Matrix Games Forums

I am currently tracking the TB requirements/needs for an 1.03.07 AI vs AI game (closed TBs).
Especially the Italian TB seems to be way ahead of the real historic path.
It's AI vs AI with closed TBs. So my assumption would be that everything should be closest to reality and 100% TBs as it can get. But it seems like it is far from maintaining the historic path with locked TBs.

This is what Wikipedia states for the Campaign:
"Date 9 July 1943 – 2 May 1945"

As you can see, the Event happened on T191, so 11th of February 1945. ~3 Months ahead of time.
Additionally some events, also from Africa, which also partially influence Italian Events when triggered.
African Surrender was alright just one week ahead of time, but the Italian Surrender happened a full month ahead of time and then the Conclusion of the Italian campaign almost 3 Months ahead of time. Note that the one Trigger on T15 was a 97% trigger, somewhat unlucky and may explain some of the early advancement, but finally reaching an advancement of three months feels weird nevertheless.





This Graph is from the AI vs AI game.
That's the Italy TB requirements/Needs, added Afrika AIR DAY requirements whicha re influencing Italy as well by (+/- 1). The following shows the Conclusion Event, as you can see its "target" Date is the 2nd of May 1945, as historically, but it can be advanced quite a bit.



I'd argue to

• delay the "What's next?" Event (increase in TB requirement for Italy) by some turns, to give Axis the time to get the Troops to Italy (see the Graph).
• delay the activation of the Italian Campaign-delay/advance-events (what happens due to NA Surrender if I got it right).
• after the Italian Armstice, since all Italians completely disappear from the whole game there should also be an decrease in TB requirement for GROUND, since as we can see there are several weeks of low TB percentage. The AIR DAY and NAVAL values get reduced by "The Shrinking Axis" Event, but what about the GROUND value? the TB is missing all the RSI Divisions and as we can clearly see this is not made up by German units.
• decrease Africa AIR DAY TB requirements starting around T50 or so by a little
  OR/AND keep the 6a Ital. FTR Group in Africa (2 of 3 Squads stayed there)!

  

  6a.png (656.64 KiB) Viewed 786 times

In my latest test I noticed the AI moved almost all Flak units from all TBs to West front, even those with TB lock.
This causes a lots of strength points losses for these TBs and may be a cause for the early end in ITA.

I see, yes Western TB also in my test has 100+ AA (T73) while other TBs have 0-1 German AAs. Finnish ones are untouched somehow, Italian AA is in western TB as well.
Here's the link to my Graphs, for those want to use it, just copy it as a Excel file, insert Data into the "Single-Letter Sheets" then stuff gets copied into the proper named Sheets and visualized.

Denniss wrote: Thu Jan 04, 2024 3:17 pm In my latest test I noticed the AI moved almost all Flak units from all TBs to West front, even those with TB lock.
This causes a lots of strength points losses for these TBs and may be a cause for the early end in ITA.

Has this been fixed?

don't know, have not run AI tests the last months.

I missed this post when it first came out. In AI tests I've seen results all over the map, but nothing to make me think it is always going to end early. I wouldn't be surprised if with all the data changes over time, that some of the garrison requirements are a bit off. Also, given the way the events can change requirements, but the reinforcement list generally tracks a given timeline, you can have odd things happen. Honestly, I'm done trying to make additional adjustments, but I'm impressed with your testing and research on this. You're welcome to use the editor to make your own adjustments.

The majority of the Flak units transferred West have not such transfers listed, some are even TB locked but the AI moves them anyway.
As Flak counts for both ground and air one will usually see West events coming late due to increased strength but other TBs usually end earlier than intended.

Yes, thanks Dennis for mentioning that. Gary had trouble getting the AI to use the flak units properly, so in the end we had him try to get the AI flak into the WE TB as at least there they would provide some benefits. Like not using the Hungarian manpower, some of these things were worked on several times but in the end we just couldn't get the AI fixed as we wanted. I think Gary had too many conflicting AI routines that had been developed over the years, and by the end, he didn't know where they all were in the code and how they were all interacting. It was like the AI had a mind of its own.

Please have him look at routines so it would not touch TB-locked units. Or Axis minor Flak units.