Warrior General Games

Originally posted by larkis
Mike, the problem is that this gives the testers more insight into the AI.
That alone is an advantage, you are going to realise what works and what doesn't and you will subconsciously going to follow it. Will anyone willingly shoot themselves and recruit a dud when they realise they can cheaply get a stud next season?
You know what plays will work, what plays don't work and what affects those plays.
No matter how anyone looks at it, it is an advantage to be a tester.

Now onto the integrity of being a tester. I would like to believe that everyone in the WEPL won't abuse their rights as a tester and will try their hardest to not give anyone an unfair advantage. However the problem occurs in other leagues where some testers do abuse their privilege, leak information out and teams have dominated due to that information.

Either make it available to all for testing or available to none


We are not testing in the same Sim AI as the Live site, what works there bears little or no relation to what works here.

We are not even testing Ai which is about to be released, we dont get to see the final version, and we dont know if anything that is changed in the AI will ever be brought over to the main site.

Often we dont even know what the change was.

If I added 40 to Jumping and got 50 PDs, it might be a direct result, or because Bort changed the Pass quality calculation, or the INT code, or something in the WR Vision code, or in the Def reaction time code, or a increase in the delay between a WR catching the ball and securing the ball, or increased the effects of Hurries, or changed the timings of the snap read delay.... etc.

There is just not the scope to link Results on the Test server directly to Results on the live server with any real validity.

With something like the Def Play Creator, there will be more familiarity if you've tested it than if you hadnt, thats obvious.
But the Testers will be writing the User Guide/FAQ for it, and while somepeople will argue that its possible we will hide some things, the chances of Bort allowing that are slim to none.

And no team has dominated because of the Test Server, it has only been going for two weeks.

The supposed 'Leak' that everyone is talking about is that higher Tackling is good against Powerbacks.
Someone said that if LBs were built for speed to stop elusive backs in the past, then they should consider adding Tackling or strength if they faced a known powerback.

Honestly does that really sound like privileged information?

There is no reason to make the test server available to all, thats not testing, thats playing for free. Opening the test server to everyone wont produce any more benefit than not having it, so thats the mootist position being suggested.

Getting rid of it puts us back where we were the last 7+ seasons, Bort makes a change in preseason, which sends everything off balance, and they adjusts stuff through the season.

So its either Testers or constant Nerfs.

The 'Percieved' advantage and the 'Actual' advantage are completely different.
There is only so much one person can learn about the game without access to Borts code, and all the Testers are at or fast approaching that stage already.

I can see that someone who has only a partial grasp on the Sim AI could learn a lot of stuff that they wouldnt have been able to otherwise, but stepping back into the real world, the people on the test server are not learning anything new.

The actual advantage we have in all honesty is knowing that Bort is looking at a certain known issue and might fix it.
Its about the same advantage as anyone able to read every Bort post in Bugs.

There is and always was a plan that the changelog would have a 'In Testing' tab or similar so everyone would know ahead of time what was being looked into.

The AI is neither here nor there. Our testing should not really provide any "AI" assistance.

Generally speaking, I think everybody already knows what works and what doesn't work - we're working to try to ensure that everything works.

With the release of the logs you will also be privy to what has been fixed and what remains to be fixed, and you can freelysubmit more ideas and fixes in the bug threads (particularly if you provide plenty of info to back up your findings).

As for the leaked test information.

Given there have been no testers prior to 2 weeks ago and no information handed out during that time, what exactly is this leaked info that has ensured all teams associated with testers in other leagues have won/dominated?

Originally posted by larkis
However the problem occurs in other leagues where some testers do abuse their privilege, leak information out and teams have dominated due to that information.

Either make it available to all for testing or available to none
How can teams have dominated because of information gained on the test server when that has been up for all but two weeks? And one of the owners of probably the most dominant team in GLB at the moment has been only a recent addition to the test server, so they certainly have not dominated because of anythign they learned on there.

As for making things available, Bort already made all games simmed on the test server available to everyone on very short notice. Or are you referring to let everyone test stuff to their hearts content? Yeah, that sure will work, as there are no people at all out there who try to find every loophole with the intent to explot rather than to fix it ... sorry, aber that is just a tad bit naive, not even considering the additional resources such a setup would require.

Originally posted by Zickzack

Of course those who want to believe this is a huge advantage won't believe anything we say anyway, but I'd hope that most people would grant those currently involved enough integrity and credibility to not consider this an issue.

I trust the people chosen to report exploits they find. That's not a concern to me. I have two hangups about it:

1) The fact that a select few people have access to 500+ different players and their builds and get to watch them play on their own terms and test out specific things that they wish to see, thus seeing which builds are better in the process.

2) Those same people then passing on said information to a few close friends, people they choose to leak it to. Cry all you want about your integrity, and I personally trust all the known testers here in the WELP, but I will post proof that all isnīt as squeaky clean as you claim.

My problem with the testing has all to do with player builds and nout to do with the AI. Man, would I like to know if a HB is better with 68 vision or if it is really as unimportant as it seems. Man, would I like to know if my HB would perform noticably different if he had 68 confidence.

This is what Iīm talking about, a post by the poster formerly known as gart888, a card carrying douchebag:

Originally posted by
Also i should add that the info that was leaked to me didn't play any roll in us winning our championship. But you can sure as hell bet it will effect how i build players from here on out.

http://goallineblitz.com/game/forum_thread.pl?thread_id=2067898&page=156 (2nd post)

Yes, the information leaked remains "tackling is more important than strength for tackling powerbacks" or something daft. Which hardly qualifies as rocket science to be quite honest.

Of course Me and Mike could go to Bort and suggest that's a bollocks piece of logic and get it all changed.

"Leaked" info can come back to haunt people if it turns out to complete rubbish, or something that is considered an "exploit" or "not working as intended".

As yet the actual Leak has not been identified

If thatīs all it is, then I guess it was just the usual grandstanding buy Fart888.

Originally posted by Kevin Smith


My problem with the testing has all to do with player builds and nout to do with the AI. Man, would I like to know if a HB is better with 68 vision or if it is really as unimportant as it seems. Man, would I like to know if my HB would perform noticably different if he had 68 confidence.



That's what I am talking about, some of these what if situations that are usually left unchanged or only slightly changed from AI to AI.
Most people only know how speed, agility, strength and blocking usually affects plays. But how about what would a lot of vision for a TE or a lot of confidence for a FB. I know I don't know what is available to test, but there are variables I think everyone would like to know to do with the new VA (even the old ones) or some SAs that are ignored at the pro level.

As I said before, I don't know what is available to test and how, what flexibility there is with the AI either. All we know is there is a testing server and that information is getting leaked. Who is doing what, I don't know.

Just to clear up some of the mess, could the testers provide screenshots of the testing server so we actually know what it is.

Oh, and currently HB vision is in my opinion completely irrelevant and untested, as is confidence, and there would be no reason to test it without some kind of logic behind it.

If Bort for instance introduced code to take into consideration the HB's vision and confidence when running certain plays, or all plays then there'd be a need to check that it was in fact doing what it was intended.

HB vision - to check whether the HB would go to a different hole in a play or would avoid certain tackle?
HB Confidence - See how the HB plays throughout the hole game, effect on the # of broken tackles - if all the other attributes were -10%?

Originally posted by Koncorde
Oh, and currently HB vision is in my opinion completely irrelevant and untested, as is confidence, and there would be no reason to test it without some kind of logic behind it.

If Bort for instance introduced code to take into consideration the HB's vision and confidence when running certain plays, or all plays then there'd be a need to check that it was in fact doing what it was intended.

Please explain. Vision irrelevant? To testing, you mean? I think it is very relevant, as there have been believable claims that it actually hurts HBs to have high vision. I am testing it on my own, but it is costing me money, months of my time and might end up giming my HB.

As it's an irrelevant attribute for most rushers. Power running for instance doesn't require vision - but that's well known. It's meant to be more for evading tackles and finding holes (which is counterproductive to a powerbacks running style). The issue comes with high vision that there are (against fast defences) actually very few holes, or the HB considers all holes covered or soon to be covered - so ends up headfaking and juddering like a loon as there's no overriding "get a yard" logic until he has absolutely no other recourse but to initiate contact.

In theory a powerback can gain from vision, when evading tackles and looking for the path of least resistence, at least it shouldn't hurt them really - but an advantage may exist in actually failing a vision check on the majority of plays because of the actual absence of holes (effectively sending the HB up the middle blindly and gaining 3+ yards through belligerence).

For testing purposes there's no much point racking up vision on a HB because we know the routing is broken and it has nothing to do with vision and a lot to do with the O-Line/D-Line interaction, blocking, Intimidation/threat levels and holes opened.

Probably, if everyone had remained at 50 speed and HB's had gone to 80, then a HB would have been able to exploit his speed and vision to hit the gaps. But because of the relative speed of the LB's in the current game, they're able to fill the gaps and/or influence the HB's decisions enough to stump their routing.

I've raised a thread on rushing logic on the server for it to be looked at.

Originally posted by larkis
HB vision - to check whether the HB would go to a different hole in a play or would avoid certain tackle?
HB Confidence - See how the HB plays throughout the hole game, effect on the # of broken tackles - if all the other attributes were -10%?

That would be a test that benefited "us" and have little educational purpose (you can do the same test with equipment).

Vision influences routing = true. Whether it's an improvement is debateable due to some of the issues mentioned above in the previous post.

Confidence functions. Whether its impact is noticeable is debateable again. Part of the issue being we don't know the energy/confidence/morale interaction rate, recovery rate - or even the loss rate.

Testing for the purpose of just putting points into confidence would stray close to just testing something for the sake of knowing an answer.

Well there is testing going on
If there are tests that are going on the benefit the game, then let everyone see the tests, not just a few.
This is so we can see what is going on and know what is being tested.

People are scared of the unknown and hate it, make it known and we will probably not mind it.

Originally posted by Zickzack

How can teams have dominated because of information gained on the test server when that has been up for all but two weeks? .


Originally posted by Zickzack

Yeah, that tester exclusive time machine that helped me to to move back this weekend to before the season and decide to bank on a power back really helped a lot. Good thing you guys don't know about that time machi... oh drat!




HMMM!