Warrior General Games

Originally posted by Malachorn
Originally posted by tpaterniti

Stamina does not work in our sims. Testing defenders against very agile KRs on Super Elusive he was trying to see if the problem was them taking bad angles to the ball carrier because of poor agility. He set speed at a constant - maybe he could have set it lower, I don't know.

...I kinda didn't believe that when I saw it, tbh.
I'm pretty sure you're using the same engine as the "live" server, tbh.

Stamina does have an effect, but a lot of the testers have just made all of their players have 100 stamina so we don't have to worry about creating and signing backups, we can just roll with the starters. So it DOES work in the sims, but it's essentially taken out of the equation in many cases.

Originally posted by r8
Originally posted by Malachorn

Originally posted by tpaterniti


Stamina does not work in our sims. Testing defenders against very agile KRs on Super Elusive he was trying to see if the problem was them taking bad angles to the ball carrier because of poor agility. He set speed at a constant - maybe he could have set it lower, I don't know.

...I kinda didn't believe that when I saw it, tbh.
I'm pretty sure you're using the same engine as the "live" server, tbh.

Stamina does have an effect, but a lot of the testers have just made all of their players have 100 stamina so we don't have to worry about creating and signing backups, we can just roll with the starters. So it DOES work in the sims, but it's essentially taken out of the equation in many cases.

First, someone should tell tpat then.
Second, I don't think that's exactly how stamina works, tbh, and it is NOT "essentially taken out of the equation"

To think so would encourage bad testing, FYI.
And bad testing yields bad results.

Originally posted by Malachorn
Second, I don't think that's exactly how stamina works, tbh, and it is NOT "essentially taken out of the equation"

It is. With even 60 stamina you can pretty much play every snap on either offense or defense and end the game with over 80 energy. With 100+, there'd be basically no energy drain at all; certainly not enough to cause any statistically significant testing errors.

Originally posted by Zurai
Originally posted by Malachorn

Second, I don't think that's exactly how stamina works, tbh, and it is NOT "essentially taken out of the equation"

It is. With even 60 stamina you can pretty much play every snap on either offense or defense and end the game with over 80 energy. With 100+, there'd be basically no energy drain at all; certainly not enough to cause any statistically significant testing errors.

Not true at all, tbh.
And this shows even more lack of appreciation that people have for in-game "breath," ending energy, and stamina in general.
Very common misconceptions, tbh.

making only players with 100 stamina because you are too lazy to make backups is bad testing imo.

Taking stamina out of the equation can be useful yet damaging. If you're simply trying to test two players at full power, then that works fine. However, if you're trying to judge things as the play and game progress (two different things), then 100 stamina is a BAD idea.

I'd like to see tests done to figure out just how drastically a player's performance breaks down as a game progresses. If it's something drastic, it needs toned down. If a player with 40 stamina can end a hard fought game with 90 energy, it needs tweaking. If a player with low stamina doesn't get tired over the course of a single play. That's something that needs to be looked at.

A RB taking off for a 90yd TD run should be fine with 50 stamina. He'll obviously tire as he runs, but it shouldn't be drastic. However, if he's got 40, 30, 20 or lower, he should show signs of fatigue as he stretches that run.

Either way, he should be quite tired at the end of that run. I'd imagine a solid energy drop until he gets some rest. You make him pull of a few of those, and he'll be relatively worn out, I'd imagine. Is that the case now?

An interesting test would be a team with never ending stamina against a team with limited stamina. Say a team with all 100 stamina against a team with all 40 stamina.

You pit them against each other with equal values in every other stat. Then you wait for the 100 stamina team to start outpacing the 40 stamina team. If it doesn't happen or doesn't happen soon enough you know that stamina isn't quite working right.

Originally posted by TheInfinity
An interesting test would be a team with never ending stamina against a team with limited stamina. Say a team with all 100 stamina against a team with all 40 stamina.

You pit them against each other with equal values in every other stat. Then you wait for the 100 stamina team to start outpacing the 40 stamina team. If it doesn't happen or doesn't happen soon enough you know that stamina isn't quite working right.

That could be something as well.

Originally posted by Staz
Originally posted by TheInfinity

An interesting test would be a team with never ending stamina against a team with limited stamina. Say a team with all 100 stamina against a team with all 40 stamina.

You pit them against each other with equal values in every other stat. Then you wait for the 100 stamina team to start outpacing the 40 stamina team. If it doesn't happen or doesn't happen soon enough you know that stamina isn't quite working right.

That could be something as well.

Wouldn't you have to factor in confidence as well? I like the idea for this test, but say someone on the 40 stamina team starts getting worn down and begins to drop passes/miss tackles etc. That causes their Confidence to drop and exhaserbates the problem leading to more dropped passes/missed tackles.

From the threads I'm starting to think two things. 1. Testers are concentrating on what will help them and their teams the most. 2. Testing is very narrow in scope and I have yet to see actual methodology posted as to what the change actually was and the results of the adjustments as testing progressed.

Can someone point out a link on any issue from when the starting point to where the "sweet spot" was determined to be?

Originally posted by Signalsgt
Originally posted by Staz

Originally posted by TheInfinity


An interesting test would be a team with never ending stamina against a team with limited stamina. Say a team with all 100 stamina against a team with all 40 stamina.

You pit them against each other with equal values in every other stat. Then you wait for the 100 stamina team to start outpacing the 40 stamina team. If it doesn't happen or doesn't happen soon enough you know that stamina isn't quite working right.

That could be something as well.

Wouldn't you have to factor in confidence as well? I like the idea for this test, but say someone on the 40 stamina team starts getting worn down and begins to drop passes/miss tackles etc. That causes their Confidence to drop and exhaserbates the problem leading to more dropped passes/missed tackles.

From the threads I'm starting to think two things. 1. Testers are concentrating on what will help them and their teams the most. 2. Testing is very narrow in scope and I have yet to see actual methodology posted as to what the change actually was and the results of the adjustments as testing progressed.

Can someone point out a link on any issue from when the starting point to where the "sweet spot" was determined to be?

Wow. Someone isn't clueless here, IMO!!!
Yeah, too many of the tests are terribly simplistic and don't account for how many variables they're not controlling and still want to believe that a couple tests somehow give them conclusive evidence about something.

...it's a pretty dangerous idea, tbh, and I'm not convinced these testers are doing a great job with coming up with meaningful results.

Originally posted by Signalsgt

From the threads I'm starting to think two things. 1. Testers are concentrating on what will help them and their teams the most. 2. Testing is very narrow in scope and I have yet to see actual methodology posted as to what the change actually was and the results of the adjustments as testing progressed.


Oddly, from someone who now is responsible to test complex systems, if you don't have a test plan/methodology, then you are rife to missing things, performing uneeded tests, and simply presenting a very poor result as a finished product.

If you don't have enough tools to actually test the product, you end up extending the length and number of tests required to completely test the product, and by any reports, there are no specific test tools for the testers on this server.

Someone might be a decent coder (and the jury is still out on that), but he certainly doesn't have a clue on how to test his product.

PinTBC

Originally posted by Malachorn
Originally posted by r8

Originally posted by Malachorn


Originally posted by tpaterniti



Stamina does not work in our sims. Testing defenders against very agile KRs on Super Elusive he was trying to see if the problem was them taking bad angles to the ball carrier because of poor agility. He set speed at a constant - maybe he could have set it lower, I don't know.

...I kinda didn't believe that when I saw it, tbh.
I'm pretty sure you're using the same engine as the "live" server, tbh.

Stamina does have an effect, but a lot of the testers have just made all of their players have 100 stamina so we don't have to worry about creating and signing backups, we can just roll with the starters. So it DOES work in the sims, but it's essentially taken out of the equation in many cases.

First, someone should tell tpat then.
Second, I don't think that's exactly how stamina works, tbh, and it is NOT "essentially taken out of the equation"

To think so would encourage bad testing, FYI.
And bad testing yields bad results.

Now I am not sure because when we started we were using the friendly engine - now maybe I am misunderstanding how that works. I can find out though.

Originally posted by tpaterniti
Now I am not sure because when we started we were using the friendly engine - now maybe I am misunderstanding how that works. I can find out though.

Trust me, stamina works on your test server.
I'd bet good money on it.

Still, having all players with 100 stamina DOES NOT "essentially take stamina out of the equation."

I think this testing is being done backwards. Rather than just throwing out players and see how they do why not

Create several players at each position (100/200/ or whatever can be feasibly tested).

Make teams from the players created

Run the games and see if results (game and player) are within acceptable parameters. If so, then run the games several more times, and see if all results are in an acceptable range. Testers also watch games to see if there are problems with the visual results and report what others may see as bugs.

When results are not acceptable in original or multiple tests and/or testers find problems with visual results then change code to try and fix what is not acceptable.

Re-run games with the exact same players. Rinse, repeat. Keep doing this until SIM is working the way Bort wants it to work.

BTW-I have made this suggestion in the suggestion forum.

Rage,

Except they aren't just testing that,

They are testing to see if outrageous builds will end up busting the code, that is a part of testing as well.