Warrior General Games

Originally posted by pottsman
But it's being adjusted with new sim games over the next few days before season 12 starts.
And will probably be tweaked in the preseason.

This is a huge thing that people don't realize...

If the numbers are off slightly, we have approximately 5 thousand games to look at as a sample size. Even if when the season rolls around, and if after 2 preseason games, NOT A SINGLE SACK happens anywhere, it can be adjusted before any meaningful game takes place.

Dammit guys, I was trying to make him realize this but you're giving the answers away.

Sorry about the way I went about this last night, I didn't express my concerns in a reasonable manner.

It is reasonable to think that the pass blocking dynamic is going to be based off of test server results isn't it? If not, why run tests on the test server? If it is, then you should be running tests with evenly matched players.

My concern is this. Bort has expressed that he wants to see somewhat realistic stats when evenly match opponents play each other. The tests label the players as "average" which implies he believes that those players going against each other would be an even match up. I am worried he is going to base the new OL/DL interaction and blocking effectiveness based on how those players perform against each other. Since the DT is much better than the C in this case, it would mean that a DL would have to be much better than an OL to function at a level Bort expects them to perform at vs. evenly matched opponents.


Lvl 47 C with no CE
http://goallineblitz.com/game/player.pl?player_id=941336

Lvl 60 DT getting offers from AAA teams
http://goallineblitz.com/game/player.pl?player_id=349185

Originally posted by Deathblade
This is a huge thing that people don't realize...

If the numbers are off slightly, we have approximately 5 thousand games to look at as a sample size. Even if when the season rolls around, and if after 2 preseason games, NOT A SINGLE SACK happens anywhere, it can be adjusted before any meaningful game takes place.

This is what I'm worried about. I'm worried about DT's having to get totally shut out by this new pass blocking system for someone to say, "Maybe the o-line is a little too overpowered?"

As it is, hurries are relatively useless, they don't have the impact that they do in real life. So sacks become the only meaningful statistic that implies an effective pass rush.

The vast majority of DT/NT builds are pretty incapable of getting to the QB. The DT build that is shown is described as being "Typical," when in actuality, it is far from typical. That's a much better than average DT build. 100/100/70 str/agi/spd, try finding DT's with that combo.

So when you have one build that is definitely superior to the oppositions build, and in 191 passing plays all a superior build can manage is 1 sack?
1 / 191 = 0.00523560209

That's acceptable? Really?

Originally posted by Deathblade
...NOT A SINGLE SACK happens anywhere, it can be adjusted before any meaningful game takes place.

So when you say this, it becomes worrying. It sort of implies that .005% is an acceptable sack rate for a DT with a superior build.

This is my thinking about the new pass rushing system. I'm worried about a perception that DT's aren't supposed to get sacks, and anything they do get, they should be grateful for.

Originally posted by blankspace
I'm going to let you think about this conversation before my next reply.

I've seen nothing to indicate what sort of philosophy the admins here have towards DT's and pass rushing. So what I wrote before, still applies. I'm worried that even if the admins go through all those sims and all those replays, they still won't see a problem.

The more games I watch the worse I feel about the new OL/DL interaction.

The last game showed 2 sacks. 1 on each team. The one by the NT is where he wasn't even blocked. The RG looked like he went to pick up the MLB blitzing and the LOT tried to come block the NT.

The one by the RDE was against a very good pass blocking LOT, but

Right now though it just looks like anything the DL gets is based on getting a lucky roll, and it appears that DTs best bet is to be as strong or stronger than the OL they are going against. From what I have seen looking at builds and results it appears agility has become next to worthless.

Originally posted by Rage Kinard
The more games I watch the worse I feel about the new OL/DL interaction.

The last game showed 2 sacks. 1 on each team. The one by the NT is where he wasn't even blocked. The RG looked like he went to pick up the MLB blitzing and the LOT tried to come block the NT.

The one by the RDE was against a very good pass blocking LOT, but

Right now though it just looks like anything the DL gets is based on getting a lucky roll, and it appears that DTs best bet is to be as strong or stronger than the OL they are going against. From what I have seen looking at builds and results it appears agility has become next to worthless.

The sacks on the test server are meant to be low...since all of the offensive lineman are supposed to be passing their vision checks.

Originally posted by Deathblade
The sacks on the test server are meant to be low...since all of the offensive lineman are supposed to be passing their vision checks.

Not a good test then :/

Originally posted by Adderfist
Not a good test then :/

It's a good test to see if the blocking code has holes in it...

I mean really, what the hell is it with people thinking that the test server is for getting fine tuned balancing?

Originally posted by Deathblade
The sacks on the test server are meant to be low...since all of the offensive lineman are supposed to be passing their vision checks.

So DL will have to rely on failed vision checks to get consistent pressure?

what concerns me is that it looks like the only legitimate way for an interior DL to beat an interior OL on a consistent basis is to be stronger. Other than that it he just has to get lucky.

The 125 strength 95 agility NT didn't really look much better against the 110 strength 110 blocking RG than the 110 strength 70 agility, 60 speed NT had against the 105/105 RG that was used in another test.

Originally posted by Rage Kinard
So DL will have to rely on failed vision checks to get consistent pressure?

what concerns me is that it looks like the only legitimate way for an interior DL to beat an interior OL on a consistent basis is to be stronger. Other than that it he just has to get lucky.

The 125 strength 95 agility NT didn't really look much better against the 110 strength 110 blocking RG than the 110 strength 70 agility, 60 speed NT had against the 105/105 RG that was used in another test.



Thank you. You said what I was going to, much better then I could.

Guys I think you need to calm down a bit. For the purpose of testing they need to see the blocking scheme work. If there are pathing issues, it will be hard to separate them from legitimate pressure. As long as the pathing and blitz pickups are fundamentally sound, the vision check can always be backed off. You just wait. When this is actually implemented with the correct settings in place, people are going to be complaining that their 20 vision G is completely useless.

From the replays I've watched, it seems like the Oline is a hell of a lot smarter than they were. Once this is stable, I'm sure a whole new debate over the difficulty of the vision check will begin.

Originally posted by steellithium
Guys I think you need to calm down a bit. For the purpose of testing they need to see the blocking scheme work. If there are pathing issues, it will be hard to separate them from legitimate pressure. As long as the pathing and blitz pickups are fundamentally sound, the vision check can always be backed off. You just wait. When this is actually implemented with the correct settings in place, people are going to be complaining that their 20 vision G is completely useless.

From the replays I've watched, it seems like the Oline is a hell of a lot smarter than they were. Once this is stable, I'm sure a whole new debate over the difficulty of the vision check will begin.

Isn't there a point when players are at lower lvls, that 20 vis should be acceptable. I can't see 1st and 2nd season players having much higher and ending up with much of a build in the end.

Originally posted by steellithium
Guys I think you need to calm down a bit. For the purpose of testing they need to see the blocking scheme work. If there are pathing issues, it will be hard to separate them from legitimate pressure. As long as the pathing and blitz pickups are fundamentally sound, the vision check can always be backed off. You just wait. When this is actually implemented with the correct settings in place, people are going to be complaining that their 20 vision G is completely useless.

From the replays I've watched, it seems like the Oline is a hell of a lot smarter than they were. Once this is stable, I'm sure a whole new debate over the difficulty of the vision check will begin.

This makes perfect sense. Just because they're running a test, doesn't mean they're testing the code in its entirety. They may have altered things and could possibly be testing one aspect of the code. Allowing the entire OL to pass the vision checks could simply be testing whether or not they make the correct movements when they DO pass vision checks.

They seem to be isolating certain parts of the code, making sure they work independently, and then possibly testing the code as a whole before making it go live. Then, when it is live, they're going to keep an eye on it in the first couple pre-season games, and attempt to solve any issues before the season begins.

Is this not the best way to go about it? How can you fix something if you don't know EXACTLY what's going wrong?

Originally posted by mwindle
Isn't there a point when players are at lower lvls, that 20 vis should be acceptable. I can't see 1st and 2nd season players having much higher and ending up with much of a build in the end.

Well, remember how vision checks work. You get one every tick or so many ticks. Even if you don't read the blitz instantly, lower level blitzers won't close the distance as quickly. Chances are vision checks at low levels will be equivalent to the lack of overall speed.