort of stuck with is having to playtest game versions that are constantly out of balance, difficulty wise.
In my view, it’s actually always better to err on the side of slightly-too-hard, if anything, although the game has currently been going the other way. Combat tends to be a push-over right now, since your army and that of the opponent tend to be similar in size, yet you have tactical cards and the opponent doesn’t.
The solution, I think, is to simply increase the quantity of monsters slightly. I don’t want monsters to have “AI” (as AI is really a pattern that the user learns and exploits quickly). It’s better if they are totally predictable, but simply outnumber you.
In fact, this points to a larger game difficulty problem, which is that of monsters simply not multiplying enough in general. Currently, a good player is able to keep them completely at bay, on the defensive, and even to basically wipe them out completely. In fact, we’ve had two of our best testers “solve” the game (for lack of a better word) by creating and endless monster-beatdown loop where they are never threatened again.
To solve this problem, we’re trying a few ideas.
- Reduce the amount of “healing” that happens when you knock out a monster lair. We’ve already implemented this, and it seems like a no-brainer now. Being able to near-completely fix the land by killing monsters resulted in a kind of positive feedback loop which allowed testers to “solve” the game.
- Simply tune up the monster production/growth knobs. We have an algorithm for this that involves mostly two factors: how much Desolation (drained tiles) there are on the world, and a timer. Perhaps we need the timer to matter more, so that even if you’re keeping the world somewhat clean, there’s still only so much you can do. Pressure should be increasing no matter what.
- Increase monster quantities in general. I’ve already talked a bit about this, and I think we’d probably do this, but just by the tiniest amount.
- Increase monster HP/Damage. I’m more OK with increasing monster HP than damage, however, I like that currently the player’s Soldiers / Cavalry / Archer stats are exactly the same as those for monsters. I think it really makes the game easier to process.
- Give the monsters some deterministic / input-random rule for combat. What I mean is, perhaps the monsters “do something” during combat. Like, they move in some non-totally-uniform way. This would have to not be just a “random movement”, because that could be totally unfair for the player. It would also probably have to be something that’s super easy to learn and understand – not some weird rule like “they always move counter clockwise around point 4,4″. It would have to be something that doesn’t increase the calculatey-ness of the game, which is already quite high.
One possibility for this last item might be something like: Every turn, one monster is highlighted, and it passes on. That one monster does something special. Maybe he’s invulnerable for that turn, or maybe he doesn’t attack but moves twice, or something like that.
Of course, this extra rule might be completely unnecessary, and we might be able to achieve what we want with the other things (which I’m thinking is the case). I just thought it might be interesting to share such a thought with you to let you see more of the process.
Early Game Challenge
We’ve said before how one problem with the Civilization games is that they’re super interesting in the beginning and then taper off quickly in the mid-game. Well, EMPIRE also has a bit of the opposite problem, where the very early game is actually the most boring part.
This is because monsters don’t get at all threatening until ~30 or so turns in, once you’ve started to suck some of the resources out of the land. It’s also because you just don’t have that much to do. There isn’t as much of the Civ-style “choosing from a ton of buildings” and micro-mangaging lots of little units. So we have some solutions to this problem.
- An “Advanced Start”. For instance, right on turn 1, your city levels up, so that you immediately have a choice. Further, a monster spawns right outside your city always. This yields us some fixed amount of improvement in this area, but I think you’re still going to have a bit of a slow early game after that initial choice.
- Have “End Turn” actually be more like a “Next Event” button. So, instead of ending your turn to go onto the next discrete turn specifically, it would fast forward through multiple turns, until you reach the next “event”. The only concern I have is that the player would need some way to also manually end turn, in the case that he wants to do something specifically next turn even though there’s no event then. So, there would need to be an extra button. Still, it may be worth this cost.
- More meaningful choices / less meaningless ones. We’ve found that exploration – manual exploration, where you select tiles to be uncovered – is actually kind of a false choice (actually, this happens to be the case in almost every game I’ve ever played that has exploration, but that’s a discussion for another time). So, we’re thinking of making exploration more of a single-button “explore all around your city” move. Doing this would remove a sort of “busy-work” early game activity.
At the same time, we’d like units to be purchasable at any time you can afford them – not just at city-level-up time. Potentially, there could be some other way that you use units. Maybe you can “assign” several of your units to do some job. Maybe it “costs” a Soldier to build a Farm, or two soldiers to build a Keep.
Or maybe monsters around your cities are “hidden” – you can’t see them at all, until you assign one or more troops to “scouting duty”. There’s some “Scout Bar” (which may be hidden to the player) for each city, and when it fills up in a given city, all nearby Monster Lairs are revealed. Again, just an idea!
So anyway, this is just a part of where we are in the development. There’s also big considerations for cards, how terrain works, and how to represent some things in combat. However, that’s not to say that nothing is working. The game is actually already quite fun, despite the horrible imbalances, and testers feedback has been very positive and reassuring.
I hope that reading about these problems was interesting for someone other than myself! Let us know if you’re interested in helping us test the game over at the forums.