CrewContracts - onUpdateBB #5596
Conversation
I'm not suggesting any changes to the PR (haven't looked at the code, yet), but if you want to see how number of available ships in the market place was done, I wrote a post on that in #3243, assuming you're facing something similar here, with number of crew contracts, and how these are updated / churn |
Those plots are mesmerizing... |
| table.insert(nonPersistentCharactersForCrew[station],newCrew()) | ||
| print("Reseeding crew candidates") | ||
| else | ||
| -- 1 in 30 to be removed and then maybe someone new inserted |
There was a problem hiding this comment.
Did you mean 1 in 101? or I do not understand correctly?
There was a problem hiding this comment.
Yes, old comment, numbers have changed.
| if #nonPersistentCharactersForCrew[station] < 1 and Engine.rand:Integer(0, 10) == 0 then -- Not much crew around | ||
| table.insert(nonPersistentCharactersForCrew[station],newCrew()) | ||
| print("Reseeding crew candidates") | ||
| else |
There was a problem hiding this comment.
10 times out of 11 this branch will fire, even if the array is empty, although now it will not cause an effect, it looks suspicious.
There was a problem hiding this comment.
If there are no applicants, array is empty AND a random number equates to 0.
That's one in 11, not 10 times out of 11. The function reseeds the table when it goes to zero and let's it stay at zero for a randomly determined time.
There was a problem hiding this comment.
I was talking about the "else" branch, yes, I said incorrectly, if the array is empty, the else branch will fire 10 times out of 11.
There was a problem hiding this comment.
OK, I go it. Yes, that looks a bit wonky.
| end | ||
| end | ||
| for k,v in pairs(nonPersistentCharactersForCrew[station]) do | ||
| if #nonPersistentCharactersForCrew[station] < math.ceil(Game.system.population) * 2 + 1 then |
There was a problem hiding this comment.
it would be nice to see an explanation of this criterion, looks like something empirical?
There was a problem hiding this comment.
It's from this equation a bit higher up on the page.
pioneer/data/modules/CrewContracts/CrewContracts.lua
Lines 394 to 395 in 0f417f2
There was a problem hiding this comment.
Oh, maybe then we should put it in a small separate function, with a descriptive name?
|
@Gliese852 Thanks for the review! I'm currently looking into the algorithm @impaktor linked to above. May need to rewrite this. I like this one though. |
|
Number of applicants over about two and a half months game time. There is no affinity to any mean number but just a free running capitalistic hell. System Administration Resting has max applicants of 11 and Cydonia 39. In these scenarios I have modified the chance of reseeding after hitting zero from one in ten to one in a hundred. That's maybe a bit too long of a wait but I think it should be pretty high. Sometimes things just go bad. :)
|
|
@impaktor OK, trying your way. I haven't tweaked the values to fit the human market yet. |
|
Graph redone for 2.5 months of human galactic migration. This is with the same variables as for the ship market. Please note that the population base is different between master and this PR. Crew Ad used Game.system.population and the one from the ship market uses station.path:GetSystemBody().parent.population. There is more drama in the earlier code.
|
Introduce onUpdateBB to the CrewContracts module.
Some points:
Fixes #5510