User Pass
Home Sign Up Contact Log In
Forum > Test Server Discussion > ***Test Server Update 8/15/10***
Page:
 
Catch22
offline
Link
 
Originally posted by AngryAnton
The plan we have come up with is to test each SA at 5, 10 and 15, and then to put that SA back down to 0 and move along to the next SA at 5, 10 and 15 points.

There has also been talk about doing further testing once the initial SA testing is done, particularly with opposing SAs such as monster hit and cover up. However, I'm sure you'll realize that this will take some time, and that testing for stacking and even interactions wouldn't possibly be finished soon.


I haven't posted the plan for testing SA's yet - all we've done thus far is the base tests for comparison purposes.
 
Catch22
offline
Link
 
As for the reason we're doing the SA testing - it's simple.

1) To see if a SA is working properly - what this means is we'll be evaluating whether the SA is working as we intend it to, and that's something only GLB staff is going to know. We'll be using the testers and test teams to run the games and compile the data but in the end, the types of adjustments that are made are not going to be revealed to the public other than "XXX SA adjusted XXX to balance" type statements in the changelog. We might also be making a few changes in regards to SA's, but are using this testing to help us identify how to best make these changes. This process will be a pretty lengthy one, so don't expect results in the next 2 months.

And yes, all testing results will be published on the live server. It's the main purpose of the Test Documenter Team - if you feel it is not getting published properly, please contact me and I will make sure it is.
Edited by Catch22 on Aug 16, 2010 08:38:41
 
bgharst
offline
Link
 
Catch, are you just lumping in the Scat Back and Zone Focus with this test? Or are you going to speed those 2 up since, you know, you kind of dropped the ball on that one?
 
Octowned
offline
Link
 
I see nothing wrong with implementing but not testing these SAs in the first place, in such a controlled manner they are being tested today. A programmer knows what his changes are going to do 99.9% of the time. If you've already coded the sim to handle your attributes, player interactions, decisions based on rolls, etc., it's all just math from there. If you want route running to give between 0-10% boost on speed while running a route, that is very easy to add to the formulas without testing the SA itself, and to be confident that it will work.

That is why I think a controlled experiment (and even better would be statistical testing) will finally help balance things out. It isn't whether or not the SA works as coded, but if the impact has any meaning in the metagame.
 
Ivanhoe
offline
Link
 
Originally posted by Catch22
We might also be making a few changes in regards to SA's, but are using this testing to help us identify how to best make these changes. This process will be a pretty lengthy one, so don't expect results in the next 2 months.


Would it be safe to say that any changes to SA's or VA's will not be implemented until the beginning of Season 19?

Half my team invested in Aura of Intimidation SA and Intimidation VA for Season 17. I can't see holding on to these worthless Abilities for 1.5 seasons and hope they work then. I will have to pull the plug on this failed endeavor, accept the heavy loss in player resources, and attempt to convert builds back to where they were previously.

 
beenlurken
offline
Link
 
Thanks for the clarification Catch!
 
Dr. E
offline
Link
 
Originally posted by SikoraP13 DTD
no, nobody was given levels at which the stuff should work at (to my knowledge). I'm fairly sure only bort and maybe catch know that, but the tests with 5, 10, and 15 in the SAs will be compared to the control (0 in the SAs) to see what/if any effect they have. Basically we're comparing what effect having the SA has in comparison to not having the SA. If there's no perceivable effect. For example, if having 15 in sticky hands doesn't increase catch percentage versus having 0 in it or if 15 in Lower the Shoulder doesn't increase BTKs, it'll be shown to catch/bort and they'll look into it. At least, to my knowledge that's how this test is going to be designed to work. Anton will have more information on this than I would but it seems obvious that Catch doesn't want people to know exactly when things start working since we're only doing testing at 5,10 and 15 and not 1-15.


The best way to test would be having the testers blind as to if it's 5/10/15. That way they aren't looking for a specific result.
 
tautology
offline
Link
 
Originally posted by Octowned


It isn't whether or not the SA works as coded, but if the impact has any meaning in the metagame.




QFT


Originally posted by Dr. E


The best way to test would be having the testers blind as to if it's 5/10/15. That way they aren't looking for a specific result.



Also a good point.
Edited by tautology on Aug 16, 2010 12:41:36
 
AngryAnton
offline
Link
 
Originally posted by Catch22
Originally posted by AngryAnton

The plan we have come up with is to test each SA at 5, 10 and 15, and then to put that SA back down to 0 and move along to the next SA at 5, 10 and 15 points.

There has also been talk about doing further testing once the initial SA testing is done, particularly with opposing SAs such as monster hit and cover up. However, I'm sure you'll realize that this will take some time, and that testing for stacking and even interactions wouldn't possibly be finished soon.


I haven't posted the plan for testing SA's yet - all we've done thus far is the base tests for comparison purposes.


Yea, by "come up with", I meant suggested, not that it had been decided. I should have made that more clear. Sorry about the confusion.
 
djhero2
offline
Link
 
the Seahawks are going to be bad this year I bet
 
BP
offline
Link
 
Originally posted by djhero2
the Seahawks are going to be bad this year I bet


I think you'd win that bet.

What I don't understand is the Bort wrote the code right? he controlled the basis for all the interaction with his formulas...now he doesn't understand how things work and we need to test it?

I can understand glitches...but writing a game from scratch requires you to understand the interactions..seeing as you created them right? Sounds like the sim is out of control TBH..maybe the "not following the AI" bug is the game becoming self aware? lol.

Anyway, users don't need to test this..like someone else said. Hire a couple of ITT tech grads to sit at home in their room and drink MD Code Red all day and knock this out...wouldn't take 2 years, entire multiplatform games are coded and fully tested in less then 2 years....I GLB can afford to do this correct with experts, no?
Edited by BP on Aug 23, 2010 18:37:13
Edited by BP on Aug 23, 2010 16:24:46
 
AngryDragon
offline
Link
 
Originally posted by BP
I think you'd win that bet.

What I don't understand is the Bort wrote the code right? he controlled the basis for all the interaction with his formulas...now he doesn't understand how things work and we need to test it?

I can understand glitches...but writing a game from scratch requires you to understand the interactions..seeing as you created them right? Sounds like the sim is out of control TBH..maybe the "not following the AI" bug is the game becoming self aware? lol.

Anyway, users don't need to test this..like someone else said. Hire a couple of ITT tech grads to sit at home in their room and drink MD Code Red all day and knock this out...wouldn't take 2 years, entire multiplatform games are coded and fully tested in less then 2 years....I GLB can afford to do this correct with experts, no?


+1

I think the time would be better served having a couple programmers looking for mistakes in the code rather than customers watching the output of the code.
 
smittdog101
offline
Link
 
No VA or SA should be too powerful or underpowered. It's the balance we are looking for in all VA's and SA's right? This way when you have an idea of what to build, at least you can have the faith that there will be some effectiveness per level with the right builds.

Without thurough testing on all of these and isolating each on vs. the other, then we will be stuck looking at plays for seasons to come and not knowing what the root cause of the buggy reaction's are.

Is it the build or is it the SA / VA causing the issues that we see in the game from time to time?

I believe that is the answer all the testers are striving towards. Hopefully there will be some patterns. Then the game can take off from there.
 
NiborRis
offline
Link
 
Originally posted by BP
I think you'd win that bet.

What I don't understand is the Bort wrote the code right? he controlled the basis for all the interaction with his formulas...now he doesn't understand how things work and we need to test it?

I can understand glitches...but writing a game from scratch requires you to understand the interactions..seeing as you created them right? Sounds like the sim is out of control TBH..maybe the "not following the AI" bug is the game becoming self aware? lol.

Anyway, users don't need to test this..like someone else said. Hire a couple of ITT tech grads to sit at home in their room and drink MD Code Red all day and knock this out...wouldn't take 2 years, entire multiplatform games are coded and fully tested in less then 2 years....I GLB can afford to do this correct with experts, no?


This is pretty naive thinking. Chaos theory, complex systems, etc etc - it's fairly trivial to take a small number of relatively simple equations and not be able to fully predict the outcome of all possible inputs. When you start considering the large number of inputs involved here (22 players in a play, with 14 attributes, 10+ SAs, VAs, and transient data like location on the field, current velocity, energy, breath, morale, etc...) it's not surprising at all that the full implications of all the various formulas require actually looking at the results rather than just thinking about it in your head for a bit.
 
Redster
offline
Link
 
Originally posted by NiborRis
This is pretty naive thinking. Chaos theory, complex systems, etc etc - it's fairly trivial to take a small number of relatively simple equations and not be able to fully predict the outcome of all possible inputs. When you start considering the large number of inputs involved here (22 players in a play, with 14 attributes, 10+ SAs, VAs, and transient data like location on the field, current velocity, energy, breath, morale, etc...) it's not surprising at all that the full implications of all the various formulas require actually looking at the results rather than just thinking about it in your head for a bit.


+1

There's a ton of stuff going into this sim there's no way to predict each little hiccup.
 
Page:
 


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