Welcome to Close Combat Series
  Login or Register Home  ·  Downloads  ·  Forums  ·  Combat Camera  ·  Help  

  Survey
Do incapacitations count as a soldier's kills?

Yes
No



Results
Polls

Votes 1179
Comments: 1

  Shout Box!!

Only registered users can shout. Please login or create an account.

  Main Menu
Articles & News  
    Help
    Player`s News
    Site News
    Multiplayer
    Terrain Challenge
    Boot Camp
Community  
    Forums
    Downloads
    Combat Camera
    MOOXE @ Youtube
    Statistics
Members  
    Private Messages
    Your Account
    Logout

  Donations
Anonymous - $25.00
08/15/2022

Anonymous - $25.00
08/15/2022

Anonymous - $25.00
12/18/2021

Anonymous - $100.00
11/08/2021

Anonymous - $15.00
04/09/2021

Anonymous - $100.00
04/05/2021

Anonymous - $20.00
02/20/2021

Anonymous - $10.00
12/29/2020

Anonymous - $1.00
11/06/2020

ZAPPI4 - $20.00
10/10/2020

Find our site useful? Make a small donation to show your support.



Search for at
Close Combat Series Advanced Search


 Author
Message
 
FreeKing

Rep: 26.9
votes: 4


PostPosted: Mon Jan 31, 2022 8:05 pm Post subject: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

A complaint, and rightly so, of the Close Combat series, especially the re-releases, was the single player AI. I had managed to push the AI as far as it could go from a modding perspective when I released my small maps mods. It wasn't quite where I knew it could be, and here's why:

Years ago, a post was made on the old CSO forums. Somebody had found a windows registry entry that improved the AI. For reference, here's a post on the CCS forums that helped spread this registry edit around:

http://www.closecombatseries.net/CCS/modules.php?name=Forums&file=viewtopic&t=6444&postdays=0&postorder=asc&highlight=registry&start=0
 
The veterans of this site will remember it. Like me, they probably made the same changes to their own games after seeing it. To quote Dauphin's observations, which had been my experience, as well:

To make it short
here are the points that I have noticed using the 003 parameter for the AI
- use of smoke
- use of off board artillery
- use of airplanes
- the use of mortar is effective against an offensive
- guns are set in better places
- the ennemy tends to do better traps
- less or much fewer crawling units
- better use of grenades

Note: the AI was much better and more active, in comparison to previous performance, not genius level AI or anything. But, it was a tougher opponent and single player was much more enjoyable. My own experience was that it felt like I was actually in a fight, instead of just stomping the AI.

To get to the point, Gateway to Caen and Panthers in the Fog have the same registry entries to be able to do the same thing. Here are the registry entry locations.

NOTE: I have the Matrix Games editions, not steam or GOG. I don't know if these will be present for either of those.

PitF: Computer\HKEY_CURRENT_USER\SOFTWARE\Matrix Games\Close Combat Panthers in the Fog

GtC: Computer\HKEY_CURRENT_USER\SOFTWARE\Matrix Games\Close Combat Gateway to Caen

Change the DIFFLEVEL parameter from 01 00 00 00 to 03 00 00 00, and you'll have the same positive effects I've mentioned above.

Now, to get to the re-releases of the classics. I had thought the registry entries had been removed, and the above was no longer possible for them. Well, as it turns out, it still is.

Rather than having the settings in the windows registry, where they were previously, the settings, including DIFFLEVEL, are now in a text document in the Documents folder (My Documents on versions of Windows prior to Windows 10). I own the entirety of the re-release series (and Modern Tactics), the Matrix/Slitherine editions, and this is where you'll find the DIFFLEVEL settings for Cross of Iron, Last Stand Arnhem, The Longest Day, Wacht am Rhein and Modern Tactics:

Documents\My Games\CloseCombatCrossOfIron
Documents\My Games\CloseCombatLastStandArnhem
Documents\My Games\CloseCombatModernTactics
Documents\My Games\CloseCombatTheLongestDay
Documents\My Games\CloseCombatWachtAmRhein

In each of those folders, you'll find a file called "OPTIONS". Open it up with notepad, find the DiffLevel row, and change it from 1 to 3. All of them, with the exception of Modern Tactics, were set to DiffLevel=1. This explains part of the poor AI performance, which seemed to get worse as the series was updated and re-released. That's because the default setting was always 1, and if a player didn't know to look for it, the player would go on thinking the AI was made worse. No, it wasn't, the setting was just set to a lower level on ALL of the re-release games, as well as GTC and PITF.

Here's an example from my The Longest Day options text file. I've bolded the field you need to change, and it is shown with the new change.

ShowDSAlert=TRUE
ShowQHelp=TRUE
PlaySounds=TRUE
PlayMusic=FALSE
PlayVideos=FALSE
GameSpeed=0
Scroll=0
OutlineType=9
BarType=7
ScreenX=1920
ScreenY=1080
Save=FALSE
ShowTrees=TRUE
DrawKIA=TRUE
Expanded=FALSE
StaticFPS=10
VoiceLang=TRUE
InsetMap=TRUE
InsetX=2000
InsetY=2000
IType=2
PlayerName=Commander
PlayerSide=0
EndDeath=FALSE
EndTimer=TRUE
TimerVal=18000
EndSDeath=TRUE
EndSDWarn=TRUE
DiffLevel=3 (Original value: DiffLevel=1)
StrengthP=2
StrengthO=2
ObeyP=FALSE
ObeyO=FALSE
FearP=FALSE
FearO=FALSE
SeeEnemyP=FALSE
SeeEnemyO=FALSE
FullInfoP=FALSE
FullInfoO=FALSE
InitiativeP=FALSE
InitiativeO=FALSE
SoldierMon=TRUE
EndBatMor=TRUE
EndCampMor=TRUE
NetLobbyConnectType=6
SoundVolume=0
Windowed=TRUE
VoiceChat=FALSE
Framed=FALSE
ScaleUI=FALSE
ScaleFrontEnd=FALSE
UseDPI=FALSE
LockMouse=FALSE

Without any other modding changes, I wanted to confirm these DIFFLEVEL settings still worked. So, I ran a test at DiffLevel=1, with no mods turned on. It's what you would expect it to be - inactive and boring. Then I ran a test with DiffLevel=3, and boom. The AI was more active, just like it was back in the day when these changes were made in the registry entry. Then, I ran a test with some modding done in the bgroups.txt, and it was even better than the DIFFLEVEL change alone.

To the veterans of the series, who know what I'm getting at. It still works, you just need to make the change in a text document instead of the registry.

So, to recap:

For Panthers in the Fog and Gateway to Caen, change DIFFLEVEL with regedit.

For the re-releases, change it in the OPTIONS.txt file with notepad.

Add additional modding with bgroups to push it even further. I assume using a small maps mod would push it even further than that.

I hope this helps everybody out. I was excited to see that these changes could still be done, because I missed being able to do it. It added so much to the experience. Now that I know I can still do it, it's brought some more life back into the series. I'll post some follow-up in here later on when I've had the chance to compare large maps to small maps with the DiffLevel=3 change.

Oh, and could a mod pin this post? Or move it somewhere more applicable to the content? This is the best place I could think of, and the changes above are important for a better gameplay experience. I'm sure alot of players would benefit from seeing it.
Back to top
View user's profile Send private message
 
Tejszd

Rep: 133.6
votes: 19


PostPosted: Tue Feb 01, 2022 5:45 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Thank you for posting this!
Back to top
View user's profile Send private message
 
Aetius

Rep: 89.2
votes: 4


PostPosted: Tue Feb 01, 2022 11:33 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Very interesting

Thanks for posting
Back to top
View user's profile Send private message
 
FreeKing

Rep: 26.9
votes: 4


PostPosted: Tue Feb 01, 2022 7:29 pm Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

My first tests were me just shoving my units off to the side of the map and turning on "always see the enemy", so I could observe the AI's movements and actions. These observation tests were done without small maps installed, and no mods except for changes to BGroups (AI reserves in the BGroup set to -5, the other three set to 5). I did a test on the stock large map, then a test on a small map version of the same map from the large map test. Here were my observations, a test in each generation of the game engine:

GATEWAY TO CAEN
Mods - BGroups changes to AI strategy, blood mod installed so I knew modded data was running and my changes to BGroups were present (I put all the mood stuff into one mod folder and used the /D switch in the shortcut's launch options). Small map mod (only the maps, no terrain element changes) with BGroups changes for small map test.
Map - I did the test on St-Mauvieu.

Settings:
-Both sides Recruit unit strength
-Always see enemy turned on for me, so I can observe the AI
-Timer of 20 minutes

AI Behavior Observations, St-Mauvieu, Assault St-Mauvieu Scenario:
    -This is the cool part! The AI actually did things on the large map. It pushed out with infantry and halftracks about 1:30 in.
    -The AI established positions in buildings, and along a wall, and in a hedgrerow, rather than just sitting in its starting positions at the beginning of the battle. There's actual activity going on, with an apparent purpose to the AI's orders, instead of troops just moving around. The buildings it occupied had a look out onto open ground, and a halftrack was placed south, also looking out into open ground, though it was exposed in the open, as well.
    -An infantry squad was pushed forward at a run, and I thought it was sent to capture the Distillerie. Nope. I had placed all of my units on the edge of the map, set to Ambush, and that squad ran down the road, past the distillerie, and right toward the edge of the map. I moved a couple squads to a hedge to see what would happen when I started shooting at AI's troops.
    -When my boys opened up on them, halftracks got moving in the rear. There was some movement with infantry squads, and the AI started their mortars on zeroing on my two squads.
    -The thing is, too, is that the AI didn't waste all of its mortar rounds like it usually does. After a short fire mission, the mortar fire stopped.
    -My attention had been on the above, so when I looked back to see what the rest of the AI troops were doing, I noticed that halftracks had been moved onto the road and were waiting. Infantry had been moved into buildings farther in the rear on victory locations, and infantry were starting to move forward onto the map.

Overall observations are that the AI actually plays on the large maps, now. For the guys that like playing on large maps, the DIFFLEVEL change makes it viable again. The AI had set up basic positions, sent out a recon to locate my troops, then reacted by setting in defense on its VLs and sending out troops. There was going to be an actual firefight going on. I won't lie to you guys - I started getting excited. So, I ended the test and installed the small maps to test again in the same conditions.

AI Behavior Observations, St-Mauvieu (Small Map), Same Scenario:

Same deal. This was purely a test to observe the AI. I stuffed all of my units into a space surrounded by hedgerows and set them to ambush. Because I was so close to the AI, due to the small map, I ordered all of my troops, except for the Churchills, to move fast to the bottom of the map.
    -Everything got kicked off so suddenly, and I had to give orders to get my troops out of the line of fire of enemy troops (the AI caught them on the move and opened up with an MG42 and mortars). Because of that, and the events unfolding differently than the first test, I scrapped the test, because the AI was reacting to different events, rather than me observing the AI, and doing an action and seeing what the AI did in response. That said, I decided to save the small map tests for the combat tests later, when I actually fought the firefight in an actual gameplay setting, instead of just observation.


THE LONGEST DAY
Mods - BGroups changes to AI strategy, GtC explosion graphics mod installed so I knew modded data was running and my changes to BGroups were present.
Map - I did the test on Bretteville. It had equal number of troops, mix of multiple unit types. It's also one of the largest new maps I could find, as most of the old maps transferred over from CC5 to TLD are the original size. The point of the test is to see what the AI would do on a large map. I already know what it does on smaller maps when I did the tests for the small maps last year.

I decided to go with The Longest Day for the large map test, mainly due to the amount of troops in the scenarios. Because I'm testing on single scenarios, it was tough to find a scenario that had a full roster of troops from the list (lots of Ostruppen and smaller engagements in Last Stand Arnhem, until you get to the scenarios featuring the Panzer divisions). Anyway, same kind of observation test, same game settings as the Gateway to Caen tests. 15 minute time limit this time, though, because TLD doesn't have a 20 minute option.

AI Behavior Observations, Bretteville, June 8 Stock Scenario:
    -The more I observed the AI in these tests, the more interested I got.
    -In this case, it was much the same as what happened in Gateway to Caen, but the AI was more methodical.
    -It was a slow start. The AI moved its tanks, and a halftrack, to positions on the road. Some infantry crawled a short way to positions in the open ground, in crops.
    -It took a minute or two, but when the AI got started I saw interesting tactical behaviors. The AI sent out infantry squads and a halftrack forward to scout and recon the area. How do I know this, you might ask? Because after the AI's recon units figured out there wasn't anything there, the rest of the units go moving. The Panthers starting moving into the area, and the rest of the infantry started forward.
    -A halftrack and an infantry squad occupied positions on the extreme flanks, like an OP or overwatch, and the rest of the units occupied the area around the rail station, which they captured with an infantry squad early on.
    -By this point it's about the 11 minute mark, and a lull has set in, with the AI static. I tried giving some orders a time or two throughout the test when the AI was static in the beginning, and when it went static at this point, and I can confirm that the game's AI does not sense if the player is giving orders or doing anything, because this didn't spur the AI into action. I used my airstrike to kill one of the AI's recon squads, and ordered my off-map mortar barrage onto some random open spot. Neither spurred the AI to do anything, this occurring with about 3 minutes left in the fight.
    -I wanted to make sure it wasn't just because the game time was about to run out, and the AI programmed to go defensive and hold on to what it has before a game ends, so I started another test in the same conditions, with no time limit.

AI Behavior Observations, Bretteville, June 8 Stock Scenario, No Time Limit:
    -It was a similar progression and deployment as the last test, but the AI was definitely more aggressive.
    -It was quite active. The lull in the beginning was about the same, but once the AI got moving it worked on expanding its control of victory locations on the map.
    -It didn't move into immediately deploying it's entire force, but parts of it. It captured the rail station early, and deployed its Panthers to the vicinity of it. An infantry squad was dispatched to scout the town of la Villanueve, and a halftrack was put on the road to move to the same area. The halftrack was used to capture the VL there while the infantry scouted the buildings near the VL.
    -The AI continued to peak my interest. After it took the la Villaneuve VL, the halftrack held the road while the infantry moved farther to the right to scout the hedgerows outside of town.
    -Infantry were also moving in from the AI's start position by 8:30 into the game, and an infantry squad had been placed this whole time a little bit south of the rail station, I assume as recon or overwatch.
    -Until about the 14 minute mark, the AI was moving troops around, bringing them up from their start positions. I'm not including the details of their movements at this point, because I'm trying to move this along, and the point of getting rid of the time limit was to see if it would push forward and not go static.
    -At 15 minutes into the game the AI had brought infantry into the open ground outside of la Villaneuve, and one of its Panthers were starting to push south of town.
    -NOTE: at no point did the AI just run units around the map. It moved them from covered location to covered location. Even the squads in the open were moved up to hay bales and such. When the infantry, and the second Panther started pushing south, the infantry were moved along the hedgerows outside of la Villaneuve.
    -At 16 minutes, the AI had placed one Panther in reserve. One infantry squad and a Panther were ordered south.
    -I ended the test at this point, because I had my answer.
    -CONFIRMED: The time limit of the game matters. If the game is going to end soon, the AI will go defensive and hold on to what it has already, instead of getting aggressive. If it has the time, it will continue across the map, and it won't do it haphazardly. From what I'm seeing, the AI in both Gateway to Caen and The Longest Day is actually quite "intelligent", meaning it will give the player the challenge, make the player feel like they're actually in a battle. I don't know whether or not the AI will pose a challenge to me yet in an actual gameplay setting, where I'm actively trying to win a fight. However, a player that doesn't use proper tactics, and who just kind of runs units all over the map, is going to get their ass kicked.

Okay! I'm VERY happy with those results. It's an active AI, that gives orders that make logical tactical sense. Now it's time to fight this AI for real. On to the actual gameplay tests.

I'll post a follow-up reply here when I play later today. But, I'm very happy with what I'm seeing.
Back to top
View user's profile Send private message
 
Pete

Rep: 118.1
votes: 12


PostPosted: Wed Feb 02, 2022 9:58 pm Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Thanks Freeking for your research and observations. I have used the registry modification myself but never bothered to save the instructions on how to do it. Thanks for the reminder.
Have you found out what difference the various battle group aggression / caution settings make?


Dulce Bellum Inexpertis
(War is delightful to those who have no experience of it.)
Our facebook page: https://www.facebook.com/buckandpete/
Our website: https://themodsection.net/
Back to top
View user's profile Send private message
 
FreeKing

Rep: 26.9
votes: 4


PostPosted: Fri Feb 04, 2022 5:08 pm Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Pete wrote (View Post):
Thanks Freeking for your research and observations. I have used the registry modification myself but never bothered to save the instructions on how to do it. Thanks for the reminder.
Have you found out what difference the various battle group aggression / caution settings make?


That's a good question. I'd assume there are differences, though it warrants testing. Instead of having walls of text, I figured I'd just do a before/after video of the DIFFLEVEL change. Basically, a video that details my findings and observations. I'll play around with the BGroups and see what happens.

EDIT: Had a chance to test the settings. I changed the settings on an AI bgroup, setting everything to the lowest settings, except for Charisma (morale), which I set to the highest setting. I tested the highest and lowest settings, each one individually. I had the DIFFLEVEL set to 3. For those that don't know what these settings are, the BGroups files has four settings for a battlegroup in the game: Aggression, Caution, Charisma, and Tactics.

Pete, I went in and did a few tests. Here's what I observed:

-I noticed that the higher the aggression setting, the more prone the AI was to sending units out alone and unsupported, which leads me to believe this setting manages how reckless the tactical AI is. This also explains why the higher the setting, the more quickly the AI pushed out onto the map. With the DIFFLEVEL change to 3, it will still move out onto the map, but will do it more or less recklessly depending on the setting. Unsupported units are sent out to far flung positions, away from the rest of its force, for example. So, speed of pushing onto the map, for the AI's aggression setting, is a measurement of how recklessly it does so, not how active it is in doing so.
-Caution looks like it manages how many units the AI uses for its orders. Basically, if the Caution setting is set to make it the most cautious setting the game allows, the fewer units it will use and the more units it keeps static in case it needs them. Think of an overly cautious commander in real life, that doesn't do much or holds back a large portion of their force, and it would accurately describe what the most cautious AI setting will do. The least cautious AI setting is the opposite, more units being used for its programmed tasks.
-I'm not sure what the Charisma setting does to the tactical battle AI. I didn't notice a change in the overall morale of the units in the unit selection screen, so I can only assume that its got something to do with how well the troops endure enemy fire and suppression and such. I'll need to come back to this one.
-The tactics setting regulates how intelligent the AI is in its tactical decisions. An example would be picking a route with more cover to move troops, instead of running them across the open all the time. Another example is sending a vehicle with infantry support, instead of the vehicle being sent all by itself. I also noticed the highest tactics setting had the AI moving troops into covered positions like hedges and buildings, rather than laying out in the open. I noticed vehicles being placed int the open if infantry were going to move out into open ground, so the tank can cover them. That sort of thing. This setting determines whether or not the AI uses better or worse tactics. There's a more combined arms warfare behavior going on, instead of just moving units around the map.

With the above in mind, here is how I would describe each setting in the BGroups:
-Aggression: How reckless it is.
-Caution: How many units are being used and how many are being held back in static positions.
-Charisma: Not sure yet. Probably how cowardly or brave the individual soldier AI is.
-Tactics: How smart it is in its tactical decisions. See above observations.

With all of this in mind, I went in and played with settings combinations until I found the one that best creates a competent tactical commander. Not reckless, but not indecisive or overly cautious. Obviously, I set the tactics setting all the way up. For the time being, I've set the charisma setting to the middle value, 0 (the settings are a range of -5 to 5, 0 being the middle), because I don't quite understand completely what it does yet. However, here are my current settings for the BGroups:

-Aggression: 0 (Not too reckless, not so careful that the AI is timid and indecisive)
-Caution: -5 (AI uses most of its units, doesn't keep them in reserve, make the AI act more dynamically and with most of the force assigned to it)
-Charisma: 0 (No positive or negative modification for the morale of units)
-Tactics: 5 (Makes logical tactical decisions, is smarter)

Bear in mind that this whole write-up discounts the whole point of the BGroups data, which is to create "personalities" for each commander of the battlegroups. I made the above settings the same for every single battlegroup in the game, because I wanted the same kind of AI "personality" in each tactical battle. The BGroups changes are necessarily required for an enjoyable single player experience, thanks to the DIFFLEVEL change to 3, because the AI will still play on the large maps and small maps, alike. So, if a player wanted to keep the battleground commanders at their stock settings, some commanders being more reckless than others, for example, there wouldn't be a reason not to. It comes down to personal preference.

With the above, I've also given players the info they'll need to tweak their BGroups settings to whatever they want, for the kind of AI personality they prefer fighting.
Back to top
View user's profile Send private message
 
Pete

Rep: 118.1
votes: 12


PostPosted: Sun Feb 06, 2022 9:53 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

-Charisma: in one of the workbooks this is described as "Commander's Charisma.  -5 takes morale away from troops, +5 gives morale to troops.  Range = -5 to 5" .

I find your observations very useful. For the Dieppe mod that I am currently working on I need extremely defensive , "static" , German battle groups as well as aggressive battle groups and this info has helped me understand better what settings to use.


Dulce Bellum Inexpertis
(War is delightful to those who have no experience of it.)
Our facebook page: https://www.facebook.com/buckandpete/
Our website: https://themodsection.net/
Back to top
View user's profile Send private message
 
FreeKing

Rep: 26.9
votes: 4


PostPosted: Sun Feb 06, 2022 7:57 pm Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Pete wrote (View Post):
-Charisma: in one of the workbooks this is described as "Commander's Charisma.  -5 takes morale away from troops, +5 gives morale to troops.  Range = -5 to 5" .

I find your observations very useful. For the Dieppe mod that I am currently working on I need extremely defensive , "static" , German battle groups as well as aggressive battle groups and this info has helped me understand better what settings to use.


Excellent, glad I could help.
Back to top
View user's profile Send private message
 
mad_matx

Rep: 0.1


PostPosted: Mon Feb 07, 2022 6:30 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Hmm.  I'm playing WaR, changed the DiffLevel to three as per the article at about Day 3, and really the only things I noticed are that Shermans and Bazookas are knocking out my Panthers with ease, and on top of that, my Panthers are turning broadside and ass-end to the enemy as they move.
Oh.  And my Panthers are getting immobilized at the slightest bump in the road.  
With all that, I haven't had the chance to see if there are any changes to the enemy AI.

Is this just what stock WaR 4.50.33 is like, regardless of difficult settings, and I just noticed it after the changes?  Comments please.
Back to top
View user's profile Send private message
 
FreeKing

Rep: 26.9
votes: 4


PostPosted: Tue Feb 08, 2022 12:50 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

mad_matx wrote (View Post):
Hmm.  I'm playing WaR, changed the DiffLevel to three as per the article at about Day 3, and really the only things I noticed are that Shermans and Bazookas are knocking out my Panthers with ease, and on top of that, my Panthers are turning broadside and ass-end to the enemy as they move.
Oh.  And my Panthers are getting immobilized at the slightest bump in the road.  
With all that, I haven't had the chance to see if there are any changes to the enemy AI.

Is this just what stock WaR 4.50.33 is like, regardless of difficult settings, and I just noticed it after the changes?  Comments please.


I haven't noticed anything like that, in particular. If I did, I probably ignored it as the stock game's issues on pathfinding.

Also, guys, see below link to the Matrix forums. The DIFFLEVEL was brought up there by Tejszd:

https://www.matrixgames.com/forums/tm.asp?m=5139124

Steve over at Matrix said that the Options.txt doesn't do what I've described. I did some more testing after reading it, and sometimes there doesn't seem to be a difference between DIFFLEVEL=1 and DIFFLEVEL=3. The AI is active in either case, and it definitely was a definite difference, from what I remember when the DIFFLEVEL was changed in the registry, though I don't know if that's still the case, anymore.

For the sake of honesty, read the above post. I can't say for sure anymore if DIFFLEVEL does anything or not, but it seems that way to me, or maybe the AI has been improved since the last time I studied the AI behavior. I leave it your own judgement. I'm going to leave my DIFFLEVEL at 3, because the DIFFLEVEL is set to 3 in Modern Tactics, and it doesn't make sense to me that it'd be 3 in Modern Tactics, but 1 in the others (If it's a legacy setting that has no effect, and only modifies both sides with a simpler and easier setting, why not match the setting for the rest of the games in the series being as you can't modify the strength of either side in Modern Tactics. Doesn't make sense to me.)

Either way - follow your own judgement, folks. If you think the DIFFLEVEL improves the AI, use it. If you don't, keep it at 1. Like I said, I'm gonna leave mine at 3, even if I can't concretely verify it's had an effect.

BGroups effects I've described are correct, though, and this was verified when I did my small maps mods, so by all means tweak the BGroups to your liking.

I hope setting the DIFFLEVEL to 3 works, but like I said, it might not be having any effect at all, other than modifiers to player and AI units, as was described in the linked thread. I'm a little disappointed that it might not be having an effect, but it is what it is, I suppose.
Back to top
View user's profile Send private message
 
mooxe

Rep: 221.1
votes: 25


PostPosted: Sat Feb 12, 2022 8:49 pm Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

Difficulty level as I understood it only changes the amount of teams in a battlegroup.


Join Discord for technical support and online games.
Back to top
View user's profile Send private message Visit poster's website
 
Hasky

Rep: 2.4


PostPosted: Fri Feb 18, 2022 5:40 am Post subject: Re: Better, More Active AI the Re-Releases - DIFFLEVEL Reply with quote

From observations of AI. At DiffLevel=3, AI controlled units behave more intelligently, attacking not along the entire "front", but concentrating on a certain point of the player's defense.
I'm currently playing the GJS 9.5 mod.
IMHO))
Back to top
View user's profile Send private message
 
 
Post new topicReply to topic printer-friendly view Close Combat Series Forum Index -> The Mess -> Tech Support


 
   
 


Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum




Forums ©





In August of 2004, Zappi, Homba, Bambam887, RedScorpion and MOOXE all pitched
in to create this Close Combat site. I would to thank all the people who have visited and
found this site to thier liking. I hope you had time to check out some of the great Close Combat
mods and our forums. I'd also like to thank all the members of our volunteer staff that have
helped over the years, and all our users that contributed to this site!