Hi! Please bear with me as I worked until 4 in the morning in the pouring rain, slipped down a hill and bruised my back and then was woken up four hours later by my well intentioned son. Who needs sleep, right? Anyways...where was I? Ah, yes: Pulsar. I set out to create a basic 3 lane rhythm game. I just wanted to make something that didn't have to completely be unique and just focus on putting music in that I liked and giving the player an experience they could feel. From there a few design choices drove me to where the game currently is. First of all, one of the greatest weaknesses of the rhythm genre, IMO, is that in the end, the game becomes this thing where the players attention becomes focused on a tiny portion of the screen, where the notes come into contact with the hitzone, whatever that may look like. It ends up giving tunnel vision. I wanted to explore the screen, so I thought "What if that zone travelled across the screen instead?" Thus the scanline mechanic was born. Second design challenge: interaction with the notes themselves. Games on this site utilize a 2:3 aspect ratio, or "portrait" screen orientation. This means, that if I made the player tap on every note it's going to feel awkward as they stretch allllll the way up the screen to get those top notes, and I knew I was designing this specifically for mobile, so my solution was to create a left and right zone, lose the center lane altogether, perfect for dual thumb tapping. At first I also had the "notes" all different neon colours just like the background particles, but that still didn't feel right and was leading to an inability to react quick enough. Solution? Two (invisible) zones, two colours. Intuitive, attractive, done. So, I had the base game, everything worked, I sat down to map out the note patterns for the songs. I opened Audacity, added a click track, synced it up, add my first bookmark and thought "This is going to take hours..." My heart sank. Briefly. Then I said to myself "Are we not living in the future? Am I not using AI??!? to create a game? There's got to be a better way." Long story short, created an automated song pattern generator, created a custom python script that analyzes the song using librosa and matched the two together to output a .js file. NOW we're living in the future. Sorry to be so verbose, the game needs a lot of polish, the patterns are not quite perfect, but they're fast to generate. I'm really eager for feedback about the difficulty so I can create multiple difficulty settings for each song. So, bring it on! My apologies once again to our PC playing friends, this is a MOBILE ORIENTED game. Thank you. Time for more coffee.