Warrior General Games

Okay, If I want to have multiple OAI package inputs for a single output with the auto adjust on, will it adjust to the best working plays within each package or just to each package (in it's entirety) that is working the best. For example if I have inside rushing package with different plays, a short pass package, an outside rush package... etc. Will the auto adjust for those plays in each particular package or to the package that is working best (maybe my inside running game is working better than the other packages in that ouput, would it favor the "inside rush package" without adjusting to the plays in it?????)

I hope you can understand my poorly worded question!

Per OUTPUT only. The whole package will be adjusted up/down.

thanks

More questions:

1) Are the adjustment relative or additive ? Does an output with 0% can be adjusted ?

2) What the sim does when the sum of % is not 100 (in a package or in an input) ? My guess is that it will first scale the inside the package to 100% and then scale the input to 100%. is that true ?

3) Can be have the information what is considered a successful play for the AA ? ( is a 2yds gain in a run play a success or failure ? what if it bring first down? , ....)

So can you create a whole bunch of packages each with a single play?

Not an OC, just wondering...

Originally posted by Shart
So can you create a whole bunch of packages each with a single play?

Not an OC, just wondering...

you could

would be useful for hail marys

ಠ_ಠ

Originally posted by Dr. Showtime
you could

would be useful for hail marys

ಠ_ಠ

Seems you could find weak links in the defense. A certain play where the OLineman is oVVning the DLineman and it keeps getting 5 yards a shot so the auto adjust just keeps calling it. Or if the team figured they would gamble and not cover the FB in a formation, that play might rise to the top and kill them.

Wait a minute, it doesn't just auto-adjust packages, does it? If you have a play in there, it will auto adjust that just like it would a package I assume. So it would be silly to create one play packages.

Like I said, not an OC, just trying to figure out what it was doing....

I saw you specify the rate at which it auto-adjusted. If you have a play that's kicking the crap out of the defense, why wouldn't you want it to auto adjust at the highest rate? Am I correct in that as the computer sees that a play is doing well against the defense that it will automatically weight that play to where it is called more often?

Okay if I understand this right (I could totally MISunderstand this as I am an old feeble minded man and get easily confused!) anyway it would seem not to be beneficial to use the packages for your AI if you plan to use Auto-adjust. It would seem the auto adjust to each input as a whole. So if you have multiple packages with many plays, maybe some of the plays are working and some are not, but as a whole that"package is working better than another package you have that does include some successful plays, just not as many. This would keep the bad plays that are not working and erase the few good plays from the other package that may be working.

However, if you just had individual plays in for an output, instead of packages, the actual bad plays would be reduced and the successful plays would be increased!

Again, IF I understand Bort's answer correctly!

Is there a link anywhere that describes what is supposedly happening with this function? Thanks.

Originally posted by tragula
More questions:

1) Are the adjustment relative or additive ? Does an output with 0% can be adjusted ?

2) What the sim does when the sum of % is not 100 (in a package or in an input) ? My guess is that it will first scale the inside the package to 100% and then scale the input to 100%. is that true ?

3) Can be have the information what is considered a successful play for the AA ? ( is a 2yds gain in a run play a success or failure ? what if it bring first down? , ....)

These are excellent questions. This Auto-adjust has the potential to have quite an impact on the game.

I know we added a bunch of plays and had the whole monkey-gate ordeal, but I'm surprised that this auto-adjust hasn't gotten more discussion and clarification. If you can set 5 plays at 20%, set the auto-adjust to "very quickly", and by the end of the game have those plays adjusted something along the lines of 80-5-5-5-5 because some play happens to have found some huge hole in the defense, I'd think that would be kind of significant.

Originally posted by Shart
These are excellent questions. This Auto-adjust has the potential to have quite an impact on the game.

I know we added a bunch of plays and had the whole monkey-gate ordeal, but I'm surprised that this auto-adjust hasn't gotten more discussion and clarification. If you can set 5 plays at 20%, set the auto-adjust to "very quickly", and by the end of the game have those plays adjusted something along the lines of 80-5-5-5-5 because some play happens to have found some huge hole in the defense, I'd think that would be kind of significant.


Well I think that is EXACTLY what it does! So If you are doing a good job scouting and put in play that shows a hole and you give that play a high percentage but the DC has corrected for it, this auto adjust can save your butt and go to other plays that ARE working! We have always had this but it was a one time choice that covered your entire offense and defense AI's. It would frequently make choices that really did not help you. Hopefully this version being much more situational specific will eliminate those bad choices and stick totally within the specific ones you have made for that output!

Originally posted by jimbcnu
Okay if I understand this right (I could totally MISunderstand this as I am an old feeble minded man and get easily confused!) anyway it would seem not to be beneficial to use the packages for your AI if you plan to use Auto-adjust. It would seem the auto adjust to each input as a whole. So if you have multiple packages with many plays, maybe some of the plays are working and some are not, but as a whole that"package is working better than another package you have that does include some successful plays, just not as many. This would keep the bad plays that are not working and erase the few good plays from the other package that may be working.

However, if you just had individual plays in for an output, instead of packages, the actual bad plays would be reduced and the successful plays would be increased!

Again, IF I understand Bort's answer correctly!

No,
it just put constraints on what you can do with it.

A nice use is have you base offense set as one package and then add a crazy play with 5% and set "very quickly". If you managed to surprise the defense that crazy play will be dominate and you role. If that play is worthless it will full to almost 0% quickly and will not interfere with your game plan.