balance test

[panda_bear_smoking]

           yello its old friend jebushunter here    <dont think anyone remembers me> 

                what if you leave a server running of only bots all with the same AI and one bot for each weapon exclude nades and secondly weapons  or isnt there some way of using bots to determine weapon balance idk throw some ideas up

[Avis]

I had the same idea already

The point is, that it would take a very long time.
You'd also have to make tests for 1on1, DM, TDM and so on since the situations might differ.
But after all your idea is just like mine.
Virtually an EVEN AI which more or less fights itself should determine the balance ratios, if you test almost any situation with every weapon - multiple times.
(But you could not test the "Recoil" value aspect ).
Also, if bink isn't too high, bots fire as fast as possible.

If someone had such a huge amount of time and was so kind to put so much effort and time in something like that, these tests could be useful for a better understanding of "balance" -
and how "balanced" the Soldat weapons really are .

[7th_account]

Unfortunally the bots aren't clever enough to pull off human moves and tactics. They also ignore bink, selfbink and moveacc, making any result irrelevant. =/

[darkangel]

agree with 7th_account

[Pie]

Not exactly
you could determine a benchmark with this method, Using duplicate bots say, admiral or another.
Give each bot a specific weapon on a cycle of maps and leave fraps open/observe them.
I've done this with screenshots, it isn't as conclusive but it does have merit, i think it's worth someone trying.
If they have enough time, i could do it on my spare computer over the period of an hour or so.

X amount of games
With Y amount of time allowed
And then you could do what you want with the data.
Find the average kills a bot makes with a specific weapon, which weapon has a higher average ect.
Sounds like a good idea
Although you wouldn't be able to use it for players ect. Just if weapons are Extremely unbalanced(in theory, every weapon should have the same average for it to be balanced this could show a large number if it is not very balanced) Although luck would have something to do with it.
I still think i'll try it. Also someone should try this with players.. Like in a match of two players with equal or close to skill, which weapon/strategy works best.

[a-4-year-old]

"There are three kinds of lies: lies, damned lies, and statistics."

I don't think there would be a whole lot of good to come out of that, since bots are so fucking stupid you might find that the chainsaw is overpowered.

[Pie]

"There are three kinds of lies: lies, damned lies, and statistics."

I don't think there would be a whole lot of good to come out of that, since bots are so fecking stupid you might find that the chainsaw is overpowered.

Maybe, i'm testing it tonight, i might post the stats later...

I was halfway through testing and i realised my soldat is a mod so the weapons.ini isn't default. So i stopped. Although the saw was over powered in my mod.
SS didn't save
Forget about it T_T

[a-4-year-old]

All bots know how to do is charge at each other and fire, so boogieman would probably do pretty well against someone with say, an auto.

[Ziem]

Bots have nobink hax xD ...
This is stupid idea... (and bots can't fire the LAW or throw the knife btw.)

[mar77a]

Unfortunally the bots aren't clever enough to pull off human moves and tactics. They also ignore bink, selfbink and moveacc, making any result irrelevant. =/

I see why you didn't want to test the balance in the czech/pole gather....

[-Vis-]

Bots can SO fire the LAW.

But yeah, this won't achieve anything. Bots and real players are much too different. Except noobs maybe.

[excruciator]

time to give bot an IQ boost?

[-Vis-]

Just give them the ability to switch weapons when they run out of ammo (or switch to knife/saw if you get too close), and to backflip out of a stream of bullets.

Though this isn't really the place to discuss it.

[.Long-Range]

I think this idea would show the weapon balance when it comes to bots but not to humans. Much like 7th_account has stated.

[panda_bear_smoking]

i think still think its a good idea maybe not as bots but as a server letting real player vs bots and record it all mah prob was just as high back then as i am now

[12th_account]

I still think it's a terrible idea. Get over it.

[panda_bear_smoking]

u cxant judge balence wpn for human with out making tthem. all the same.       
the changes r just doing dmg i miss the no fire delay days.      the saw is the over powerfuled followed by soccom then knife. tbe otheer wpn r undeerpower
 edit: i dont even understand what i was saying up the. tsk tsk

maybe i was trying to say is that. i beleave that soldat is magic in the sense of mind over matter and its a null point to keep changing the wpn keep tweaking em. ive seen un-armed beat chainsaw 1v1 soccom beat barret. ive seen first time player destroy whole servers.  what would be nice is all wpn start with default values and you picck what u want. get to customize ye guns not rpg like but inthe player section choise the stats for ur gun. kinda like letting people MD just a little. ie extended clip, faster fR,less bink, blaa bla

[Shoozza]

You could speed up the whole game to get faster results however like 7th_account already said: Bots aren't clever.

Another way to test balance is to have global stats for all played games (or specific ones like SCTFL) with hit %, kills, popularity etc.
There will be local weaponstats in the next Soldat version so it's going in that direction.

[xstyle]

At this point there isnt even a need for a balance test, because there were atleast 3 sctfl seasons with the 1.5 weapon settings. Its obvious that minimi and spass are massively overpowerd + that deagels are too weak. Also making balance tests with bots makes no sense at all because they dont know how to play to use the weapons full potential.

[panda_bear_smoking]

At this point there isnt even a need for a balance test, because there were atleast 3 sctfl seasons with the 1.5 weapon settings. Its obvious that minimi and spass are massively overpowerd + that deagels are too weak. Also making balance tests with bots makes no sense at all because they dont know how to play to use the weapons full potential.

u füçk up the wpnz and people like u always whinning bout these are over powered those are underpowered. u cantt. judge based on humans are not tthe same we use skill bla bla. bots are the same sure there sstupid but they never play worst or better alwwys the same. too maany factors with humans. its like trying to get every one to aggree whhat ice cream flavor is the best...blue moon btw