User Pass
Home Sign Up Contact Log In
Forum > Discuss GLB Issues With Catch22 > NOT A BUG - CLOSED > Combine some threads: the G-OT Gap "Exploit"
Page:
 
Link
 
Originally posted by sjmay


Good point, though I am curious why would you get a speed acceleration from shedding a block, or maybe max acceleration from shedding a block, thinking it out, DLine engages Oline, Dline is blocked, DLine sheds block, Dline has to accelerate again, sounds like he should be starting again from zero, is that what is happening? Or is he still maintaining some kind of momentum while being blocked that would account for a speed burst?


June 13, 2009
- Tunnel vision SA being active allows DE's to get a short speed boost when successfully stunning their blocker.
 
jdbolick
offline
Link
 
Originally posted by j10er
Personally, I think there's some kind of slingshot/pinball effect going on where the DE bounces between the LG and LOT, gaining acceleration/speed as he goes, and it produces a "shot out of a gun" issue.

I said the same thing to Bort in a thread the other day and he agreed, then said that he's taking a look at the speed boost.


Originally posted by fallingmercury
June 13, 2009 - Tunnel vision SA being active allows DE's to get a short speed boost when successfully stunning their blocker.

By the way, it's not just TV because there are DTs who are having the same sort of success.
Edited by jdbolick on Jul 17, 2009 13:14:44
 
sjmay
offline
Link
 
Originally posted by fallingmercury
Originally posted by sjmay



Good point, though I am curious why would you get a speed acceleration from shedding a block, or maybe max acceleration from shedding a block, thinking it out, DLine engages Oline, Dline is blocked, DLine sheds block, Dline has to accelerate again, sounds like he should be starting again from zero, is that what is happening? Or is he still maintaining some kind of momentum while being blocked that would account for a speed burst?


June 13, 2009
- Tunnel vision SA being active allows DE's to get a short speed boost when successfully stunning their blocker.


Oh, that answers the question of should a DE get a short speed boost,

Please, do me a favor, just use this,

http://greasefire.userscripts.org/scripts/show/46121

Edited by sjmay on Jul 17, 2009 13:14:18
 
jdbolick
offline
Link
 
Originally posted by sjmay
Oh, that answers the question of should a DE get a short speed boost,

Bort felt that defenders who won their rolls weren't actually benefiting from it without reverse pancakes because they usually didn't get far before the blocker re-engaged. So essentially the DE/DT had to win several rolls in a row to get by. The idea was to reduce the number of "wins" necessary, which it did, but this problem came in because of how the speed boost interacted with the new limit on blocking angles.
 
Link
 
Originally posted by jdbolick
Originally posted by sjmay

Oh, that answers the question of should a DE get a short speed boost,

Bort felt that defenders who won their rolls weren't actually benefiting from it without reverse pancakes because they usually didn't get far before the blocker re-engaged. So essentially the DE/DT had to win several rolls in a row to get by. The idea was to reduce the number of "wins" necessary, which it did, but this problem came in because of how the speed boost interacted with the new limit on blocking angles.


Look at the LT fail because of blocking angle... Even though he's overlapping almost half of the RE...

http://goallineblitz.com/game/replay.pl?game_id=590111&pbp_id=12174450

 
kurieg
offline
Link
 
Blocking angles should be limited. Too much hook blocking as it is in the game.

 
jdbolick
offline
Link
 
Another potential fix that I suggested to Bort was having blocker dots "rotate," so their blocking angle isn't just from 10 o'clock to 2 o'clock, but could rotate to any comparable area, like 8 o'clock to 12 o'clock. You don't want the LG to go from blocking a DT to his left to suddenly blocking someone to his right, but if he is engaging the DT to his left then it makes sense for him to be able to shift his acceptable blocking angle to the left as well.

I don't know how difficult that would be to code, but it would be the best solution in my view, simply because it would allow penetration for DTs & DEs who win that first roll, but the LG would still have a chance to prevent them from getting to the quarterback. So it would disrupt the pocket and maybe force the QB to his right, but wouldn't result in such a quick sack.
 
Link
 
Originally posted by kurieg
Blocking angles should be limited. Too much hook blocking as it is in the game.



True.

But it's also true that when the DE lines up between the LT and G, that he should be in the LT's blocking cone even before the play starts.
 
Link
 
Originally posted by jdbolick
Another potential fix that I suggested to Bort was having blocker dots "rotate," so their blocking angle isn't just from 10 o'clock to 2 o'clock, but could rotate to any comparable area, like 8 o'clock to 12 o'clock. You don't want the LG to go from blocking a DT to his left to suddenly blocking someone to his right, but if he is engaging the DT to his left then it makes sense for him to be able to shift his acceptable blocking angle to the left as well.

I don't know how difficult that would be to code, but it would be the best solution in my view, simply because it would allow penetration for DTs & DEs who win that first roll, but the LG would still have a chance to prevent them from getting to the quarterback. So it would disrupt the pocket and maybe force the QB to his right, but wouldn't result in such a quick sack.


Holy crap, blocking angle is fixed in a forward direction?
 
sjmay
offline
Link
 
Originally posted by jdbolick
Another potential fix that I suggested to Bort was having blocker dots "rotate," so their blocking angle isn't just from 10 o'clock to 2 o'clock, but could rotate to any comparable area, like 8 o'clock to 12 o'clock. You don't want the LG to go from blocking a DT to his left to suddenly blocking someone to his right, but if he is engaging the DT to his left then it makes sense for him to be able to shift his acceptable blocking angle to the left as well.

I don't know how difficult that would be to code, but it would be the best solution in my view, simply because it would allow penetration for DTs & DEs who win that first roll, but the LG would still have a chance to prevent them from getting to the quarterback. So it would disrupt the pocket and maybe force the QB to his right, but wouldn't result in such a quick sack.


It shouldn't be that hard, he already has that kind of programming with the HB's cone of vision, so you make a few tweaks and put it on the Oline etc, but the essence of that is already programmed with HBs.
 
j10er
offline
Link
 
Originally posted by jdbolick
Another potential fix that I suggested to Bort was having blocker dots "rotate," so their blocking angle isn't just from 10 o'clock to 2 o'clock, but could rotate to any comparable area, like 8 o'clock to 12 o'clock. You don't want the LG to go from blocking a DT to his left to suddenly blocking someone to his right, but if he is engaging the DT to his left then it makes sense for him to be able to shift his acceptable blocking angle to the left as well.

I don't know how difficult that would be to code, but it would be the best solution in my view, simply because it would allow penetration for DTs & DEs who win that first roll, but the LG would still have a chance to prevent them from getting to the quarterback. So it would disrupt the pocket and maybe force the QB to his right, but wouldn't result in such a quick sack.


Much more realistic, imo.

Plus, it coincides with Bort's current project to introduce more vision cones into the game. There are hardly any currently.
 
jdbolick
offline
Link
 
Originally posted by fallingmercury
Holy crap, blocking angle is fixed in a forward direction?

I don't know that for sure, it's just a guess. For run blocking it isn't, but in pass blocking it doesn't really seem like there is any rotation.
 
Link
 
Originally posted by jdbolick
Originally posted by fallingmercury

Holy crap, blocking angle is fixed in a forward direction?

I don't know that for sure, it's just a guess. For run blocking it isn't, but in pass blocking it doesn't really seem like there is any rotation.


This play would seem to confirm the fixed-cone theory. Even though the LT is moving to the right, right into the side of the RE, he can't interact with him. It's like the AI knows it's supposed to block that DE, but it just can't because the cone only points downfield...

http://goallineblitz.com/game/replay.pl?game_id=590111&pbp_id=12174450
Edited by fallingmercury on Jul 17, 2009 13:27:11
Edited by fallingmercury on Jul 17, 2009 13:26:38
 
jdbolick
offline
Link
 
Originally posted by fallingmercury
This play would seem to confirm the fixed-cone theory. Even though the LT is moving to the right, right into the side of the RE, he can't interact with him. It's like the AI knows it's supposed to block that DE, but it just can't because the cone only points downfield...

I also previously pointed out the April 16th entry in the changelog which says: "- Blockers rotate their bodies better when run blocking to better get in between the ball carrier and tackler." I speculated that because it only says run blocking, that blockers previously didn't rotate, and that they still don't when pass blocking. I suggested to Bort that he apply the same change to pass blocking, but I have no idea if he saw it.
 
Link
 
Would be cool to have a slider for o-linemen to slightly increase the size of your blocking cone, in exchange for an increased chance of being called for holding...
 
Page:
 


You are not logged in. Please log in if you want to post a reply.