Comrades! The time has come in this app's development to figure out one of the most important parts of edging someone to the very limits of their sanity, figuring out how close they are.
But first!
I've finally added spatial audio. It was incredibly easy. All I had to do was abandon the 5-10 hours I had sunk in trying to get Oculus's spatial audio plugin to work. For those of of you who are unaware, which is probably most of you reading this, Oculus/Meta's Unity integration plugin has been a freaking nightmare since it's release. They'll just break that shit for months at a time with no communication whatsoever. It's a small miracle that any Unity apps have been created for the Quest headsets at all, in my opinion. I can't speak for Unreal dev, but I imagine it's the same. So, I'm using the Steam Audio solution now! Which worked for me like a dream. I think it took me 20 minutes to get it all downloaded and implemented. About 2 minutes into the video you'll see it occur to me that I should test the spatial stuff for the video so I'll move my head around and you'll notice that the sound profile of the pumping sound changes. I'm not all happy with that pumping sound btw, gotta work on that. Thinking of breaking the pump sound in to In / Out pieces and then firing them off as part of the animation's timeline. I think that'll sound better.
Now, back to the Topic I'm sure you're actually interested in, that's right, it's Milking Theory time. I'll spoiler this bit for those of you aren't interested.
spoiler
So, as I've mentioned before I'm initially designing this software for The Handy Sex Toy, it's a stroker for penis-y parts. The Handy basically consists of two pieces; A sleeve, much like a fleshlight, and then a stroker that's basically a linear screw rod and a servo motor with a slide that holds an attachment for the sleeve.
The Handy API only really offers one variable to control the stroker in HAMP (Handy Alternate Motion Protocol) mode which I'm using and that's basically velocity. I can control how fast it moves up and down as a percentage of it's full power. But, in all modes, it's possible to control the length of the stroke via setting the minimum and maximum range the slide can move in. This is import as not only do penis-y parts come in all shapes and sizes, but sleeves do as well. All those little bumps and things on the inside of the sleeve feel different depending on where on your penis they're rubbing against.
So what that means is, that in addition to accounting for the length of the users penis, I've also gotta account for the fact that they may be using a different sleeve with an entirely different feel to it.
So here's the plan so far. For my alpha testing release I'm working on a Reinforcement Learning algorithm. Before and after each stroking session it will ask them how close they are to cumming, with the four options:
-
- Not at all.
-
- A little bit.
-
- Getting there.
-
- So close!
Then, for a while, my algorithm will simple assign random values to the stroke length and speed to build up a database of sorts until it has enough information to reliably edge the the user among the different states. Ideally, this will mean that each individual user will have a custom profile built just for their sleeve and body. The longer they use it, the "smarter" it should hopefully get. And at some point I can make it so they can load and save multiple profiles. Long term, I even hope to have them capable of exporting the profile somehow so they can be used in future apps I want to create.
So, working out the algorithm is the next step. To be honest, this is my first time working with this type of algorithm. I have no idea how many times the algorithm will need to work on the random settings before building something that resembles personalized. But, I hope to make 
proud.
Thanks! I'd really like to get an alpha out and do testing soon, so I'm pushing as hard as I can without ignoring my other projects. I totally forgot to mention that I started working on a transitionary scene that the player pops into the office upstairs beforehand. I spent most of today working on it and now I'm not sure I wanna use it. I'm trying to figure out where the best place to do all the tutorial stuff about controller buttons and like sleeve choosing should take place. I was thinking a transitionary scene in the upper room where it's a "normal" milking room and then they "wake up" downstairs in the communist area brainwashing area, but I'm not sold yet. I've also thought about maybe putting like a bulletin board in that opening farm scene that shows controller stuff and maybe streaks or other gamification type things. IDK. I'm also trying not to go too wide in scope when I really just need a prototype for an alpha first.