v0.1 Retrospective

Hello everyone!

Last Saturday, I released the very first open-beta version of my latest game, Golemancer. After that, I took a break from development for the next few days as I got people I know to try out the game. I showed it to my family members, friends, and to a local game-development club, looking for feedback on the demo. So... how did it all go? What went well, and what went badly? What will I change or add in future updates? Well... read on to find out!

The Good

To be honest, I wasn't very happy with v0.1 when I released it. I thought the game was unbalanced, dull, and had major design flaws that would stick out like a sore thumb to anyone who had ever played a video game before. I was fully expecting to get mostly-negative comments on itch.io, and that everyone who played it would judge me to be hopeless as a game developer and hate me forever. (Okay, maybe not that last part.)

But to my surprise, people seemed to enjoy it a lot! It's still not perfect, of course (more on that in the "The Bad" section) but most of the comments on the game page were along the lines of "It was a fun game, but could use some work on X." Here are some positive anecdotes about what people thought of the game:

Wanting higher difficulty

I sent the game to a close friend of mine. He gave me a lot of useful feedback, and since he also has experience making games, we discussed some elements of its design. After a few attempts, though, he started to find the game was too easy. He started challenging himself to defeat the Rat King in as few turns as possible (I think his record is two, and no, I'm not sure how he managed it). 

Since he had been finding the game too easy, I threw together a special modified version of the game where every enemy would gain customizable stat buffs. Additionally, enemies would sometimes learn random bonus skills. My friend's record (that I know of) is beating the game with every enemy having +4 health and +4 attack. He said that the bonus skills never mattered that much... except for one time where a Viper learned Snipe and one-shot one of his golems.

This is what nightmares are made of

The good news is, my friend was invested enough in the game that he wanted to keep playing it and challenging himself further, which is a really good sign! Having more difficulty options for future updates is something I'll definitely want to add - though, to be fair, the current version only has the first area, so I expect it to get harder anyway.

Blocco strats

Earlier today (at the time of writing), I presented the game at a local university's game development club. One person played it on a large screen (a couple others tried it on their own computers). They got the hang of the game pretty fast, and when they were offered their first golem to choose from, they chose Blocco - not because they thought it was the strongest choice, necessarily, but because they liked its colors.

One of these things is not like the others 

This is another good sign - apparently, my designs are varied and interesting enough that they can weigh into people's decision making. I had worried that some golem designs were out of place, since a lot of them didn't look anything like traditional golems, but maybe that's not such a bad thing! However, that wasn't the only interesting thing that happened while this person was playing.

Blocco has 4 health, 4 attack, and - in addition to the classic Step and Strike - knows the skill "Rearrange", which swaps its current attack and health. There are rooms throughout a run where you can permanently buff a golem's health or attack, too. I figured that attack was more impactful than health - "the best defense is a good offense", after all - so health-buff rooms give +2 health, whereas attack-buff rooms give only +1 attack. (You might be able to see where this is going...)

The person playing got quite a few health-buff rooms, and in each one they chose Blocco to receive the buff. By the time they reached the boss room, Blocco's stats were 12/5 - which, of course, could be swapped in an instant to a 12-attack unit. After clearing out the other enemies, Blocco made short work of the Rat King, and that was the end of their first run.

I had thought that Blocco was a pretty "meh" unit strength-wise, but this playtester proved me wrong. Apparently, there are strategies in the game that I hadn't even considered. This is definitely a good indicator that the game has some depth!

The Bad

I got more positive feedback than I was expecting, but the game is still definitely not perfect. There are a few issues I had already been aware of (and that players confirmed), and a few that came to my attention afterwards. For some of these, I'll write more detailed devlogs in the future where I attempt to tackle them, but here are brief summaries:

Misunderstanding the tutorial

Tutorials in games are often really boring. No one likes having to read through thirty pages of explanation before they can start playing. I tried to keep the tutorial as brief as I could - just five popups in the first battle telling you what to click on and what your goal was. Unfortunately, this was so brief that a few players misunderstood the basic concepts of the game - thinking that only one golem could act per turn, or not realizing it was their turn to act at all. A few context-sensitive tips (e.g. "this golem still has an action available!") should help with this.

Lack of polish and decoration

The current version of the game has only blank black backgrounds (try saying that five times fast!), and animations are very simplistic. This isn't too unusual for an open beta - this is one of many things I'll improve in the future. However, the animations can sometimes affect gameplay experience - for example, when the rats in the final battle get stronger, it's not obvious that the Rat King is causing it.

Path generation

In Golemancer, you only get to see the rooms right in front of you at a crossroads, but there is a complete map behind-the-scenes. This map generation is way too random at the moment; sometimes you'll get nothing but battle after battle, whereas other times you can breeze through taking buff room after buff room. On a related note, there's currently no reason to take a battle instead of a buff room, which makes for less interesting decisions at crossroads.

[Edit: I have another devlog where I work on improving this!]

Enemy AI

At the moment, enemies are pretty darn stupid. They try to move toward golems using taxicab distance (not actual pathfinding) and attack whenever they have the option. In the future, I hope to implement something like Monte Carlo tree search so that enemies are able to make smart moves without me specifically telling them how to behave. They also walk right next to golems without a care, leaving themselves open to counterattack. That leads me into my next point...

The first-approach problem

If two opposing units both know Step and Strike, the first one to step into range leaves itself open, and its opponent will get to strike first. This is bad for several reasons. One is that it encourages the player to hang back and let the enemies come to them, which is much more boring than going on the attack. Another is that it's way too easy to attack enemies and then prevent them from counterattacking, using a skill like Shove, Overclock, or Telekinesis. Almost every unit knows Step and Strike - or some variation of those skills - which makes this a very prevalent issue.

Code quality and project layout

This isn't something that affects the final game directly, but it certainly makes it harder for me to work on it. The current systems for things like actions in battle and the interface for buff rooms is a huge, tangled, horrible mess. This makes adding more content really difficult (since I have to hardcode a bunch of things to make them work), so I'll probably want to completely overhaul these systems before I add much more to the game.

Conclusion

So, ultimately, this v0.1 launch went a lot better than I expected! But I've still got work to do. I had to force myself to keep these summaries somewhat brief, since I have a lot of thoughts on each of them - so if you're interested, check in on my future devlogs where I'll go into more depth.

Thank you to everyone who tried out the game, and thanks for reading this post!