Historical outcome of events in 1.10

[Pang Bingxun]

In 1.10 unlike 1.09 historical outcome of events does ignore ai_chance. Instead of choosing the option with the highest ai_chance simply always option a will be chosen.

There are two ways to address the problem. Return to the solution used in 1.09, that from my point of view is the elegant solution. Or all events not complying with the presumption of the current approach need to be rewritten. In a first test run i identifed two events(919194 SIK and 919195 CHC) related to the second united front in china that need to be changed, but possibly there are a more.

 

[bosman]

I also prefer higher chance selection like in 1.09

 

[Pang Bingxun]

In a first test run i identifed two events(919194 SIK and 919195 CHC) related to the second united front in china that need to be changed, but possibly there are a more.

For now in inverted the options in those 2 events to keep at least the effect as intended, else any test run is sort of voided by wrong option choosing.

 

[Zsar1]

If one would put the historical option to the top, and then give the (estimated) optimal option the highest chance, mayhap with this behaviour restored one could have two non-standard ways to play: with historical decisions and with optimal decisions...

 

[Pang Bingxun]

Well, in the 2 events here the "historical" option is the second option for the semigood reason that there are 4 other events with the same options but different likelyhoods. The 4 warlords have IIRC 100% chance for acceptence, Xinjiang only 30% and Mao only zero. It makes some sense to keep the order of the option as they were in 1.10. When modding events it is often more convient to only change the ai_chances instead of having to mod the events and possibly the corresponding translation files aswell.