FC. Realism vs. Arcade UTD.

[Robalex]

This is only about exhibition mode CPU vs CPU games so be advise in order not to waist your time reading.

This thread is created because I couldn't find a solution for FIFA 12 neither FIFA 13. I'm looking for help to create a mod to solve below issue.
AndreaPirlo told me he will help on this when he finish to upgrade his latest FIFA12 GPMod which also works on FIFA13, his GP mode is superb!.

In the meantime I wish I could receive some comments/help from masters who understands the game better than anyone (regularcat for example, or Fidel...)

I'm sure most of you never watched or managed a game without using a joystick, probably never pay attention how a CPU team controlled behave against other CPU controlled team (exhibition mode of course, it's the only way to see it)

The only concern is: When a CPU strong team faces a much weaker team, what happens is that strong team will create plenty chances and shootings but "mysteriously" will score 1 or 2 goals. The poor facing team will create only one or two chances or shootings and will score 1 or 2 goals. It happens most of the games.

In real football the team which creates more goal chances than the rival will generally win, not always, but most of the time.

Question is: Is there any possibility to change or adjust this side of "arcade version" of the AI in order to make the game more realistic?

Thanks in advance to all who can provide any help

br

robalex

 

[regularcat]

i could help you if you posted in the correct forum.

this is for mods, as in patches for a game, not questions about mods.

moved to fifa 13 forum.

 

[bangus]

Adjust team tactics, that fixes so many issues with this game. Set weaker teams to high defense, low attack, slow buildup, few shots, etc. Set stronger teams to high attack and lots of shots. Example: the shooting rating goes up to 99, but the highest any team is rated in that catagory is 80, and that's just two teams. The top 2-3 dozen teams should be rated in the 90s in shooting. The default team tactics are incredibly vanilla, EA did nothing with them. But tactics can fix so many of the game issues people here are continually complaining about. It just takes a bit of editing time and effort and you'll be amazed at how different and better the game plays.

 

[Robalex]

Adjust team tactics, that fixes so many issues with this game. Set weaker teams to high defense, low attack, slow buildup, few shots, etc. Set stronger teams to high attack and lots of shots. Example: the shooting rating goes up to 99, but the highest any team is rated in that catagory is 80, and that's just two teams. The top 2-3 dozen teams should be rated in the 90s in shooting. The default team tactics are incredibly vanilla, EA did nothing with them. But tactics can fix so many of the game issues people here are continually complaining about. It just takes a bit of editing time and effort and you'll be amazed at how different and better the game plays.

Hi Bangus, I've already try adjusting each team tactics, it doesn't work. Its something beyond that. Gameply is realistic, amount of shooting is realistic. But what is not realistic is when strong team with good players shoot dozen times and score once and poor team with mediocre players shoot one or two times and they score. See the point?, that's not real, it could happens, but not most of the games.
EA try to keep you anxious when playing, that's why help bad teams make them score and strong team shooting wide, hit the post, poor team super keeper, etc. I tried with tactis, sliders, and CPU still "cheating"

 

[Robalex]

i could help you if you posted in the correct forum.

this is for mods, as in patches for a game, not questions about mods.

moved to fifa 13 forum.

Thanks Regularcat, do you know how to fix what I've posted?

 

[bangus]

Not to argue, but it's hard to believe there's a cheat code in the game that makes weaker teams play better against stronger teams when playing CPU-CPU. It makes no sense that they would bother to even put a cheat code like that in the game.

 

[bangus]

Are you playing career mode? Because career mode is full of cheats like that. Plus aren't there player form ratings in career mode? That would account for some of it. Anyway, there are career mode cheat codes in the product file that make teams play better against other teams, depending on the situation. Take a look at those.

 

[bangus]

From Fidel's product file:

// Career Mode difficulty tuning
[CAREER_MODE]

// negative means that the opposition is better than us
// positive means that the opposition is worse than us
OVERALL_RATING_DIFF_0 = -25
OVERALL_RATING_DIFF_1 = -20
OVERALL_RATING_DIFF_2 = -15
OVERALL_RATING_DIFF_3 = -10
OVERALL_RATING_DIFF_4 = -7
OVERALL_RATING_DIFF_5 = -5
OVERALL_RATING_DIFF_6 = -4
OVERALL_RATING_DIFF_7 = -3
OVERALL_RATING_DIFF_8 = -2
OVERALL_RATING_DIFF_9 = -1
OVERALL_RATING_DIFF_10 = 0
OVERALL_RATING_DIFF_11 = 5
OVERALL_RATING_DIFF_12 = 10
OVERALL_RATING_DIFF_13 = 15
OVERALL_RATING_DIFF_14 = 20
OVERALL_RATING_DIFF_15 = 25
OVERALL_RATING_DIFF_16 = 30
OVERALL_RATING_DIFF_17 = 35
OVERALL_RATING_DIFF_18 = 40
OVERALL_RATING_DIFF_19 = 45
OVERALL_RATING_DIFF_20 = 50

// 100 represents one full difficulty level
// negative reduces the difficulty
// positive increase the difficulty
DIFFICULTY_MODIFIER_0 = 100
DIFFICULTY_MODIFIER_1 = 100
DIFFICULTY_MODIFIER_2 = 100
DIFFICULTY_MODIFIER_3 = 75
DIFFICULTY_MODIFIER_4 = 60
DIFFICULTY_MODIFIER_5 = 50
DIFFICULTY_MODIFIER_6 = 40
DIFFICULTY_MODIFIER_7 = 30
DIFFICULTY_MODIFIER_8 = 20
DIFFICULTY_MODIFIER_9 = 10
DIFFICULTY_MODIFIER_10 = 0
DIFFICULTY_MODIFIER_11 = 0
DIFFICULTY_MODIFIER_12 = 0
DIFFICULTY_MODIFIER_13 = -10
DIFFICULTY_MODIFIER_14 = -20
DIFFICULTY_MODIFIER_15 = -40
DIFFICULTY_MODIFIER_16 = -60
DIFFICULTY_MODIFIER_17 = -80
DIFFICULTY_MODIFIER_18 = -100
DIFFICULTY_MODIFIER_19 = -100
DIFFICULTY_MODIFIER_20 = -100

// we want to simulate that player a team that's better than you
// that is rated in the D+ 80s is harder then if they were in the 50s
// so we scale the user rating between MIN_VAL and MAX_VAL
// and apply the the modifier to the difficulty modifier above
// so if I'm rated 85 and I play someone better than me, I multiply
// the difficulty modifier by 3 - if I was rated 80 it would be 2.3333
OVERALL_RATING_DIFF_MIN_VAL = 70
OVERALL_RATING_DIFF_MAX_VAL = 85
OVERALL_RATING_DIFF_MIN_MOD = 1.0
OVERALL_RATING_DIFF_MAX_MOD = 3.0

 

[Robalex]

From Fidel's product file:

// Career Mode difficulty tuning
[CAREER_MODE]

// negative means that the opposition is better than us
// positive means that the opposition is worse than us
OVERALL_RATING_DIFF_0 = -25
OVERALL_RATING_DIFF_1 = -20
OVERALL_RATING_DIFF_2 = -15
OVERALL_RATING_DIFF_3 = -10
OVERALL_RATING_DIFF_4 = -7
OVERALL_RATING_DIFF_5 = -5
OVERALL_RATING_DIFF_6 = -4
OVERALL_RATING_DIFF_7 = -3
OVERALL_RATING_DIFF_8 = -2
OVERALL_RATING_DIFF_9 = -1
OVERALL_RATING_DIFF_10 = 0
OVERALL_RATING_DIFF_11 = 5
OVERALL_RATING_DIFF_12 = 10
OVERALL_RATING_DIFF_13 = 15
OVERALL_RATING_DIFF_14 = 20
OVERALL_RATING_DIFF_15 = 25
OVERALL_RATING_DIFF_16 = 30
OVERALL_RATING_DIFF_17 = 35
OVERALL_RATING_DIFF_18 = 40
OVERALL_RATING_DIFF_19 = 45
OVERALL_RATING_DIFF_20 = 50

// 100 represents one full difficulty level
// negative reduces the difficulty
// positive increase the difficulty
DIFFICULTY_MODIFIER_0 = 100
DIFFICULTY_MODIFIER_1 = 100
DIFFICULTY_MODIFIER_2 = 100
DIFFICULTY_MODIFIER_3 = 75
DIFFICULTY_MODIFIER_4 = 60
DIFFICULTY_MODIFIER_5 = 50
DIFFICULTY_MODIFIER_6 = 40
DIFFICULTY_MODIFIER_7 = 30
DIFFICULTY_MODIFIER_8 = 20
DIFFICULTY_MODIFIER_9 = 10
DIFFICULTY_MODIFIER_10 = 0
DIFFICULTY_MODIFIER_11 = 0
DIFFICULTY_MODIFIER_12 = 0
DIFFICULTY_MODIFIER_13 = -10
DIFFICULTY_MODIFIER_14 = -20
DIFFICULTY_MODIFIER_15 = -40
DIFFICULTY_MODIFIER_16 = -60
DIFFICULTY_MODIFIER_17 = -80
DIFFICULTY_MODIFIER_18 = -100
DIFFICULTY_MODIFIER_19 = -100
DIFFICULTY_MODIFIER_20 = -100

// we want to simulate that player a team that's better than you
// that is rated in the D+ 80s is harder then if they were in the 50s
// so we scale the user rating between MIN_VAL and MAX_VAL
// and apply the the modifier to the difficulty modifier above
// so if I'm rated 85 and I play someone better than me, I multiply
// the difficulty modifier by 3 - if I was rated 80 it would be 2.3333
OVERALL_RATING_DIFF_MIN_VAL = 70
OVERALL_RATING_DIFF_MAX_VAL = 85
OVERALL_RATING_DIFF_MIN_MOD = 1.0
OVERALL_RATING_DIFF_MAX_MOD = 3.0

Exactly, I do not know what would happen if I use those codes and change [CAREER_MODE] into FRIENDLY... the problem is I do not know if that "friendly" line even exist in .exe file

 

[bangus]

If you're playing career mode, I would just change all those values to 0.

 

[regularcat]

yea thats career mode, but i think he is asking about match day/kick off, which is a good question.

ill have to take a look when i get my hardware in & i can actually test some things.

 

[Robalex]

yea thats career mode, but i think he is asking about match day/kick off, which is a good question.

ill have to take a look when i get my hardware in & i can actually test some things.

Thanks Master, hope when you have your super machine up & running you'll have the time to go back to this post

 

[AndreaPirlo21]

Thanks Master, hope when you have your super machine up & running you'll have the time to go back to this post

I'll look right now.

 

[Robalex]

I'll look right now.

Thanks my friend, hope you find codes to edit kick off -> exhibition match

Remember that the problem is only when playing top teams against poor teams.

good luck!

 

[bangus]

Remember that the problem is only when playing top teams against poor teams.

I played two FIFA 12 CPU-CPU games, Liverpool (team rating 78) hosting Swansea (team rating 65), then Swansea hosting Liverpool. 10 minute halfs, Legendary difficulty. I use custom tactics: I've set up Swansea as a defensive, ball-possession team and Liverpool as an attacking team with a quicker buildup.

I can't say I saw any cheating. The games and the results played out about how one would expect between those two teams. One thing I did notice: when playing CPU-CPU, the players and player animations move more quickly than they do in User-CPU games. So something is definitely a bit different there.

Game 1:
Liverpool - Swansea
1-0 score
9-4 shots
2-1 on target
45%-55% possession

Game 2:
Liverpool - Swansea
0-0 score
11-9 shots
1-4 on target
43%-57% possession

 

[AndreaPirlo21]

I played two FIFA 12 CPU-CPU games, Liverpool (team rating 78) hosting Swansea (team rating 65), then Swansea hosting Liverpool. 10 minute halfs, Legendary difficulty. I use custom tactics: I've set up Swansea as a defensive, ball-possession team and Liverpool as an attacking team with a quicker buildup.

I can't say I saw any cheating. The games and the results played out about how one would expect between those two teams. One thing I did notice: when playing CPU-CPU, the players and player animations move more quickly than they do in User-CPU games. So something is definitely a bit different there.

Game 1:
Liverpool - Swansea
1-0 score
9-4 shots
2-1 on target
45%-55% possession

Game 2:
Liverpool - Swansea
0-0 score
11-9 shots
1-4 on target
43%-57% possession

The animation difference you see, is, at least I think, because the AI doesn't pressure the AI as much, and the AI uses the same programmed movements against itself because it isn't adapting to the situation as it would against a user.

 

[Robalex]

I played two FIFA 12 CPU-CPU games, Liverpool (team rating 78) hosting Swansea (team rating 65), then Swansea hosting Liverpool. 10 minute halfs, Legendary difficulty. I use custom tactics: I've set up Swansea as a defensive, ball-possession team and Liverpool as an attacking team with a quicker buildup.

I can't say I saw any cheating. The games and the results played out about how one would expect between those two teams. One thing I did notice: when playing CPU-CPU, the players and player animations move more quickly than they do in User-CPU games. So something is definitely a bit different there.

Game 1:
Liverpool - Swansea
1-0 score
9-4 shots
2-1 on target
45%-55% possession

Game 2:
Liverpool - Swansea
0-0 score
11-9 shots
1-4 on target
43%-57% possession

Using your GP Mod, movement are realistic because I've got a TV in the room and when is real football and I can compare the speed of movement, passes and ball coefs.
I've increased SHOT_ATTRIBUTEs in order to see 5 minutes games with more realistic final scores. Would you try a worst team than Swansea? much worst?

 

[Robalex]

The animation difference you see, is, at least I think, because the AI doesn't pressure the AI as much, and the AI uses the same programmed movements against itself because it isn't adapting to the situation as it would against a user.

YES!, that's one of the reasons, AI pressure, but also poor players shooting accuracy against CPU team is too high. I tell you, I've seen many many games of bad teams against top teams and find unrealistic that they can score 1 or 2 goals, last sample was a very modest Arabian team beating Arsenal 2-0

 

[bangus]

I did some more testing: Liverpool vs Bray (team rating 53). Bray as I've set them are a one-dimensional team: all defense and possession, no attack, with 7 players set as backs or defensive midfielders. Again, nothing unusual going on in these games IMO, pretty much what one would expect, maybe too few Liverpool goals I suppose:

Liverpool - Bray
1-1(PK) score
13-3 shots
3-1 on target
33%-67% possession

Liverpool - Bray
1-1 score
16-2 shots
6-1 on target
41%-59% possession

I then tried an experiment. Thinking that these low, tight scores might be due to slider settings, I then lowered Shot Error and Pass Error sliders to 0 (were set on 55 and 70). My theory was that better teams' superior shooting ratings (and therefore scoring success) is maybe being nullified by the high Shot Error slider. My thinking was if I lowered that slider, a couple things should happen:

1. Players' shooting ratings should mean more, so better teams with better shooters should score noticably more goals.
2. Stronger, offensive teams that attack more than weaker, defensive teams, will have more scoriing opportunities and, again, should score noticable more goals.

The result:

Liverpool - Bray
7-2 score
12-12 shots
10-12 on target
35%-65% possession

From that experiment, I'd say lowering the Shot Error slider way down is the solution. Liverpool scored 5 more goals on less shots on target, obviously because their players are better shooters, and because the Shot Error slider was no longer nullifying their players' shooting ratings. What didn't work was lowering the Pass Error slider. Bray ended up with the same number of shot opportunities as Liverpool, probably because they were able to pass the ball around successfully until they could get a shot on goal. Their shot average in the first two games was more in keeping with the type of team they are, so I would set the Pass Error slider back to where it was.

 

[bangus]

The animation difference you see, is, at least I think, because the AI doesn't pressure the AI as much, and the AI uses the same programmed movements against itself because it isn't adapting to the situation as it would against a user.

Interesting points, that makes a lot of sense.