Home

Adaptive AI - Play Based vs Route Based

This is a discussion on Adaptive AI - Play Based vs Route Based within the Madden NFL Last Gen forums.

Go Back   Operation Sports Forums > Football > Madden NFL Football > Madden NFL Last Gen
WWE 2K22 Trailer Announcement and Breakdown
EA Sports PGA TOUR to Exclusively Have All Four Majors
Next-Gen Tony Hawk's Pro Skater 1+2: Is the Upgrade Worth It?
Poll: What's more important to you, when the time comes to purchase a game? (Click to vote)
Reply
 
Thread Tools
Old 04-12-2009, 02:55 PM   #17
MVP
 
OVR: 21
Join Date: Jul 2002
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by mikeacard
i can only control one player on defense, so when someone is using a play that can exploit multiple players in the defense the adaptive a.i. is needed for the human. for example using a fast qb like mcnabb and rolling him out with a receiver running an 8 yard crossing route across the field. right now i have a choice i can either control a player and have him cover the scramble of the qb or cover the crossing route by the wr. whatever i do the person im playing against can exploit the a.i. of the cpu for the other matchup.

so lets say i decide im gonna cover mcnabb with a linebacker while he rolls out, and ill keep my db in man coverage on the crossing route. the other player will see that i am covering the roll out so he will make the quick throw to a the wr for a decent gain each time, because my db will never "learn" that the crossing route is coming. this means he will never jump the route no matter how many times it is run. so i get tired of getting beat by the pass so i tell the backer to spy the qb and manually cover the wr. now the player sees that the scramble will be available because the a.i. is too dumb to cover the scramble by its self. the backer may get into position but nobody else on the defense will realize that a scramble is coming no matter how many times it happens in a row, leaving the player to make one person miss in order to get a decent gain while im covering the crossing route manually.

correct play call is only half the battle. it is still neccesary for a humans a.i. teamates to "learn" and cover things that the human player is not able to cover.

If the defensive plays stop what they are designed to stop(within reason of the player talent on the team) and the offense is not given the ability to exploit the defense in a cheating way, there is no need for adaptive ai on the human team.

If someone is constantly throwing a 5yd out you will either call a zone defense that puts players in that area of the field or call man and put your best cover man on that WR.

If the defender is not noted for playing zones well and keeps allowing a WR to beat him. You gotta make a adjustment in personel or play calling. The player should not be made to do what he normally cannot.

You say if he recognises that play let him be able to jump it. OK now the O has to know this too, so now they call a in instead of a out. Shouldnt the Db still get burned because now he is overplaying for the out ? If I called a play expecting him to cover inside and he plays outside based on a tendeny tracked by the cpu, and the offense goes inside, I am gonna be pissed.

They need to make sure the players and the plays do what they are suppose to do. You the coach have to call your plays based on tendency and the ability of your defense.

If I have slow CB's I cant expect to play man against a team with fast Wrs without deep help, OR getting some kind of consistent pass pressure. I also cant expect the game to cheat up my defense to do what they are not able to do in real life.

Not saying you, but some people think there are defenses that stop everything and there are not. If someone is good enough to make reads and exploit what you are doing to them they should be able to without worrying about the cpu cheating the defense up to play how they really should not.

We have not had a Madden game that did not give one side of the ball some type of advantage that could be stopped by sound play calling. We have always had to find a way to game the game and exploit the exploit.
TombSong is offline  
Reply With Quote
Old 04-12-2009, 03:04 PM   #18
Rookie
 
OVR: 5
Join Date: Jul 2007
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by TombSong

You say if he recognises that play let him be able to jump it. OK now the O has to know this too, so now they call a in instead of a out. Shouldnt the Db still get burned because now he is overplaying for the out ? If I called a play expecting him to cover inside and he plays outside based on a tendeny tracked by the cpu, and the offense goes inside, I am gonna be pissed.

you make good points in the rest of your post, but i don't see this as a bad thing. in real football this exact situation can and does happen. if a db has been getting beat by a certain route all day he is going to be looking to stop that first. so if the tendency has been an out all day long and it all of a sudden gets changed to an in i wont have a problem with a db overplaying the outside even if the play called for him to play the inside. players in the nfl get beat because of things like this, i wont have a problem with my a.i. db jumping the wrong route sometimes either.
mikeacard is offline  
Reply With Quote
Old 04-12-2009, 03:17 PM   #19
Rookie
 
OVR: 4
Join Date: Mar 2009
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by mikeacard
you make good points in the rest of your post, but i don't see this as a bad thing. in real football this exact situation can and does happen. if a db has been getting beat by a certain route all day he is going to be looking to stop that first. so if the tendency has been an out all day long and it all of a sudden gets changed to an in i wont have a problem with a db overplaying the outside even if the play called for him to play the inside. players in the nfl get beat because of things like this, i wont have a problem with my a.i. db jumping the wrong route sometimes either.
That's my main beef with it. I don't want my guy jumping certain routes unless I tell him to. If this was a feature where I can tell my guy to overplay certain routes or a certain part of the field that would be great. IMO leaving it up to the CPU to make this decision could generate glitches and cheese tactics. This is all speculation on my part of course but I don't think it's out of the realm of possibility.
MoneyMoesh is offline  
Reply With Quote
Old 04-12-2009, 03:19 PM   #20
MVP
 
OVR: 21
Join Date: Jul 2002
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by mikeacard
you make good points in the rest of your post, but i don't see this as a bad thing. in real football this exact situation can and does happen. if a db has been getting beat by a certain route all day he is going to be looking to stop that first. so if the tendency has been an out all day long and it all of a sudden gets changed to an in i wont have a problem with a db overplaying the outside even if the play called for him to play the inside. players in the nfl get beat because of things like this, i wont have a problem with my a.i. db jumping the wrong route sometimes either.

What would be better is not to have the cpu decide whether the DB plays inside or outside. They should allow the user to call inside or outside shade. Then when the player gets beat outside or inside, you will know why and wont throw your controller at the screen because the cpu made the wrong decision for you, LOL

Thats my 2 cent though.
TombSong is offline  
Reply With Quote
Old 04-12-2009, 03:22 PM   #21
Rookie
 
OVR: 4
Join Date: Mar 2009
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by TombSong
What would be better is not to have the cpu decide whether the DB plays inside or outside. They should allow the user to call inside or outside shade. Then when the player gets beat outside or inside, you will know why and wont throw your controller at the screen because the cpu made the wrong decision for you, LOL

Thats my 2 cent though.
Exactly Sir, me and you are on the same page.
MoneyMoesh is offline  
Reply With Quote
Advertisements - Register to remove
Old 04-12-2009, 03:56 PM   #22
Rookie
 
OVR: 5
Join Date: Jul 2007
Re: Adaptive AI - Play Based vs Route Based

Quote:
Originally Posted by TombSong
What would be better is not to have the cpu decide whether the DB plays inside or outside. They should allow the user to call inside or outside shade. Then when the player gets beat outside or inside, you will know why and wont throw your controller at the screen because the cpu made the wrong decision for you, LOL

Thats my 2 cent though.

i have no problem with that. i still wouldn't mind the cpu having some type of "human" error in it.
mikeacard is offline  
Reply With Quote
Old 04-12-2009, 07:44 PM   #23
All Star
 
Valdarez's Arena
 
OVR: 51
Join Date: Feb 2008
Location: Austin, TX
Posts: 5,091
Blog Entries: 33
Re: Adaptive AI - Play Based vs Route Based

You have a lot going on in the NFL to help with adjustments. You have coordinators siting in the booth watching the game. You have 11 guys on the field playing individually (though as part of a team) each recognizing the players, their routes, and their tendencies. It's pretty simplistic to say that you need to call the right defense. Lets be honest, you can pretty much boil an online video football game down to 5 to 6 plays, and which point it becomes little more than a guessing match at calling the right play to stop them. Rock, paper, scissors anyone?

I don't mind being schooled by someone who uses their entire playbook, but when you come upon the 5 play wonders who run out of one formation, it becomes boring, frustrating, and an extremely unauthentic experience. This is what I hope the Adaptive AI addresses.

I do agree with folks points about the AI not doing something you haven't told it to do. What we don't want to see is the defense expose itself by trying to cover repetitive patterns. For example, if you have a cover 2 in play, and your CB decides to go deep to cover a streak pattern by the outside WR, thus leaving the outs exposed. This would be bad. Very bad. Instead, the safety should get a speed boost, or a read boost to make a play on the WR and effectively knock down the ball.

My main point is that if we are going to be penalized for using the same play or route over and over again (which it sounds to me like we are, and I believe is a good thing), then we need some type of feedback in the game to warn us.
__________________
Follow Me On Twitter: http://twitter.com/Valdarez
Read My Blog - Vision Is Everything
Valdarez is offline  
Reply With Quote
Old 04-12-2009, 10:49 PM   #24
Rookie
 
OVR: 9
Join Date: Jul 2006
Re: Adaptive AI - Play Based vs Route Based

I never read anything that said the human teams DB would change assignments or anything like that.

The feature is just an awareness boost based on recognition of the same pattern. IE, the corner "recognizes" the out pattern earlier in the route then he did the first few times that pattern was run.

For a human controlled defense this isn't an audible or a hot-route defensively, it's a simple awarness boost once a play is ran a certain amount of times.
Sven Draconian is offline  
Reply With Quote
Reply


« Previous Thread | Next Thread »

« Operation Sports Forums > Football > Madden NFL Football > Madden NFL Last Gen »


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -4. The time now is 03:10 PM.

Top -