EIA v1.27.03 Beta Hotfix #4 Available

[JimmiMagnus]

And here's a savegame where both Austria and Spain made uncons in the same month.
Log shows +3/-5 for both.

[pzgndr]

It's the latest patch, you posted just above.

I got these and I'm reviewing them. Fixed the Free State contol, Morocco conquer, and Russian leader issues. Looking at the rest. I sent you a PM. Take a look and get back to me?

[pzgndr]

Hotfix #4 posted. This is again a comprehensive update with Data and Setups files included. Changes include:
- added nation_clear_all_land_factors_being_built function to clear minor nation pending land builds upon being conquered
- added warning message if eligible enemy city is not under siege
- adjusted naval AI for France and Russia to build some ships in Toulon and Sevastopol
- added delete_unit_from_area checks for leaders being captured
- fixed nation_scan_all_areas_for_control logic error for city control
- fixed Nation's Statistics screen to not show false wars with kingdom minors
- fixed is_loc_blocked for loaned units check blocking some legal moves
- fixed ai_is_my_capital_in_danger for # turns for enemy to reach
- adjusted ai_build function for minor free state bonuses (now 10%/20%/30%)
- fixed ai_build function for various money and manpower discrepancies
- adjusted naval AI for invasion supply
- adjusted AI economic manipulation with additional considerations
- reduced AI combat die roll bonus from +2/+1 Hard/Med to just +1 Hard
- fixed check_for_freestate_control to resolve roll for control issues plus implemented a random order check
- fixed combat_find_retreat_area to resolve retreat issues
- fixed nation_scan_all_areas_for_control to remove some v1.27.02 'fixes' that work in debug but not in release
- fixed load_units_to_battle and combat_find_leader to find and load leaders
- fixed surrenderparmscode for setting leaders available to be removed causing an error
- fixed exe_surrender for unconditional surrender PP changes
- fixed area_can_i_lift_siege to verify siege status
- adjusted AI redeployments for France to support potential allies at war with Great Britain or Russia
- fixed area_find_over_garrisoned_city and area_does_area_have_garrison (again) for game hang
- fixed repatriate_forces for release of leaders issue
- fixed Admin Menu to create city garrison with correct nationality
- fixed combat_retreat_units for run blockade battles where losing blockaders must retreat
- adjusted ai_set_move_order for Great Britain in 1805 if France gains control of additional minor navies
- fixed AI naval invasion issue for reinforcements exceeding transport capacity for minor DOW
- fixed ai_select_casualties for naval losses
- fixed combat_adjust_pps for multiple naval combat victors

I had again hoped that this would be a minor hotfix and perhaps rolled into a new v1.27.04 version as a release candidate. But I keep finding and fixing more than I would like. Sometimes I wonder if I will ever actually finish this! Regardless, I think we are closer to the finish line with every update. A lot of what I have been working on are AI adjustments and more obscure bugs that get noticed as more folks play and pay attentions. And I am thankful for the feedback.

Maybe tomorrow I'll say some more about this hotfix and what's left on my ToDo list. If the adjustments and fixes with this hotfix are in fact minor then I'll try to get a new v1.27.04 release candidate out by the end of the month. We'll see. For everyone who downloaded Hotfix #3 (33 players!), please download this latest update and give it a go. Ebjoy!
.

EiANW 12703 Beta[4].zip

(3.21 MiB) Downloaded 17 times

[Naxias]

This is a savegame, playing as Prussia.

The Spain AI declared war on France, stacked it's corps for invasion, and then it's just been sitting on it's thumb, not moving for I think over a year now.

Even with France, at some point, as weak as can ever be expected it still did nothing. Russia and Austria also sent troops to just skim the French border, but they too are very passive in terms of engaging France.

[pzgndr]

The Spain AI declared war on France, stacked it's corps for invasion, and then it's just been sitting on it's thumb, not moving for I think over a year now.
Even with France, at some point, as weak as can ever be expected it still did nothing. Russia and Austria also sent troops to just skim the French border, but they too are very passive in terms of engaging France.

Thank you for this game save. This is the kind of AI strategy stuff I need to troubleshoot. Short answer here is that ai_redeploy_available_units had Spain checking for redeployments into France, but Prussia is not at war with France and I failed to include this condition possibility. I need to do a thorough review of all of the logics and make some adjustments.

The problem is trying to include the most common combinations and permutations of who's at war and who's not, who's an ally, etc. and set what percentage of total forces to fight with. A little, a lot, a full court press? I also need to consider various ahistorical conditions to ensure the AI does something worthwhile.

So, longer answer is, rather than shoot out a quick fix I'll be working on this for the next update. When I started looking at this, I was worried that I may have made a late adjustment somewhere and broke the code. Good news is that's not what happened, but I need to do some more homework. I assume you got to December 1808 without any significant gameplay issues, and that's good. We're haggling about some AI performance issues, and that's where I'd like to be right now. Thanks!

[pzgndr]

I mentioned that I would say something more about Hotfix #4. So, a few comments.

I had some complaints about the AI die roll bonuses, so I reduced AI combat die roll bonus from +2/+1 Hard/Med to just +1 Hard. This helps make the AI more challenging, and when one considers all of the other shortcomings of the AI then a little help here should not be a game-breaker. And it's not, unless you're trying to beat the computer and conquer all of Europe. Personally, I don't play wargames to do that but some folks do.

A lot of AI naval issues were addressed and resolved. A couple of late items included the naval AI casualties being wrong and the PPs for multiple naval combat victors were not happening. I had to verify the rule and then add loaned and allied victors to the PP checks. In my 1813 playtest last week as Turkey, Spain DOWed Morocco and I got control during diplomacy but then during Spain's naval phase it failed to execute an amphibious invasion. Why? Usually if it meets the condition for amphib during diplomacy then it should be good during naval. But apparently during reinforcement Spain added factors to the landing corps which exceeded the capacity of the amphib fleet. So I had to add extra checks to the AI reinforcement code.

Speaking of builds and reinforcements, I spent a bit of time going throught the entire AI build code, again. I had noticed some "leaks" of men and money and it took time to track down where it was happening. And then I noticed some discrepancies for AI placing factors during reinforcement, and that took some time to figure out. I think we finally have 100% accountability in the bookkeeping for builds and all factors are being accounted for during reinforcement. All good now.

I continue to watch for little issues during my playtests, and then stop to investigate and resolve them. I'll be working on Hotfix #5 but if all goes well and the fixes are minor, then I'll roll everything into v1.27.04 as a release candidate later this month. Based on what I just posted above, I have some work to do on AI redeployments plus I have a few items on my ToDo List. It's all coming together. As always, if anyone sees any issues, please let me know and provide a good game save of the prior turn/phase so that I can recreate the issue for troubleshooting and resolution. Thanks. Enjoy!

[pzgndr]

The Spain AI declared war on France, stacked it's corps for invasion, and then it's just been sitting on it's thumb, not moving for I think over a year now.
Even with France, at some point, as weak as can ever be expected it still did nothing. Russia and Austria also sent troops to just skim the French border, but they too are very passive in terms of engaging France.

FYI, I want to provide a follow-up. I've made several (many?) AI adjustments based on my 1813 and 1805 playtests this past week. I was still seeing behaviors for redeployments, advances, and withdrawals that I didn't like so I made improvements. I was also still seeing some lapses of war that shouldn't happen so I made some adjustments. There will likely still be occasional lapses because I cannot foresee every situation, but the AI should not be making more than 1 or 2 minor DOWs at a time and the priority for attacking minor capitals is now higher than attacking other enemy units. It should be better now.

I plan to playtest again with a second round of 1813 and 1805 to verify all is well before I post an update. Anyone interested in playtesting my draft updates is welcome. Just send me your email and I'll add you.

[beamslam]

Hello, Good to see this game is still being worked on!
Having a few notices after playing with the latest update version.
The 1813 scenario. French solo. The date is now February 1814.

I'm not an experienced player so might have missed some other issues. The interface is a bit weird sometimes, when selecting and deselecting units etc. But anyways, something should be corrected; I have the Third part combat system on to test it as well, and a couple of times it couldn't load the battle file. It said: "Invalid aa Value (Area)!" But the strange thing is the battle file was not saved for this battle?? Instead I had saved another battle file but far away from this area one or two turns back. I couldn't load the previous file and proceed with the internal battle resolution instead.
By the way, is it only asking for this option when I am attacking? I haven't let me chose the option when defending.

A minor error: it shows a wrong game year right after the first turn in the Show national characteristics window. And it kept the last date in a newly started 1815 game. The date at the bottom of the map window is still correct.

Generally the AI is good. The Coalition was mostly defensive in the spring and summer. Sticking most of the forces around berlin. Hard nut to knack. After a failed attempt on a direct assault on berlin my army was badly mauled and retreated behind the Elbe to wait for reinforcements. I deliberately made a flank weak as a bait, and Alexander couldn't resist and crossed the river, where I stuck him with almost all who could still walk. Alexander was defeated and retreated. He tried twice, latest time another place in middle of winter with his now depleted army. Still have to resolve this battle. The Prussians have not moved forward in support yet.
In Spain I'm now conducting a withdrawal after some back and forth movements and battles.

So in all, except for the Battle export function i'm pleased so far.
I do wish it would be possible to have any reinforcements for a battle incorporated in a battle export file, but guessing that would be out of scope of this project atm.

[pzgndr]

Hello, Good to see this game is still being worked on!

I have the Third part combat system on to test it as well, and a couple of times it couldn't load the battle file.
By the way, is it only asking for this option when I am attacking? I haven't let me chose the option when defending.

A minor error: it shows a wrong game year right after the first turn in the Show national characteristics window. And it kept the last date in a newly started 1815 game. The date at the bottom of the map window is still correct.

Generally the AI is good.

Good feedback. Thank you! I fixed the year display in Nation Info and Free State panels. For the battle files, are you following Section "11.6 Third Party Combat Resolution (Optional)" as described? We haven't looked at this in years but players that used it back in the day verified it was working. I'll take a look and see what might be the issue. And I'll look at maybe allowing battle files for defending.

In my 1813 playtest ongoing now, I've made several more minor AI adjustments to improve performance. Over the years, the challenge was first to get the AI to do something that it wasn't doing at all, then to work out the issues where it was being stupid or not performing correctly, and now to fine-tune it to optimize behavior. We're almost there. Close, and getting closer! I should have another update out next week.