Road to Bimbo Sequencer 2.0: Lessons Learned (Patreon)

It's Wednesday, May 1st 2019. I've just finished coding the first working version of Bimbo Sequencer 2.0. I'm two days ahead of schedule, but I'm not happy. It's functional, but it doesn't feel fun or exciting. Fuck.

Ever since the first game, I've wanted to make a sequel. Bimbo Sequencer 1.0 was drawn and coded in 24 hours, including the time spent learning Twine, a simple game engine. I assumed making games was easy. I underestimated the amount of work another game would involve. As a result, my next two attempts never saw the light of day (until now). Even so, Bimbo Sequencer 2.0 wouldn't have happened without them. This is the story of how two failed games led to a successful one.

Everything starts with Design

One crucial game element that wasn't included in 1.0's 24-hour creation time frame was its design. I had been toying with the idea of a TF game in the style of EYEZMAZE's GROW series for years. I had spent a lot of time figuring out the aspects of the transformation. What would each button do? How many levels for each aspect? What would the "correct" order look like? What about Easter eggs? Before I ever started drawing or coding, I had a clear idea of what I wanted my game to look like.

This was the single most important factor in 1.0's success. Lack of a coherent design was also the reason behind the failure of my next two attempts.

I first started working on a Bimbo Sequencer sequel in August 2018, almost a year ago. It was ambitious. Six buttons instead of five. Full color. Shifting backgrounds. Instead of changing the victim directly, you would manipulate their environment, adding exercise equipment, media with subliminals, sex toys, etc. I was excited. I started sketching. 

A few days and about 500 layers later, I realized I was getting nowhere. In my enthusiasm I had added things with no sense of how they'd fit within the bigger picture. There was no bigger picture. I hadn't established what the buttons would be or how they'd affect the transformation. I didn't know what the winning combination would look like, or how high each attribute would go. I started extrapolating the amount of work I had left and realized I had barely scratched the surface. What had started as a quick side-project would require significantly more time. Overwhelmed, I gave up.

I still hadn't learned my lesson, though.

Realistic Expectations

In November, a few months later, I made another attempt. TFGamesSite was running a contest titled "Virtual Worlds" and the deadline was just a few days away. Perfect. A strict time constraint would keep me humble. I wouldn't overdo it with new features. Instead of full-body, we'd see the victim from the waist up. Instead of full-color, just grey-scale. Instead of a Sequencer game, it would be a branching path Choose Your Own Adventure. I called it LadyQuest.

In LadyQuest, you'd play as a guy who's playing a VR game. It would be a text-based CYOA. I wrote a quick outline of the plot: explore the map, forage for materials, farm low level enemies, pick a specialty (fighter/mage), craft stronger weapons, defeat the evil wizard. As you got deeper into the game, your character would change. It would start subtly. You might catch a glimpse of a dragon behind the clouds, the city's skyline would be replaced by a mountain range, plants would grow outside the window. Soon enough, your body would start to shift. Finally, you'd end up in the real world with your transformed body.

I had a lot of fun with the artwork. Two paths: warrior and mage, each with their own body type, armor and weapons. In a few hours I was done. I even added a secret bimbo path:

Then I started coding. And it all fell apart.

Bimbo Sequencer 1.0 was made using Twine. It was a good choice. It's straightforward and well-documented. However, the first game was already pushing against its limits. For anything more involved, its simplicity would be unwieldy. I needed something that offered more control. An HTML5 game engine. I chose Phaser.

With more control came more complexity. Things I took for granted in Twine, I had to code from scratch in Phaser. With the "Virtual Worlds" contest deadline looming, I started cutting corners. During an all-nighter, about 5 hours before the deadline, I realized: I had no game.

Sure, I had a plot outline, but I had overlooked the most important aspect of a Choose Your Own Adventure: the writing. The concept of slow, subtle changes was cool, but it only worked alongside an engrossing story. If you finish reading an interesting passage before noticing "Hey, the skyline is now a mountain!" it adds to the sense of immersion. On the other hand, when the descriptions are as basic as "click here to farm goblins" - which is what I had - it just becomes another sequencer game: click the right buttons to win.

I thought "maybe I still have time to turn this into a sequencer game!" Unfortunately, the TFs were designed as linear progressions. It wouldn't work. Once again, skipping the planning phase had led to failure. Strike two. Exhausted and disheartened, I went to sleep. 

Focus on what works 

For my third attempt, I decided to go back to basics. My goal was "Bimbo Sequencer 1.5". Nothing flashy. At this point, I just wanted something playable. I'd focus on learning Phaser. The only new features would be color and transition animations. Five buttons, like the original. Keep things simple. This time, I had a clear design idea: each button would add an element to the scene. Tapping clothes would add clothes, shoes would add shoes, etc. Then, on each subsequent turn, added items would level up. After a few weeks, I exported all the artwork and coded the first build of the game. This is what the start screen looked like:

That's what I had on May 1st. Pretty close to the final game. However, after playing for a bit, I realized something was "off". It wasn't fun or exciting. I remembered being delighted with the first game's first build. This time, I was not.

You might have already figured out the problem, dear reader. It took me a bit. 

She starts out already naked! In fact, most endings feature less skin! For an erotic game, that wouldn't do at all!

Thankfully, there was a simple solution: shift the clothing levels. She'd start with the sweatsuit and end-up naked. I'd have to do some more drawing, but it was salvageable. Since I had a couple days ahead of me, why not add a sixth button and take this to the next level? Thus, tanning was added.

It's easier to add onto a solid foundation.

According to number theory, five buttons make for 120 possible combinations (5! = 5x4x3x2x1 = 120). Six numbers would bring the total up to 720! To keep the extra work manageable, I decided the sixth button would only level up once (she's either tanned or not). Still, it'd be interesting if she tanned differently depending on her clothing. It would require way more work, but it would be a clear visual indicator of the interactions between the buttons, and helpful in figuring out the winning combination. I decided it was worth the effort. A sixth button also meant I had to add an extra step/level to every existing button. Now it had become a proper sequel. Bimbo Sequencer 2.0 was on the way!

I knew the extra button would take time, but I underestimated how much that would be. By Friday, May 3rd, there was still a lot to do: Easter eggs, loading screen, help screen, general debugging. Coincidentally, I had forgotten to clear my post queue. Cursed Ugg Boots, which I'd been saving for June, got posted early. It bought me some time.

Failure is a huge part of success.

And that brings us to the happy ending! 

There's a tendency to sweep missteps under the rug, only focusing on successes. However, the things I learned from my failures and the adjustments I made ended up being crucial. I'm recording them here so I don't forget those hard-earned lessons. Without LadyQuest and my other unsuccessful attempt, we wouldn't have Bimbo Sequencer 2.0. 

I hope you enjoyed this post. Thank you for reading!

Sortimid <3