I've been working on my game for roughly a year now. Even though I went through VERY strong cycles of [IT'S OVER] - [WE'RE BACK] (even to the point of thinking the project hit a dead end) and some pretty big changes in how I view game design, I kept it under wraps. Now that I've probably settled on a set of mechanics and don't feel quite as hopeless about the project, I should probably document a bit of my process before it gets flushed by my dumbass goldfish brain. I'm writing this primarily to sort out my thoughts & timeline so forgive me if this shit's long winded + boring.
During the development of my first game Redpulse, I've had this unspecific but very strong desire to make a mechanically complex shmup. One that'd make me go "oh shit I gotta use x here, and y there", and one that made me feel smart when I put the square peg into the square hole. I wasn't really sure why I wanted this, or what even it was that I wanted exactly. I just had a general feeling, or even something like a kinaesthetic sensation.
I first got this feeling when I was playing Psikyo's games, Dragon Blaze in particular (using my boy Rob). The way I had to think ahead and figure out helpful ways to set up the dragon in advance, clearing out waves as soon as they appear and planting the dragon somewhere while I fuck off to do other stuff. It felt pleasant and "big brained" while still remaining a very simple, direct & immediate like a good arcade game should.
I was also playing a lot of beat 'em ups. I didn't think of them in those terms at the time, but in retrospect they indirectly tap into a similar type of playstyle. They are games of setting things up. You have some pieces (enemies), some ways to affect them (grabs n throws), a strong central goal (group up and crush) and some "cash out" mechanics (stroooong jabs) that you can use once you set your house of cards up. Unlike shmups though, they have very random enemies, which in practice just means that they let you feel smart (or stupid) for longer. The situations constantly shuffle around, and the house of cards you're building has to be built differently. Yet it still ultimately gets built and you get to feel cool once it does.
Around this time, I checked out the arcade version of Gun.Smoke and it SHOOK ME TO MY CORE. It's perhaps the only game in the shmup genre which bridged the gap between shmups and beat 'em ups, atleast in my mind. Sure, it doesn't have the house of cards element to its gameplay because you don't have many strong ways to affect enemy position, nor does it have strong cash-out mechanics. But what it does have is some proper beat 'em up enemy AI, and even some kinda strange version of footsies. This game made me realize just how much unexplored space there is in shmups with regards to dynamic enemy AI, and enemy manipulation based gameplay, even though again, I didn't quite think of it in those terms.
So it started coming together - I wanted a game that somehow naturally merged the beautiful, elegant design of shmups with the more setup & enemy manipulation focused gameplay of beat 'em ups.
My first proper attempt was at the end of 2019 in the form of a little prototype with placeholder assets. It sorta worked, but the set-up element was lacking - you could group enemies but it all felt very reactive in the sense that you'd see a situation happen and make the best of it, instead of creating situations yourself with active play. The prototype also suffered from the typical twin stick problem - after a certain point it becomes too easy to circle the arena and pick off enemies around the periphery. Shooting was also in constant tension with the state-based melee attacks, since it let you move around. Which forced me to increase enemy health and reduce hitstun from shots. At some point, I also ran into the problem that enemies simply wouldn't work together, and I was trapped in the bell curve of difficulty - no matter how many enemies I added, the gameplay didn't meaningfully change or even get more difficult.
I gave up for the time being and focused working on Mechanical Star Astra hoping to learn more shmup fundies before working on a much more unique project. I would sneak in more dynamic enemies and more complex mechanics into shmups here and there (Astra's Exoshell, Gunvein's ship enemeis) but there was no major push towards my goal until late 2022. That's when I made Zattan for a shmup game jam.
Here, instead of starting with a top down beat em up framework, I started with a more traditional shmup and saw what I could do by adding more complex moves & enemy AI/interactions. It lost a lot of the negatives of my previous demo - there was no more circle strafing, adding enemies was meaningful, combos felt somewhat distinct, the player had a game plan. But it also had even less of that house of cards element than I wanted. What's more, because it was a shmup, beat 'em up encounters didn't work that well - they all blended together. This game gave me a good reason to rethink things.
I needed throws for set up, I needed more environmental interaction so even the same groups of enemies could feel distinctly different to fight. I needed more enemy states, and I needed to figure out how to more smoothly integrate shooting and melee.
So, like Zattan, I started with a shmup framework of only being able to look in 1 direction and shoot up, with the enemies coming from above. I knew the game had to take place on foot, so my initial idea was going with some kinda fantasy (or YuYu Hakusho esque urban fantasy) setting. But since animating walk cycles is a pain, I went with the highly economical but cool looking VOTOMS & Megaman ride armor sliding. To shift the gameplay towards set-up, I knew I wanted a more interesting throw game which my previous games lacked. So I added a Zelda/Goof Troop esque pickup & throw mechanic, which jived well with the new focus on environments. I thought about how to integrate melee & shots so I can balance them similarly, and realized that I can simply treat shooting like Resident Evil 4 and Senko No Ronde treat it - a move with startup, active & recovery frames which either slows you down or locks you in place.
Everything was coming together and finally I could start the project that I've been wanting to work on for years! A bit into it, I started to think about its beat 'em up elements more seriously as well. Instead of just being a shmup/bmup hybrid, I wanted it to reflect what I valued in beat 'em ups as well.
So with all that said here's what I ultimately want :
- I want to make a game which combines the simple elegance of shmups with the set-up heavy gameplay of beat 'em ups.
- A game with a small but distinct set of moves that are all useful.
- A game with aggressive, dangerous enemies and a lot of states/interactivity.
- A game with strong cycles of tension and release.
- A game that has the pacing of a shmup, without losing the methodical "plan ahead" feeling of bmups.
- A game that has enough applications to its mechanics that I can create varied pacing by just slightly changing which sides of the game I wanna stress, instead of relying on forced pace breakers.
- A game that relies on natural balancing and has no easy, non contextual situation reset mechanics.
I want to fight cheesy repetitive strats as much as I can, and hopefully make a game that's rock solid and resists the player's attempts to break it without resorting to artificial limitations. Or at the very least I want to find out for sure that I naively underestimated how hard this is.
I wanna make a unique arcadey game that'll trap & frustrate players who connect with it because they'll have nowhere else to turn to for a similar kinda experience. Just as how the games I admire did with me.
Let's hope I'll make this shit work!
