Kristof73 fixes do DD

[unmerged(53922)]

Kristof73 MOD/Fixes to DD 1.2 '36 GC

I am working on MOD which be more ballanced then vanilla. With better and more aggressive AI, but without any cheats - like unit spawning. Human is not penalized either. I can guarantee high quality of this MOD. Soon I show some screens from Japan AI in action.

Kristof73 MOD/Fixes for DD 1.2 '36 GC ver. 1.0 (01.2007)

*********************
******* Bugfixes *******
*********************
- Fixed wrong money value in /db/resource_values.csv from 12 to 160 to the same ratio as placed in db/misc.txt (supplies/money - 20:1). Corrected trades values on money in every scenario. See more info in thread: Supplies to money ratio - that is how to rob AI ...

******************************
******* Game Improvements *******
******************************
- 100% working NoExpForces MOD v 1.1c to DD 1.2 all scenarios, to human players only (AI countries still will be using EXP forces) in Single & Multiplayer with information how to use it in any MOD

******************************
******* Gameplay Balancing *******
******************************
- No overhelming Airpower: Changed in /db/misc.txt from - 0.02 to - 0.10 "Combat Modifier: Total Air Overstacking Mod for each division above two". Now the only effective air unit stack will be 4. It is done becouse underpowered human Air vs AI Air abilities.
- Strategic bombardment is 2 times more effective: Changed in /db/misc.txt from 1.0 to 2.0 "Damage factor versus IC from bombing" and "Damage factor versus resources from bombing"
- Puppets are not true allies: They are building 80 % of garrisons and start every month with 2% dissent. Don't expect that people in puppeted countries love their master and want to die in his wars.
- Brigades are additional equipment only: Changed all brigades manpower cost to 0. TAC, NAV and STRAT manpower cost raised +1 to game ballance. It is done becouse 2 factors. First - brigades are highly ineffective in power/ESE/manpower ratio to vanilla units. Second - units with added brigade has the same amount of manpower as without it. Disbanding brigade is giving to you 0 manpower too.

**********************************
******* General AI Improvements *******
**********************************
- AI airpower tweaked. Interdiction, logistical strike, runway cratering, installation strike missions are turned off, as totally useless in AI management.
- Invasion AI fixed. Now AI will do invasions faster.
- Technology research and building schemes tweaked for all countries.
- No big stacks of AI units in useless places
- All major AI will produce nukes.
- Offensive supply turned off for all AI's. EX. SOV often has units without supplies becouse that ....

***********************************
******* Countries AI Improvements *******
***********************************
- Portugal AI will be building convoys, when needed.
- SPA has removed passivity and handicap vs SPR
- Reworked all major AI's. Now AI really want to destroy enemy, not sitting home and doing nothing.

********************
******* Events *******
********************
- Changed ai_revolters.txt events. Now all countries after every liberation will have free resources/free IC/free manpower. All countries will have leaders and ministers when one time liberated. Second need game reloading before liberation.

 

[unmerged(19841)]

Soon you will find here some fixes and improvements to game play balancing to pure DD 1.2, suggested by me. It will be part of my future Kristof73 1936 MOD. AI will be much improved, but human and any AI country will not be penalized at any form. There will be no cheats like unit spawning eighter. Simply better AI and a few restrictions to human (like 4 planes unit stack will be only effective), to better SP game balancing. Some events will be fixed too, but AI chances will be the same as in original. AI will be build nukes and play after 1946. So ex. Japan after India and Australia conquer, will strike SU if they will be still at war with Germany. SOV after destroying Germany, will spread communism in Asia and Africa. etc. Now I am improving Japan AI. No more defending and conquering useless islands. Their home land will be all the time well defended. After China, they will strike India at full force, and do only landings in Philipinnes and Australia. Next goal will be or taking Suez or DOW on SOV.

I am glad to hear your comments. You can use any of them in your MOD without my permission, with information about original source only.

Kristof73 MOD/Fixes ver. 1.0 for DD 1.2 (01.2007)

************************
******* Bugfixes *******
************************
- Fixed wrong money value in /db/resource_values.csv from 12 to 160 to the same ratio as placed in db/misc.txt (supplies/money - 20:1). Corrected trades values on money in every scenario. See more info in thread: Supplies to money ratio - that is how to rob AI ...

******************************
******* Game Improvements *******
******************************
- NoEXPforces MOD to human players only in Single & Multiplayer with information how to use it in any MOD

**********************************
******* Gameplay Balancing *******
**********************************
- Changed in /db/misc.txt from - 0.02 to - 0.10 Combat Modifier: Total Air Overstacking Mod for each division above two. Now the only effective air unit stack will be 4. It is done becouse underpowered human Air vs AI Air abilities.

- Changed all brigades manpower cost to 0. TAC, NAV and STRAT manpower cost raised from 2 to 3 to game ballance. It is done becouse 2 factors. First - brigades are highly ineffective in power/ESE/manpower ratio to vanilla units. Second - units with added brigade has the same amount of manpower as without it. Disbanding brigade is giving to you 0 manpower too. Now brigades are treated as additional equipment only, as seen by game mechanism.

*******************************
******* AI Improvements *******
*******************************
- AI airpower tweaked. Interdiction mission is turned off, as totally useless in AI management.
- Invasion AI fixed. Now AI won't forget about units stacks prepared to planned invasions.
- Portugal AI will be building convoys, when needed.

**********************
******* Events *******
**********************
- Fixed China Surrender event. Now puppeted Nat China will have leaders, except sleeped by original event

looks good were is the download at?

 

[unmerged(53922)]

looks good were is the download at?

Thanks for interesting. Now it's done in 50%. I think v 1.0 will be ready and 100% tested till end of the month.
Next improvement will be reworked Japan/Manchuria/Puppeted NatChi AI, but game not be balanced with this change.

In final version I will need good speaking English man to correct my errors.

 

[unmerged(49083)]

I'm english! just send it to me by PM, perhaps.

 

[unmerged(53922)]

I'm english! just send it to me by PM, perhaps.

Thanks a lot. When It be ready I will send to you English text. So your name will be as coproducer

 

[unmerged(53922)]

New version NoExpForces MOD 1.1b to vanilla DD 1.2 for all scenarios is out. It works for single and multiplayer. No more any unwanted EXP forces !!! At game start you type your country or countries TAG's for multiplayer and that's all. You can use it in your MOD with some modifications.

Direct download: http://savefile.com/files/412316

 

[unmerged(63886)]

Thanks for the updated NoExpforces mod!

And I'll be downloading your mod when it's available.

 

[nigo]

make this for HSR 2.08

HSR is the best mod outhere

 

[unmerged(53922)]

make this for HSR 2.08

HSR is the best mod outhere

Everybody could use my work, so ...

Soon will be relased the most important fix to hoi2 DD. Fixed AI invasions, so no more AI invasion stacks waiting forever For now it works great !

The same, when turned on only ground attack; strategic, naval and convoy bombardment - AIR AI is doing much better work.

 

[unmerged(49083)]

What difficulty are you testing on? Just want to know, so if you were doing it on Very Hard/Furious I wouldn't put it on normal/normal, see a brain dead AI and start complaining!

 

[unmerged(53922)]

What difficulty are you testing on? Just want to know, so if you were doing it on Very Hard/Furious I wouldn't put it on normal/normal, see a brain dead AI and start complaining!

I do tests on normal/aggressive, but frozen AI invasion is on every level.

Game is only ballanced on normal, becouse on higher levels, countries with big mapower stock has huge advantage vs minors. There is a second problem too. Ex. Nat China is starting with 20% dissent. Dropping dissent down will be some faster with more IC, but Japan will overproduce China in this time .....

 

[unmerged(19841)]

Thanks for interesting. Now it's done in 50%. I think v 1.0 will be ready and 100% tested till end of the month.
Next improvement will be reworked Japan/Manchuria/Puppeted NatChi AI, but game not be balanced with this change.

In final version I will need good speaking English man to correct my errors.

you can pm it to me,I don't have a lot of time during the week,but the week end is almost alway free for HOI2,,

 

[unmerged(63886)]

Anything to report, Kristof?

We want it now!

 

[Gormadoc]

Now this look interresting.

No more Frozen invasion AI. I think i read a post a while back (maybe yours) that talked about a solution with an event that every month turned invasions on/of to stop the AI from forgetting the troops that are waiting to invade. Is that what you do ?

Gormadoc

 

[unmerged(49083)]

Never. Ranger Mike is an amis, and the americans tried to make themselves separated from england by f****** up the english language, with words like 'Armor' or 'Color'!

 

[unmerged(53922)]

Anything to report, Kristof?

I've done:

1. Changed ai_revolters.txt events. Now all countries after every liberation will have free resources/free IC/free manpower. All countries will have leaders and ministers when one time liberated. Second need game reloading before liberation.

This fix problems with annexed countries and liberated with no leaders.

2. With fixed money ratio to other resources, human has much less money from trades.

3. Restriction in 4 planes stack as the only effective, gives more chances to AI in airbattles. AIR AI is mainly focused on ground bombing, so AI will probably kill some more units this time. Not wasting airpower on useless interdiction of units not in battle.

4. First will come more aggressive Japan AI. Such aggressive vs vanilla that Africa or Siberia could be their new promised land .

Now this look interresting.

No more Frozen invasion AI. I think i read a post a while back (maybe yours) that talked about a solution with an event that every month turned invasions on/of to stop the AI from forgetting the troops that are waiting to invade. Is that what you do ?

Gormadoc

Yes. I am testing this right now. I can tell more about it soon. I am not sure if it works. I must check times of invasions with and without my event. BTW. Big stacks of AI units are often becouse "garrison" values. When you set up very low values ex. like in USA files, then AI don't know where to put these units, and often chooses some island. How to fix it ?
US has high value of exp forces to ENG, but they don't want them Why ? ENG have too small "max_front_ratio". Raise this value much higher, and ENG will demand troops from US. At once some big stacks of US units from nowhere are going as British to frontline.

 

[Gormadoc]

1. Changed ai_revolters.txt events. Now all countries after every liberation will have free resources/free IC/free manpower. All countries will have leaders and ministers when one time liberated. Second need game reloading before liberation. This fix problems with annexed countries and liberated with no leaders.

what did you change, put a persistent = yes on all countries ? "All countries have leaders", does that include China. I was contemplating to try to fix china leaderproblem with creating a dublicate set of leaders with new leader id's and put them as dormant in leaderfile. but hasnt gotten around to that. what did you do ?

2. With fixed money ratio to other resources, human has much less money from trades.

hehe, ye i saw that tread too, your proposed changes sounds reasanoble, but i will probably have to look more into that before i adobt your proposed changes.

4. First will come more aggressive Japan AI. Such aggressive vs vanilla that Africa or Siberia could be their new promised land ..

Carefull with that one it is very easy to go overboard on aggressivenes and end up with the kind of situation like france walking 50% of its forces into southern germany, only to get them killed making rest of germany's france campaign and even bigger cakewalk. Ai agressiveness is fine against AI opponents but when they meet a human it's ussually end in... well its not pretty. Good luck with it though.

BTW. Big stacks of AI units are often becouse "garrison" values. When you set up very low values ex. like in USA files, then AI don't know where to put these units, and often chooses some island. How to fix it ?
US has high value of exp forces to ENG, but they don't want them Why ? ENG have too small "max_front_ratio". Raise this value much higher, and ENG will demand troops from US. At once some big stacks of US units from nowhere are going as British to frontline.

aha thats sound like a good solution. Actually when i come to think about it, its brilliant, I always has been pretty annoyed by the inability of the allies to work together, have you playtested it, did ENG/USA forces make a seriuos D-day ?
What did you specifically change? i gather you made a switch event that raises ENG max_front_ratio when USA enters the allies or when the US reach a specific amount of divisions ?

Keep it up and please share if you get more off those ideas.

Gormadoc

 

[unmerged(53922)]

First, I want to make handsoff playing AI successfull without cheats. Especially Vanilla US/UK is pathetic. So If DDay is not possible, then probably I consecrate true history over game ballancing. Ex. in later year Turkey will have a choose, go to Allies or go to Axis. It will open new front. But to ballance war beetwen Germany vs SOV and Allies, Japan (winning all the time in most games) will probably must DOW SOV. So the game will be much more interesting. With widely opened post war fronts ..... So the invasion problem will be not so big. Game will be going to the end, all the time with action.

Second thing - I made more ballanced human vs AI. As you can see human is using underpowered air and a lot of puppets. This is game killer. Who on the Earth saw the puppets like Scandinavia fighting together with their master as friend ? It's nosense. Master is an enemy, partisans will still act in such country, army will no have will to fight ... In fact puppet goverments was only a cover to easier occupation. So I probably made 2 % dissent/per month to every puppet and I will limit them by 80% garrison production or big IC reduction to real II war puppet simulation. The same - limited airstack effectivness to 4 units only, prevents AI from losing their plains very quickly and it will not be possible to murder AI armies by air only. So no overhelming 16 planes stacks any more.

what did you change, put a persistent = yes on all countries ?

Yes - and I added a local_flag which is automatically removed, when country is annexed. And all nations will have always their free resources with every liberation, what has sense.

My addition to ai_revolters.txt:

event = {
(...)
trigger = { NOT = { local_flag = is_independent } }

persistent = yes

action_a = {
(...)
command = { type = local_setflag which = is_independent }
}
}

"All countries have leaders", does that include China. I was contemplating to try to fix china leaderproblem with creating a dublicate set of leaders with new leader id's and put them as dormant in leaderfile. but hasnt gotten around to that. what did you do ?

I did the same. I added a duplicate leaders with other "id numbers" to .csv files. They all are dormants in .inc scenario files. When country is annexed and liberated by anybody, then dormant at day one are loaded. It works only one time. But you must use an event which sleeps original leaders, becouse somebody could reload in time between annexing and liberatio and country could have 2x the same leaders. I tested it and it works 100% at every condition. (be accurate in ending "enter" in .csv and probably not too long lines in .inc files)

ex. for China:

event = {
id = 150840000
trigger = {
ai = yes
}
random = no
country = CHI

name = "AI_EVENT"
desc = "AI_EVENT"
style = 0

date = { day = 1 month = january year = 1936 }

action_a = {
command = { type = local_setflag which = is_independent }
}
}

event = {
id = 150840001
trigger = {
ai = yes
NOT = { local_flag = is_independent }
}
random = no
country = CHI
persistent = yes

name = "AI_EVENT"
desc = "AI_EVENT"
style = 0

date = { day = 2 month = january year = 1936 }
offset = 1
deathdate = { day = 29 month = december year = 1953 }

action_a = {
# sleep original leaders just in case of game reloading before country liberation
command = { type = sleepleader which = 50000 }
(here all original leaders )
command = { type = local_clrflag which = is_independent }
}
}

hehe, ye i saw that tread too, your proposed changes sounds reasanoble, but i will probably have to look more into that before i adobt your proposed changes.

Ok, In my calulations money is far to cheap. I saw that AI don't buy money at all. (or very very rare) and when trading money to human it generates it from nowhere. Human could collect this cheap - not existent money and do spy missions, when most the AI have no money at all.

Carefull with that one it is very easy to go overboard on aggressivenes and end up with the kind of situation like france walking 50% of its forces into southern germany, only to get them killed making rest of germany's france campaign and even bigger cakewalk. Ai agressiveness is fine against AI opponents but when they meet a human it's ussually end in... well its not pretty. Good luck with it though..

Thanks. But my definiton of agressivness was another. Ex. AI Japan will can't mass their troops on every beach x 10, but on the borders or fronts (when at war). My first move as Japan is moving all troops from islands to Commintern border. Sorry - this time will not be easy DOW on Japan as SOV in february 36. And none of that troops will go again to these islands. For US AI it changes nothig. It's even better when they will place units on these islands. Human could go in one - two steps to Tokyo at once, so these islands has any value in war AI vs Human. It's better to have 4-5 units in every beach in Japan most the time, when facing human US.

Of course when I mod US, Japan furher conqest will be probably not possible, becouse US will pump units to India as exp. forces to ENG. My MOD is thought as what if ? Ex. when Japan successfully conquered India, then they have a choose - or invade Iraq and Saudi Arabia for oil through Persia, or DOW hated SOV communist. I know it's semi historical - but the game is boring, when human can everything, but AI should go only historical way ..... Japan taking in snail speed these useless islands.

aha thats sound like a good solution. Actually when i come to think about it, its brilliant, I always has been pretty annoyed by the inability of the allies to work together, have you playtested it, did ENG/USA forces make a seriuos D-day ?

AI is doing invasion very lazy. From day one (is initializing invasion in invasion file) to landing (is launching) often is passing a 2-4 months ... why ? Transports are free, unit stack is waiting ... I tried enforce them to do it faster by switching invasions on and off. I saw some results, but I am not convinced if it truly works. I have to test the same situation with my event and without many times. What I know today. If the stack is ready, and I turn off invasion in save file, and I reload the game, they invade in day one. (mayby without turning off, reloading is forcing to invade at once - I don't know it now) ... I have some ideas to make better invasions without unit spawning, but I have to test them. For now AI is teoretically using "target" variable as preferred area to invade .. but it's only teoretically. So I add all useless island and not wanted places to invade as target = -1. Then AI is forced to land only in some places. Ex. US will not invade any of the Japanease small islands (I don't want be historical, but effective in game) but ex. only China or Korea. An event will count forces in specific provinces and will add these targets with all other on minus. (hundreads of beaches). And it will set up on some time "area_multiplier =" on these beaches on very high level, to force AI to pump here units". Mayby it will work. I will test it.

What did you specifically change? i gather you made a switch event that raises ENG max_front_ratio when USA enters the allies or when the US reach a specific amount of divisions ?

For now I only testing it without making events. And raising max_front_ratio tho ENG is forcing US, to use most of their non-garrison units (stacked in some random places in high amount) as exp force to ENG fronts. So Italian's have no chance in Africa. Then I will probably dismantle their nosense Africa conquest and they all will go to Barbarossa as Exp. forces. Mayby this time Moscow will fall.

Keep it up and please share if you get more off those ideas.

Thanks for interest in my work. Every, what I made is free for all HOI community

 

[Gormadoc]

First, I want to make handsoff playing AI successfull without cheats...... Ex. in later year Turkey will have a choose, go to Allies or go to Axis. It will open new front...... So the game will be much more interesting. With widely opened post war fronts.....Game will be going to the end, all the time with action.

This is not dissimilar what iam trying to do. I am in process of making start event for all major and some semi major country where the AI will chose a long term strategy. ex. Italy can either choose original vanilla path. or Allies with germany in 1936 for blueprint (What most Human players does, why not the AI). It can get directly involved in SCW on nationalist side. Try to make a complete fascist alternative alliance across the board. SPA, ARG, POR, JAP, etc.

I have also changed triggers for Albania, Anscluss-München event chain so it is more random. Germany could in theory speed up or slow down anneksation of these areas. This means that the human player can never be complete sure when war actually breakout. This should force the Human player to do a more balanced buildup since we can no longer be sure the war start sept. 1939. So disbanding the intire army is not without any risk of defeat.

Second thing - I made more ballanced human vs AI. As you can see human is using underpowered air and a lot of puppets. This is game killer. Who on the Earth saw the puppets like Scandinavia fighting together with their master as friend ? It's nosense. Master is an enemy, partisans will still act in such country, army will no have will to fight ... In fact puppet goverments was only a cover to easier occupation. So I probably made 2 % dissent/per month to every puppet and I will limit them by 80% garrison production or big IC reduction to real II war puppet simulation. The same - limited airstack effectivness to 4 units only, prevents AI from losing their plains very quickly and it will not be possible to murder AI armies by air only. So no overhelming 16 planes stacks any more.

Again i completely agree on both accounts. My approach on the major puppet countries like CHI, SCA, IND, etc. is that i have made AI switch files. That makes them focus on internal affairs. So instead of spawning mass infantry, they will embark on massive economic buildup. Both IC, Bases, an i will also make some infrastructure builds event. This is imo. not far away from what we have seen historically. Defeated countries as Japan, Germany, Italy all embarked on big internal buildup programs after they where defeated. And did't spawn 100 inf. divisions army's the first year. On the military side I make them build mostly Garrisons and militia, with only MP brigades, simulates effort to keep dissent down. Beside af that only a very small mobil force of few inf divisions and HQ. So the Puppet troops should only serve as an interior force to keep controll of country. Btw. I have also made events that will make UK and french colonies revolt if the players remove all or majority of troops from them. Ex. Malaya will defect if there isnt' atleast one British land division in that country, and ENG is not atwar with japan or china. This keeps the Human UK from emtying (or atleast not without some penalty) all overseas colonies and just steamroll AI germany in 1939. havent made any change to airstacks, i will seriously consider your approach it sounds practical.

Ok, In my calulations money is far to cheap. I saw that AI don't buy money at all. (or very very rare) and when trading money to human it generates it from nowhere. Human could collect this cheap - not existent money and do spy missions, when most the AI have no money at all.

I did a bit of research on this, a while back an as far as i can see, the AI will overproduce CG with 10% of available IC if the AI has below 250. This will offcourse only add to the necessity to change the parameters since it further adds to the exploit.

I know it's semi historical - but the game is boring, when human can everything, but AI should go only historical way ..... Japan taking in snail speed these useless islands.

Agreed. I has same approach. If Human can do it, why not the AI. I have however made a start event in my mod that give the human player a choice to sleep all the New AI files and go with forced historical AI Strategies but still include all my AI enhancement on research, spies, production, ministers etc

AI is doing invasion very lazy. From day one (is initializing invasion in invasion file) to landing (is launching) often is passing a 2-4 months ... why ? ....What I know today. If the stack is ready, and I turn off invasion in save file, and I reload the game, they invade in day one. (mayby without turning off, reloading is forcing to invade at once - I don't know it now) ...

It is welknown that save and reload tends to wake the AI up. The prime example of this is the instant cancelling of trade deals when reloading. Now in my oppinion this is one of the biggest flaws in the current game engine. Now having to reload every say 7 days in war time gets kind of boring real fast. I dont really know what to do. But shuffling AI files every now and then could perhaps be a solution. Atleast this is what Lothos did with the AIR AI, which is working reasonoble well.

So Italian's have no chance in Africa. Then I will probably dismantle their nosense Africa conquest and they all will go to Barbarossa as Exp. forces. Mayby this time Moscow will fall.

I am contemplating the same. Italian AI should only go for Egypt/midleeast if they have been allied with either SPA or Germany since 1936.

On the spy issue what i have done is that i have removed all spy parameters from the AI files that gets loaded from start of scenario. When the scenarios start there is a set of events that sets the AI spy parameters according to which country is human controlled and what game setting the player choose.

One thing i have done is setting the influence global opinion to 0 for all axis countries. There is no reason for Germany to try to raise ENG or any other countries belligerence. waste of time and money. The next thing i will test is to lower the chance of succesfuld counter espionage, raise the domestic spy number to 10 in AI files. So it wil be harder for Human to kill of all spies. Hopefully the increased number of domestic spies will offsett that effect for the AI

Gormadoc

 

[unmerged(53922)]

Some screenies from modded JAP AI vs vanilla CHI AI, tested on normal.
China has surrendered after 16 months in October '38.