Decided to run some tests to make sure tagging was working correctly. Started out with a simple test in the following game:
http://test.goallineblitz.com/game/game.pl?game_id=1448
Basically, I tagged the running QB as "Runner" and created a specific input that used a goal line formation with all players blitzing if the running QB entered the game. Goal line was called in every instance the running QB came in.
The general default builds for the test server teams are pretty terrible so I created a offense with pro level type of builds for Milwaukee and a defense with pro level type of builds for Kansas City. Milwaukee's defense was the default and Kansas City's offense was the default. I ran one test game to make sure the builds I had created worked - returner was too powerful so I changed him to be much slower/less agile. Some concerns from the sim, especially regarding pass blocking/passing, but that was not what I was testing but I made note of my concerns and posted them:
http://test.goallineblitz.com/game/game.pl?game_id=1458
Originally posted by thehazyone
Some definite concerns came from the game and not just from Team #1 Pro Offense vs Team #2 Pro Defense as it was happening on both sides of the ball.
Passing is atrocious - between the two teams they were a combined 24 of 84 - Team #1 Pro Offense had moderate success in terms of yards (304.5) but 1/2 of that came on two plays.
Interior line pressure from DT's and NT's is out of whack. Test NT1 had 7 hurries, Test DT1 had 8 hurries, Typical NT (Backup) had 7 hurries, Typical NT had 2 hurries. For the test defense this was against very good interior line builds. These hurries are what led to the atrocious passing numbers as hurries definitely have a more profound effect on the quality of the pass.
I'll put in a more extensive AI tomorrow with a little more play calling variety and more typical game plan than basic AI, but those two things worry me.
Ran a test game with tagging utilized in more depth. Essentially, I tagged the Passing QB, Running QB, Power HB, Elusive HB, Blocking TE, Receiving TE. I then created specific defensive inputs for the following scenarios:
Running QB in game
Power HB in game with Blocking TE
Elusive HB in game
Passing QB in game with Receiving TE
I had specific defenses in for each of those scenarios. Each time those scenarios hit, the correct play was called.
http://test.goallineblitz.com/game/game.pl?game_id=1461
Again, had concerns with pass rushing/OL blocking/pass game and noted them (see comment in Testing the New Pass Blocking Code thread).
There were a few cosmetic issues with tagging as far as wrong position's showing up in drop boxes. Pointed those out for Bort.
General observations about Tagging:
Works as intended. Teams that utilize a player in only one way (i.e. Running QB solely as runner, Elusive HB solely as pass receiver, Power HB solely as inside runner) will no longer be able to do so without facing a defense that likely shuts it down completely. It will require that these players do more than one thing. It will create more cumbersome and complicated defensive AI's, which is the biggest reason I am not a fan of tagging - not because of how it works, but because of the work it creates. If any one has a specific situation they want tested, post here and if it needs testing (meaning I'm not just going to test anything for the sake of testing it), I'll do so and post results in here.
http://test.goallineblitz.com/game/game.pl?game_id=1448
Basically, I tagged the running QB as "Runner" and created a specific input that used a goal line formation with all players blitzing if the running QB entered the game. Goal line was called in every instance the running QB came in.
The general default builds for the test server teams are pretty terrible so I created a offense with pro level type of builds for Milwaukee and a defense with pro level type of builds for Kansas City. Milwaukee's defense was the default and Kansas City's offense was the default. I ran one test game to make sure the builds I had created worked - returner was too powerful so I changed him to be much slower/less agile. Some concerns from the sim, especially regarding pass blocking/passing, but that was not what I was testing but I made note of my concerns and posted them:
http://test.goallineblitz.com/game/game.pl?game_id=1458
Originally posted by thehazyone
Some definite concerns came from the game and not just from Team #1 Pro Offense vs Team #2 Pro Defense as it was happening on both sides of the ball.
Passing is atrocious - between the two teams they were a combined 24 of 84 - Team #1 Pro Offense had moderate success in terms of yards (304.5) but 1/2 of that came on two plays.
Interior line pressure from DT's and NT's is out of whack. Test NT1 had 7 hurries, Test DT1 had 8 hurries, Typical NT (Backup) had 7 hurries, Typical NT had 2 hurries. For the test defense this was against very good interior line builds. These hurries are what led to the atrocious passing numbers as hurries definitely have a more profound effect on the quality of the pass.
I'll put in a more extensive AI tomorrow with a little more play calling variety and more typical game plan than basic AI, but those two things worry me.
Ran a test game with tagging utilized in more depth. Essentially, I tagged the Passing QB, Running QB, Power HB, Elusive HB, Blocking TE, Receiving TE. I then created specific defensive inputs for the following scenarios:
Running QB in game
Power HB in game with Blocking TE
Elusive HB in game
Passing QB in game with Receiving TE
I had specific defenses in for each of those scenarios. Each time those scenarios hit, the correct play was called.
http://test.goallineblitz.com/game/game.pl?game_id=1461
Again, had concerns with pass rushing/OL blocking/pass game and noted them (see comment in Testing the New Pass Blocking Code thread).
There were a few cosmetic issues with tagging as far as wrong position's showing up in drop boxes. Pointed those out for Bort.
General observations about Tagging:
Works as intended. Teams that utilize a player in only one way (i.e. Running QB solely as runner, Elusive HB solely as pass receiver, Power HB solely as inside runner) will no longer be able to do so without facing a defense that likely shuts it down completely. It will require that these players do more than one thing. It will create more cumbersome and complicated defensive AI's, which is the biggest reason I am not a fan of tagging - not because of how it works, but because of the work it creates. If any one has a specific situation they want tested, post here and if it needs testing (meaning I'm not just going to test anything for the sake of testing it), I'll do so and post results in here.