How In The World Can This Game Ever Be Balanced?
Harry_S_Truman
Join Date: 2002-11-21 Member: 9568Members
Join Date: 2002-11-21 Member: 9568Members
<div class="IPBDescription">Too many things change at once....</div> I don't see how we can ever get a semblance of balance in NS. Every changelog contains dozens of point changes, changes in build times, changes in weapon damage, etc., etc.
In a game like this changing one single element can have a significant impact upon the game, yet we constantly see numerous changes at a time. How can you ever pinpoint what effect these have when you do so many at one time?
In a game like this changing one single element can have a significant impact upon the game, yet we constantly see numerous changes at a time. How can you ever pinpoint what effect these have when you do so many at one time?
Comments
There is no quantitive way that even a few thousand community members can detail how these changes effect each other when so many are done at one time.
Perfect balance isn't exactly possible, but a "workable" or "playable" balance, most certainly is.
NS is a complex mod, but as far as games are concerned, it's not that complex. Try balancing Hearts of Irons or even some more advanced, full-fledged RTS games.
A few thousand people (or even a dozen) can easily tell you what works, and what doesn't. Some things are obvious -- the time it takes to electrify an RT -- others are less obvious, such as gorge armor points. But the goal isn't to balance the game perfectly (i.e, there's an equal, and opposite tactic for everything for both sides).
The goal is just to make the game playable. So.. balancing isn't that hard, really. And like I said, a few thousand testers help you learn what works, and what doesn't.
I guess I don't really understand what it is you're getting at...
The lesson is about keeping the number of variables in a situation small so that you can track exactly what changes are caused by which "solutions."
It's keeping the game both balanced and <b>fun</b> that is difficult.
Having elite players test it for us <b>is</b> good for the public. We strive for perfection.
"OMG! DIE GORGEH" <!--emo&::asrifle::--><img src='http://www.unknownworlds.com/forums/html/emoticons/asrifle.gif' border='0' style='vertical-align:middle' alt='asrifle.gif'><!--endemo--> <!--emo&::gorge::--><img src='http://www.unknownworlds.com/forums/html/emoticons/pudgy.gif' border='0' style='vertical-align:middle' alt='pudgy.gif'><!--endemo-->
I see this so many times.
<i>"Buddy, learn to aim!"</i>
It seems to me the weapons damage values are balanced for NS. One less thing to worry about, Flay!
In a game like this changing one single element can have a significant impact upon the game, yet we constantly see numerous changes at a time. How can you ever pinpoint what effect these have when you do so many at one time? <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
Yes, exactly right. <!--emo&::marine::--><img src='http://www.unknownworlds.com/forums/html/emoticons/marine.gif' border='0' style='vertical-align:middle' alt='marine.gif'><!--endemo-->
Its balanced
2.01e is I mean
What to see a cool trick
It dneso't metatr waht oredr you tpye wrdos in as lnog as the frist and lsat lteters are in the rhigt pclae you can siltl raed the wrod. Tihs is bauecse we dnot raed wrods by the letetr but by the wrod. So a wrod can be msseed up rlealy bldaly and we wlil sitl be albe to raed it.
And many changes are made in balancing patches all the time, even by the likes of Blizzard (SC, and WC3 etc).
What complicates NS really is the commander vs no commander. Exceptional aliens can make a larger difference than exceptional marines. This is because no matter how good a marine is, he is limited by his commander. An exceptional alien however can do infinitely more damage on their own...
RC1 - Lowered HMG cost
RC2 - Lowered Hive cost
RC3 - Raised HMG cost
RC4 - Lowered HA cost
RC5 - Increased Nade damage
RC6 - Reduced Electricity research time
RC7 - Increased skulk attack power
RC8 - Reduced Fade Cost
RC9 - Reduced Fade Damage
Kinda get the point. In researching a bug in a computer, program, or design that an individual is working on, it is easy to narrow down the problem, over time. However when trying to balance a game it isn't as easy to make single changes at a time, that is where taking feedback and tweaking come in to play.
So maybe it is a better idea instead of looking at these changes as a lot quickly, look at the bigger picture.
This game with the balancing is more like a complex mathmatical equation where the marines = X and the aliens = Y, and where you are trying to get the values of X and Y to be as equal as possible. So when you increase the value of X and decrease the value of Y you will either get a power shift or a closer balance of the game.
Now for those of you that are math gods, come up with an equation which equates this game into numbers and determine what needs tweaked. <!--emo&;)--><img src='http://www.unknownworlds.com/forums/html/emoticons/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif'><!--endemo-->
Marines need for a commander -1
Aliens independent resource structure -.5
Aliens ability to place a node for each alien on the team in the first 30 seconds +5
Marines have long range +2
Marines slow movement -1
Aliens fast movement +0
Marines electrified nodes +1
Marines cost to electrify -2
Marines GL cost to damage +1
Keep going with that and try to get "aliens = marines"
This statement is incorrect. It take a player who can shoot 12 bullets into a vanilla skulk to kill him. You have to think beyond the numbers when it comes to factoring balance.
Ultimately, you can't balance anything 100% when player skill varies so intensely. Quite often the source of many people's woes is the fact that they are just outskilled or outstrated, not "outbalanced" as it were.
A team of good aliens vs. a team of good marines <i>with a bad commander</i> = Marine Loss.
A team of good aliens vs. a team of bad marines <i>with a good commander</i> = Marine Loss.
A team of bad aliens vs. a team of bad marines and a good commander = Marine Win
A team of bad aliens vs. a team of bad marines with a bad commander = Tie
A team of good aliens vs. a team of good marines with a good comm = Tie, (from what I've seen of CAL matches) possible Marine Loss.
The problem is that in order for the Marines to win they either have to face terrible aliens or win an attition battle with a great commander.
This statement is incorrect. It take a player who can shoot 12 bullets into a vanilla skulk to kill him. You have to think beyond the numbers when it comes to factoring balance.
Ultimately, you can't balance anything 100% when player skill varies so intensely. Quite often the source of many people's woes is the fact that they are just outskilled or outstrated, not "outbalanced" as it were. <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
I meant this, really. I thought that the part where bullets have to hit them was going to be assumed by everyone.
But yes, you are correct, playerskill is very important to judge, and this is why you recruit people like vets to help balance the game.
The lesson is about keeping the number of variables in a situation small so that you can track exactly what changes are caused by which "solutions." <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
Ahh, but this isn't an electrical circuiit. YUou can't simply take something out and test it to see if it works. You can't even be sure only one thing is broken... Replacing one 'broken' component can often break one that worked fine before.
Balancing is not the same as electronics.
All things take time... <!--emo&::nerdy::--><img src='http://www.unknownworlds.com/forums/html/emoticons/nerd.gif' border='0' style='vertical-align:middle' alt='nerd.gif'><!--endemo-->
The lesson is about keeping the number of variables in a situation small so that you can track exactly what changes are caused by which "solutions." <!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
Ahh, but this isn't an electrical circuiit. YUou can't simply take something out and test it to see if it works. You can't even be sure only one thing is broken... Replacing one 'broken' component can often break one that worked fine before.
Balancing is not the same as electronics. <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
Still, coding is very similar to balancing. It's just that the algorithms in balancing are FAR more complex.
It's really perfectly simple. For example, in the lastest release 2.01 the cost of Heavy Armor was reduced from 20 to 15. Why? I've never seen a lack of heavy armor in any game I've played. How does this change effect the gameplay? Obviously it allows the TSA to field more heavy armor and field it faster. Since any decent marine team with HA and some welders never really needs an extra suit of heavy armor, why would the cost be decreased? There's about 100 more factors that can be impacted by this change.
Now, take the number of changes total and you can see that the changes made seem to only wildly skew the game in a different direction.\
For anyone that says NS is balanced, they must play the Alien side a lot. Because on all the pubs I play on, the aliens still win just about every game. The whole "It's balanced for clan play" is a silly argument as well, since about 99% of us never play "clan" games.
For all you "fan bois", I'm not bashing your game, just trying to figure out what the devs are doing by changing so many things so quickly.
Obviously the sentiment of an Alien fan.....
The lesson is about keeping the number of variables in a situation small so that you can track exactly what changes are caused by which "solutions."<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
Yes, that's a good way to troubleshoot, but it's not a good way to optimize a complex system (in the sense of mathematical optimization, not tuning performance).
Max
Exactly...I seem baffled by some of the changes they make.
Still, coding is very similar to balancing. It's just that the algorithms in balancing are FAR more complex. <!--QuoteEnd--> </td></tr></table><span class='postcolor'> <!--QuoteEEnd-->
Coding is fairly similar, but not so much electronics.
However, in BOTH cases the vital difference is the obvious one... coding and electronics are basically mathematical. You can predict that, if you alter thing A, it will change thing B a certain way.
But balancing a game doesn't work like that, thanks to that good 'ol human element. People will throw a curve at anything you do anyway... you can't calculate for that. (Well, you can, but you're probably going to be wrong.)
I guess they just wanted to see how well it would work. Trying out new things seems to be pretty much what the betas are for (along with bug stomping).
I hope they fiddle around with the JP costs next. (:
The lesson is about keeping the number of variables in a situation small so that you can track exactly what changes are caused by which "solutions."<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
Yes, that's a good way to troubleshoot, but it's not a good way to optimize a complex system (in the sense of mathematical optimization, not tuning performance).
Max<!--QuoteEnd--></td></tr></table><span class='postcolor'><!--QuoteEEnd-->
Well, I don't see Flayra whipping out the Monte Carlo or gradient optimization methods, so I don't think that the game is getting "optimized" either. I'm not knocking the way Flayra is doing it. I was simply trying to explain the thread parent's point. With as many people chomping at the bit for the next patch as there are, he doesn't have the time to take the slow, methodical method. The only metric he has for balance is server statistics on the win/loss ratio. He doesn't have the time to make small incremental changes and then wait two weeks for some win/loss statistics to be generated.