So it took me a while longer than I expected to get my thoughts together, and a bit longer than that to actually get some time to type them out!
Before I get into it I should explain real quick, Devtober is an event where devs spend the month of October focusing on steady progress and regular updates for a project in order to develop better habits, break bad ones, and generally just get into a good pattern for working. I had hoped to join in last year but things kind of fell through and I wasn’t able to so I jumped at the chance to do it this time around.
Up till now I’ve kinda been absent online for a while, between crazy hours at work and side jobs I kinda let things get away from me for a while. For a while now I’ve been working on a bit of a long-term project with my teammate sbq92, it’s something we’ve been slowly building up for a long while now so we didn’t want to try and rush into it for Devtober. Instead we considered a few different ideas and landed on one we thought could reuse some of our existing code and be a bit more doable.
The funny thing is though that I kinda just jumped right into it, and this is actually the first time I’ve covered the idea in detail!
So, the game is tentatively called Arena Quest, and our idea was to do a bite sized RPG that centered around monster battling. The general idea is that the player is a Tamer working their way up the ranks in an arena through 1v1 monster battles. We’ve got a pretty solid idea of how big the game is and while it won’t be very long we’re hoping that it has some nice replayability. I’ll be covering things in a bit more detail as the features are done though.
For now, let’s get on with it!
To-Do Lists, Planning, and Trello
Now I’ve always been big on planning ahead, I have way more notes than I can handle half the time and I usually just get lost in it. So rather than ending up with some endlessly long checklist I tried to push myself to use Trello more. I’ve used it in the past but up till now it always felt like too much work compared to the results, or I’d just forget to update it at all.
For October I decided early on that I wanted to use it more and I actively worked on keeping it updated. It made a huge difference in keeping my thoughts a lot more organized and giving me a good way to jump back into work every day. I’d set myself a goal every day of finishing at least one task and I’d just look for the easiest one each time, a quick bug fix, some polish on a finished feature, whatever. By the time I finished it’d usually feel like a waste to stop with just that and I’d really get into things, it was a great way to motivate myself to get more done, and even if I stopped after that first task it was still progress.
This is something we talked about right from the start, unlike with some of my personal projects we set ourselves some hard limits on exactly how much we wanted in the game right away. Usually when I work on a project I start with one big mechanic and just expand out from there and end up dealing with feature creep, where it’s always just one more little feature until you end up with a project that never ends.
Obviously limiting features means less work in the code, but one thing I hadn’t considered is how much it helps with the features you do include. When you have a set guideline for what the game will handle you don’t have to plan for every outcome, there’s no “What if I want to do this later?” or “Should I futureproof this?”.
You just make what you know you’ll need, which lead me to something I’m surprised I ended up doing, hardcoding features. In the past I’ve avoided hardcoding as much as I can, if I could absorb something into another system I would. But doing that has just led to more and more vague “systems” for handling things that are eventually too much of a mess to work with. Throughout the month I found myself doing it here and there in small bits, rather than bending some existing code to support something I’m only gonna do once I just used some clean new code to handle it, and it went great!
Kinda funny to see planning both as something that went right and something that went wrong. When it came to handling code and making progress my planning helped a lot, but one thing I didn’t do as much was come up with a plan of action for the project in general.
For most of the month I was jumping from feature to feature with no real order in mind. Sometimes I’d get halfway through setting something up only to realize I needed another system in place to finish it. You can imagine the complete mess involved with debugging two different pieces of code at once since one relies on the other to be finished.
Early on in the month it wasn’t too bad, I was importing my systems from my other project, trimming them down to what was needed and making sure to finish anything that had been left incomplete. But as I got into new territory my priorities were a mess, half the time I just tried to work on stuff that would have nice visual results I could show off.
Not making long term plans for what I’d be doing was probably my biggest mistake and the one I’m the most set on taking care of first.
Focusing too much on the "engine”
This is something that’s been an issue for me for a while, despite regularly seeing other devs fall into it and knowing that it’s a common problem. Like I mentioned above I’ve always liked to expand my systems to handle a lot of different stuff, whenever I can I try to bend my code to support new features.
Which in itself isn’t a huge problem, but the problem comes when you try to expand what something can handle to account for new stuff constantly. Or even worse, when you try to account for all of it right from the start by building a framework that can handle anything you through at it.
Early in the month I saw some other devs talking about this and they had a pretty good rule. The first couple times you need to do something, just code it, don’t bother trying to plan ahead or making some complex system to handle it. If it comes up a third time then you know it’s worth some reusable code, and by then you know the types of issues you’ll run into and what you need to plan for.
Whenever I tried to plan out this great new system that could handle all my problems it just became an endless loop of over-engineering it for problems I might not even have.
Well, there’s probably more I can get into but I think that’s more than enough for now!
Anyway, looking forward my biggest goal is to make a long-term plan and make use of what I’ve learned over the month. I need to decide which features take priority and the order to do them.
When it comes to development of the game my main focus going forward is to finish the core gameplay loop: Getting monsters > Fighting battles > Increasing your rank.
Devtober ended up as a really productive month and pushed me through a lot of features I had been struggling to finish in my other project. I learned a lot about how I can avoid some of the issues I ran into that I’m hoping I can apply as I keep going.
Arena Quest still has a little while to go, we had privately hoped to have a demo ready for the end of the month but I guess that was a bit too ambitious given how busy I am with work. We still plan on finishing and releasing it though so look forward to continued updates!
pixelskylines
germansingergirl
yousharknotpass
jacobvanloon
izzy806
gummygunk
2087
dailybridgerton