Permuter Retrospective 2
this is basically a devlog at this point but playtest 2 is done! im really feeling that playtest 3 will be the release canditate lol
this process has been really nice. its made me realise how much back and forth + playtesting games like this require
i'm going to keep going until it's Steam-able but i've already got a lot of big takeaways from this project:
1 - balancing is really really hard. for example, this game has a "merchant" upgrade, where enemies have a chance to drop an extra coin. enemies drop coins on death. one common game feel tip is to increase the amount of enemies but decrease their health. however, if i increase the amount of enemies but decrease their health, this makes the "merchant" upgrade soo much stronger, because its strength implicitly scales with how frequently an enemy death occurs. balancing one thing can throw something else out of balance. i kinda understand why balacing games are a constant effort years after a game's launch
2 - ...im never ready for how much time polishing takes lol. i made basically the entire game in 3 days and yet i estimate it's going to take more 3 months to get it to the level of polish i want. that's not even adding more features or making more content, just tuning what's already there
i now understand that if you're making a systems-y game, it's just so unlikely that the first draft will be good. it'll probably go through many huge sweeping iterations. these iterations could touch every single facet of your game, so the more surface area your game has, it'll require exponentally more iterations
in my case, taking on feedback and iterating on my game required so many changes and touched so much surface area that the second pass of the game took 10x more effort than the first
i think this has put into perspective how impossible my other projects would be if i tried to continue them
some of my previous prototypes took months to make, (e.g. Autoficer, left image) and knowing what i know now, i can't even imagine how long it would take to complete
doing stupid probably-not-real math, if a 3 day prototype ends up taking 3 months to finish (1:~30 ratio), then a 2 months prototype would take ~5 years to finish. which actually sounds like it could be an accurate timeline for a game like this lol. Autoficer is also a game with not many similar examples to pull from, so i'd additionally have to spent a very very long time climbing over design walls. this has really cleared the fog on which projects are unrealistic
since life is getting busier (in a good way), i really dont want to keep spending months on something im going to throw away. projects like this are very good if you're learning, but eventually you have to cash in what you've learned and actually make and ship stuff. for now im quite happy making things that are tiny
so ya.. balancing is hard. polishing is hard. quality games take far, far longer than youd think
these insights are quite obvious in retrospect, but when you're really in the depths of practice, planning, and daydreaming, realities like this find it hard to get your attention until you're far too deep into a project and fate has an i-told-you-so moment when you come back with soot on your face
i will keep it simple from now on!!! im going to make smaller experiences that are better and give them to wayyy more friends to try out early. my friends have definitely been the best thing about my game dev hobby... thats what its all about, making something for the girliesss
excited to keep working on this and i feel like i can see a finish line lol. all that's next is more upgrades, 2 small bosses and a nice art pass!! tysm for reading
