>>13175254https://www.youtube.com/watch?v=Fegs-XVKgnM>>13175263In that case I optimize for minimal time series and maximum impact. I decide on my wank as a reward/relief function, when I don't have a woman to rely on, and minimize the time series data I collect by speed polling while parallel imagining something.
Usually the shorter the wank period the better the reward/relief function as evaluating a long time series of wank metadata ends up reducing the impact due to time spent on processing large amounts of data.
Shorter polling periods of data gathering also helps maintain the impact of the wank session so if I do find something in my non-wank time I simply add it to the cache of short-term potential and let the desire to wank grow, which ends up making the wank better.
>t.time optimistYou can substite 'wank' for 'sexual activity with preferred partner/object/human/animal' or whatever as here wank would be x and y would be the time series data modulo tinglies.
