[B135] Island migration is not working correctly?

For whatever reason, I had the scenario that one of my 2 islands' initial generation finished much quicker than the other one. This resulted in that the second island was 16 generations ahead, while the first island was still filling its initial population. However, I have selected island migration every 5 generations, how can this work if one island is 16 generations ahead? Shouldn´t the second island have been waiting at generation 5 until the first island is ready so that it can migrate the population as specified in the settings? Otherwise, the migration between the 2 islands isn´t possible at all. Which now makes me wonder: does the island migration work at all?
Attachments
Untitled.png
(594.74 KiB)
Untitled2.png
(505.48 KiB)
  • Votes +2
  • Project StrategyQuant X
  • Type Bug
  • Status New
  • Priority Normal

History

g
#1

geektrader

25.01.2022 18:31

Task created

g
#2

geektrader

25.01.2022 18:31
Voted for this task.
b
#3

bentra

25.01.2022 22:54


May I direct you to this bug that was fixed:
https://roadmap.strategyquant.com/tasks/sq4_7841


Now that this is fixed, if I understand correctly what mark said, the migrants will NOT queue and wait for the next available island to become the correct age as I suggested. BUT it will search for another island of the correct age and if it does not find the correct age the migrant is simply lost. I can add that when I was really examining the copulation function of SQX it usually produces a child as one tiny variable from one parent and 99% of the other parent (mimicking the mutation pretty much,) The copulation function does seem to work as intended but could probably be optimized.

As for filling up equally, that is weird, I reported a similar bug here but it seems to be fixed when I use 12 islands they all fill up equally fast now:
https://roadmap.strategyquant.com/tasks/sq4_7723





b
#4

bentra

25.01.2022 23:06
After a quick reread of:
https://roadmap.strategyquant.com/tasks/sq4_7841

maybe they do wait in a queue for the next available island to be the correct age. Either way, I saw some signs of migration and no signs of migration to the wrong aged islands anymore.
g
#5

geektrader

25.01.2022 23:32
Thanks for the information, appreciate it. I´ve been closely watching this. While Island #2 was filled up fully, it started to go into the net generations, up until generation #10, Island #1 was still in the initial population phase (not sure why it was THAT async!). So no migration could have happened at generation #5 at all and it most likely was discarded. That should be addressed in my opinion.
E
#6

Emmanuel

29.01.2022 14:50
Voted for this task.

Votes: +2

Drop files to upload

or

choose files

Max size: 5MB

Not allowed: exe, msi, application, reg, php, js, htaccess, htpasswd, gitignore

...
Wait please