Ultimate Videogame Topic 3: The Next Generation

[talah]

I finally got Prison Architect, and I'm obsessed with blueprinting my entire map for maximum efficiency before I build anything. I'll spend an hour planning out my prison, spend in-game days building it, and then get bored and start over. So I'm probably doing something wrong there, too.

Well - epic necro here ("Hi Epic Necro, I'm Dad" :D ) but yeah I looked at this a while ago and have picked it up because it was a few bucks in the sale.

And, am suffering a bit from the planning phase problem as well - I want this thing to work well, and it really evokes -feelings- actually, for me, that I'm being fair and doing the right thing.
(The other side of it I guess appeals to my logical propensity - a well planned prison looks a hell of a lot like a CPU and that's obviously a design intent given there are logic gates available as a mechanic in the game.)

So here's a question.

I've built a minimum security prison in my first quadrant which has good "needs" facilities and comprises cell qualities which range from crappy (holding cells), through intermediate and all the way up to excellent. I punish prisoners sensibly for violence and contraband (above default for attacks or weapons, and sober-up time for drugs and intoxication). I also have classes, and reform programs in place and these score fairly highly - ie the "reform" stat is high - whereas the "punishment" stat is quite low.

So the question is - do I set a low "reoffending chance" threshold for parole candidates or do I set it higher and wonder who decided just where that chance comes from?
I had it set at 20% (ie, prisoners will get released if they are up for parole if their "reoffending chance" is lower than 20%), but right now there are two prisoners in the parole room who have chances just above that threshold - 24 and 28 percent (and 15 others up for parole as well).

I'm thinking that, I'm sort of thinking through the answer to my own question as I'm typing anyway - ie, 'it's a game so the reoffending chance is a simple statistical reality' - but I don't think that's fair, as the actual reoffending rate comes from some other hidden calculation along with the way your prison is run.

And this hidden quantity is driving me crazy!
I want to set the parole threshold to 40% "chance" of reoffending, because I think that makes sense if I don't know what determines the "chance".
But I also want to set it to 20% because I would hope that the influence of losing freedoms, and being given every opportunity to change, would have a bigger influence than "oh well this guy might or might not reoffend".

What thinks folks? Simple answers are great - I'm obviously going to decide *something* pretty much right now because I have the game paused :P

(But a bit of background, my 24% guy is in for 1 year on a repeat offense of vandalism. My 28% guy is in for a fourth infraction - insider trading - for 3 years, and he's previously served 24 years for robbery, money laundering and burglary. WHY IS THE VANDALISM GUY IN JAIL OMG LUL 2016)

[talah]

Spoiler: I set it to 25%. Still interested in opinions if this isn't the wrong place to ask :)

[talah]

I think you'll find it's pronounced "Mega Stan"

[talah]

baba is you is a phenomenal game

I'm not gaslighting you here but I feel that baba is a strange exercise in learning public key cryptography (https asymmetric encryption).
The solutions later in the game are extremely hard, and not, I don't think, particularly solvable through logic, rather through trial and error.
They feel very much like predetermined solutions with initial states obfuscated based on the tools at hand
Like a lowkey thermodynamics situation where an ant starting at a leaf on a branch has no problem getting to the base of the trunk, but an ant at the trunk has little hope of finding a specific leaf.


Also I came here cuz I just started playing don't starve and geez - it's a satisfying game even if i have been randomly slaughtered and had that wtf moment more than a few times :)
Was hoping to find a don't starve thread, is there one?

[talah]

baba is a strange exercise in learning public key cryptography (https asymmetric encryption).

yeah go ahead and do that.