Balance in NS2

Tweadle · February 2012

<div class='quotetop'>QUOTE (Harimau @ Feb 10 2012, 03:09 AM) <a href="index.php?act=findpost&pid=1901634"><{POST_SNAPBACK}></a></div><div class='quotemain'>rebirth is right, and I will disagree with Tweadle on two points.</div>
Not quite sure where we disagree.

<div class='quotetop'>QUOTE (Harimau)</div><div class='quotemain'>So it's not so much a question of balance, as it is of understanding</div>
I don't want balance at the expense of fun and I also agree with your point on understanding/knowledge. However, this knowledge can come about intuitively (mechanics that make sense) *or* forcefully (tutorials). Natural Selection maybe suffered from a lack of both and I've always advocated a comprehensive tutorial because it has no drawbacks besides the time taken to make it. Intuition is important too but is attached to various other issues which I think are important to discuss in their own right. The case for framing things in terms of knowledge is all very well and good when there is something to actually know. Right now, there isn't and I'm saying there should be. Wallhopping is, so far, unsuccessful and I see its limitations making it super-hard to adequately replace bunnyhopping.

I also never claimed that balance should only be achieved at any particular level, nor that its possible to achieve at all levels. I do, however, think that a solid movement system which scales somewhat with aim is crucial in alleviating these problems. I pretty much entirely agree with puzl, for the record (even on sometimes balancing with pubbers in mind), but I'm not quite as elegant or as eloquent on the subject.

<div class='quotetop'>QUOTE (Harimau)</div><div class='quotemain'>The question of fun may be subjective while balance is not, but it cannot be denied that there are concepts of fun that are common</div>
I agree here too. I'm not trying to make a case for ignoring what is fun but this discussion started as a balance issue and I maintained that theme. It was only when rebirth started pointing out some difficulties in reconciling balance and fun and why fun should take priority that I chimed in. Perhaps I should have been clearer, though. It was not necessarily a flaw with the logic of rebirth's spiel, but with his use of it. I think he takes his philosophy of game-design too far and comes to a conclusion that has too little an emphasis on balance, certainly in the case of aim/movement scalability.

Firstly, I think that his philosophy works very well as a general concept but particular scenarios sometimes require a stronger focus on balance than others - the marine/skulk relationship being one of them. This is predominantly due to how often these two classes interact at all stages of the game. I truly believe that the skulk requires some sort of mirror to the scaling of skill-to-aim marines and that it needs to be factored in the design decision at one of the early passes. I would even go so far as to suggest <b>starting</b> with this in mind (because its such an elusive, but important, goal) and <b>then</b> work on getting it fun. We do disagree on the prudency of this ordering though, I guess.

Secondly, I don't think our talk of balance has done anything to belittle the 'fun' argument in the first place. Rewarding movement mechanics need not be boring or tedious and I think bunnyhopping was neither. The reason I didn't go down that tack was because this thread was focused on the importance of balance and because the subject of whether it is fun or not is pretty polarised between those who can do it (well) and those who can't. In fact, you could perceive the demand for the reintroduction of bunnyhopping or its equivalent not simply in terms of balance but in terms of fun. After all, we could be arguing for major changes in the skill ceiling of the gorge, lerk or fade to compensate for marine aim. But this would leave the whole skulk experience really dull and unforgiving, with all skulks at top levels avoiding marines in favour of eating rts or something which is a thankless task.

Jiriki · February 2012

To be honest, I think balance is overemphasized issue. You can always teak some values to get decent balance, but that doesn't mean the game is fun or has depth.

<div class='quotetop'>QUOTE (puzl)</div><div class='quotemain'>It's important to listen to pro players but you have to use your instincts in deciding if their advice is correct. Pro players are usually great at finding flaws but often poor at providing solutions. Sometimes a really talented musician is not a great composer ( and vice versa ).</div>
I think the problem with better players providing solutions, is unaccountability. That is even, they don't directly bear the costs of not providing a good solution, and don't take into account exogenous factors like difficulty for beginners, those which the person calling the shots must. But you should watch out for intuition. Unless you are an expert, you should be careful of using it since that is the most biased system of our cognition. This was actually a topic of Kahneman's book of <a href="http://www.amazon.com/Thinking-Fast-Slow-Daniel-Kahneman/dp/0374275637" target="_blank">Thinking Fast and Slow</a>.

In fact, most disagreements (with experts) are probably <a href="http://hanson.gmu.edu/deceive.pdf" target="_blank">dishonest</a> which is based on Bayesian mathematics. I have meta-level concerns about most disagreements.

<div class='quotetop'>QUOTE (puzl)</div><div class='quotemain'>Games are too complex to heuristically make design decisions.</div>
I don't know about making design decisions, but players of either NS, chess or any other game play mostly on intuition. I'd guess some of the decisions in Starcraft would struggle the best mathematicians (artificial intelligence hello), but players develop intuitions to calculate the right answer to avoid these logical atomic calculations. This is also my experience in NS. Sometimes our brain leaves us some final bits to make a decision on but the heuristics provided the premises. It gives an illusion of us making a very reductionist analysis. I though Sirlin pointed to this and some studies backing this up in his articles.

<div class='quotetop'>QUOTE (puzl)</div><div class='quotemain'>And the issue of team stacking is not relevant to this discussion.</div>
Well maybe not to this discussion but team, if not server stacking was always a problem in NS. In fact, the problem was that you mixed bad and very good players together. We loved the ways someone could be good, but if such palyer used all that on public server, the end result wouldn't be pretty. In any game, whether it is Starcraft or some other competitive game, you have mechanics nobody below some skill-level knew or had to worry about. Removing those mechanics could deprive the game, because that is treating the symptoms, not the root cause. Take for example Wave Dashing from SSBM. It was removed in Brawl, yet it added a lot to pro games. Most who played this casually never knew how to do that as it took frame-level practice to perform, and lots of time to master (much harder than bhop tbh). The thing is though, in SSBM, casual gamers played with equal opponents. Likewise, most problems related to NS were related to teams being stacked. Who cares in Bronze if you'd have to write the collected works of Oscar Wilde at the grandmaster level in Starcraft?

<div class='quotetop'>QUOTE (puzl)</div><div class='quotemain'>I mean, even look at who is in the video above. Idra is on the record for thinking that Zerg is the worst balanced race in NS2, and he has, on many occasions ranted about how retarded protoss are ( too easy to play ) and how OP terran is ( too easy to capitalise on small advantages ). However, in SC1 he played Terran and, surprise surprise, Idra found that Terran was the weakest race, often moaned about its problems despite the fact that 4 out of the 5 SC1 bonjwas were Terran. So has Idra magically picked the weakest race each time, or is it, perhaps, in his character to overly focus on the negatives of the race he plays?</div>
Yeah but Idra is one player. To minimize biases, you need to take an average. However whose opinion would you pick over the average opinion of the best in Starcraft to balance it? I don't see any other group who would be more qualified to talk about it.

<div class='quotetop'>QUOTE (puzl)</div><div class='quotemain'>Absolutely.. I'm not trying to explain away any of the problems.. I'm just trying to explain that there is (still) no anti-competitive agenda. If you look at Cory's excellent reply in the topic on skill based movement he elaborates on this point. They have identified that there are issues.. they have expressed a desire to fix them.. but there are only so many hours in the day.</div>
<div class='quotetop'>QUOTE (Fana)</div><div class='quotemain'>It's telling that not even once has any of them replied to my posts on the subject, which tend to be on point as I like to think I know a thing or two about the subject. While the reason for it is unclear, it seems abundantly clear that the they've taken the stance of the "casual players" over the "competitive players" regardless of the arguments made for or against it. Does that mean there's a conspiracy against competitive players? No, but there's an obvious bias, and that bias leads to this:</div>
I've come to trust Fana's opinion on most things related to NS, as I should, but I disagree on this, at least semantically. I don't think devs have the same intuitive understanding as Fana and many other competitive players do, which leads them to different conclusions. I don't know if I'd describe that as a bias though. I'd also like to emphasize, I never thought devs had any intentional ill-bearing towards us (apart towards manners which is understandable).

The problem are some value disagreements but I think most disagreements are not about values. They seem so <i>prima facie</i>, but in fact are not, and are masked by veil of ignorance, for lack of better words. This applies to the previous Idra quote aswell. That Fana's comment on fade's movement system is a good example. Some people think that reintroducing <i>3.2</i> blink would be making a choice between favoring competitive groups in detriment to casual gaming, while I honestly believe it is not. It'd add to both levels of gaming.

Quite frankly, words are cheap. I think it is important to have good intuitive understanding of the topic you are talking about. Words are are just smoke and mirrors. They aren't Q.E.D. like math, but rely on very abstract premises and causality, something whose <a href="http://en.wikipedia.org/wiki/Operationalization" target="_blank">operationalizability</a> isn't obvious. Something like art criticisim.

About the fade change you were talking about; I remember interviewing Tane as a playtester about this and he disagreed with it like did some others. At least now safely in hindsight I think it was a good update, not only because it could have been scripted and made bad fades much better, but I think it also added some depth <i>maybe</i> against jetpacks, just like allowing binding wheel to mouse makes bunnyhop easier, but adds more depth as it you don't have to struggle with every jump [hah, I'd want to see the <i>fiasco</i> of competitive players whining about game being too hard if that were disallowed]. Ultimately there're few things wrong with 3.2. What happened is history, but I'm very happy with the end product. I thank puzl, among many others, for this.

Harimau · February 2012

<div class='quotetop'>QUOTE (elodea @ Feb 12 2012, 10:24 PM) <a href="index.php?act=findpost&pid=1902343"><{POST_SNAPBACK}></a></div><div class='quotemain'>I actually feel with rfk gone, super-unit scarcity is back but more in the sense that pres is now valuable and tradeoff decisions must be made where you effectively have to decide what lifeform you want at gamestart. I like how it feels currently in that it strikes a nice balance between super-unit scarcity and new player frustration at not being able to experience the 'super-unit' for themselves. You cannot gorge + cyst and expect to then fade/onos in a timely fashion. Neither can you lerk and then fade/onos. And saving for fade definitely locks you out of going onos within a reasonable timeframe. Although the lifeforms come in waves, i certainly do not agree that super-units are too frequent.

Also, using the level of 'scarcity' in NS1 as a yardstick makes no sense when it was hugely tied to number of players.</div>
That's a great thing, then, the removal of RFK.
Whatever the current situation is, the ideal to strive for would be:
<i>The power of super-units should be inversely proportional to their abundancy and frequency.*</i>
And <b>raw cost</b> often simply doesn't cut it in terms of limiting abundancy and frequency.
I use super-units as a loose term, as it can refer to literally any class on either team, where you are "paying for power".
The biggest issue currently facing NS2 is the lack of <b>opportunity cost</b>, and introducing more opportunity cost would go a long way to limiting the abundancy and frequency of super-units.
If the removal of RFK has moved towards this ideal, then that's a great thing, but we should <i>continue</i> to strive towards it.
*Of course, when you apply the philosophy of "everyone should be able to have a go at every class", then naturally the power of super-units has to be reduced, to keep the game both balanced (balanced at the top-level) AND fun (balanced at all levels).

Standard NS1 games were balanced for 6v6. In a standard game, there was absolutely super-unit scarcity. (So in those terms, yes, it makes perfect sense.)
Every other game-size is therefore inherently less balanced. Actually that was one of the big reasons for the introduction of the TRes-PRes system: scaling the game for other player counts. It is not, however, working as well as hoped.

<div class='quotetop'>QUOTE (Tweadle @ Feb 13 2012, 10:28 AM) <a href="index.php?act=findpost&pid=1902499"><{POST_SNAPBACK}></a></div><div class='quotemain'>Not quite sure where we disagree.

I don't want balance at the expense of fun and I also agree with your point on understanding/knowledge. However, this knowledge can come about intuitively (mechanics that make sense) *or* forcefully (tutorials). Natural Selection maybe suffered from a lack of both and I've always advocated a comprehensive tutorial because it has no drawbacks besides the time taken to make it. Intuition is important too but is attached to various other issues which I think are important to discuss in their own right. The case for framing things in terms of knowledge is all very well and good when there is something to actually know. Right now, there isn't and I'm saying there should be. Wallhopping is, so far, unsuccessful and I see its limitations making it super-hard to adequately replace bunnyhopping.

I also never claimed that balance should only be achieved at any particular level, nor that its possible to achieve at all levels. I do, however, think that a solid movement system which scales somewhat with aim is crucial in alleviating these problems. I pretty much entirely agree with puzl, for the record (even on sometimes balancing with pubbers in mind), but I'm not quite as elegant or as eloquent on the subject.

I agree here too. I'm not trying to make a case for ignoring what is fun but this discussion started as a balance issue and I maintained that theme. It was only when rebirth started pointing out some difficulties in reconciling balance and fun and why fun should take priority that I chimed in. Perhaps I should have been clearer, though. It was not necessarily a flaw with the logic of rebirth's spiel, but with his use of it. I think he takes his philosophy of game-design too far and comes to a conclusion that has too little an emphasis on balance, certainly in the case of aim/movement scalability.

Firstly, I think that his philosophy works very well as a general concept but particular scenarios sometimes require a stronger focus on balance than others - the marine/skulk relationship being one of them. This is predominantly due to how often these two classes interact at all stages of the game. I truly believe that the skulk requires some sort of mirror to the scaling of skill-to-aim marines and that it needs to be factored in the design decision at one of the early passes. I would even go so far as to suggest <b>starting</b> with this in mind (because its such an elusive, but important, goal) and <b>then</b> work on getting it fun. We do disagree on the prudency of this ordering though, I guess.

Secondly, I don't think our talk of balance has done anything to belittle the 'fun' argument in the first place. Rewarding movement mechanics need not be boring or tedious and I think bunnyhopping was neither. The reason I didn't go down that tack was because this thread was focused on the importance of balance and because the subject of whether it is fun or not is pretty polarised between those who can do it (well) and those who can't. In fact, you could perceive the demand for the reintroduction of bunnyhopping or its equivalent not simply in terms of balance but in terms of fun. After all, we could be arguing for major changes in the skill ceiling of the gorge, lerk or fade to compensate for marine aim. But this would leave the whole skulk experience really dull and unforgiving, with all skulks at top levels avoiding marines in favour of eating rts or something which is a thankless task.</div>
1) I was referring to intuitiveness (which is a subset of the knowledge/understanding argument, and therefore affects balance at all levels of play except at perfect knowledge/understanding). You had said that the discussions on intuitiveness were irrelevant to the discussions of balance.

2) And then I wanted to make a point of not disregarding fun when considering balance AND make a point that there are concepts of "fun" that are common to most people. As much as fun is subjective, rebirth was simplifying and distinguishing the two main motivators in a competitive online multiplayer game, the two main sets of "fun": "playing for fun" - in which case the fun comes during the activity; and "playing to win" - in which case the fun comes at the end of the activity (by succeeding: whether that is a single engagement or an entire game). In a way, fun and balance (or rather, people that support one over the other) are contrary arguments, even if they are not mutually exclusive when applied in practice: someone who plays for fun cares less about top-level balance, and so talks in those terms (something seems overpowered or unfun to the player because of lack of understanding and a different motivation); someone who plays to win only cares about balance at the top-level because they get their greatest satisfaction from winning (something that is a boring or unfun mechanic is still balanced at the top-level, so the player doesn't care so much that it's boring because they have a simpler motivation). Yes, there are mechanics that are both always fun and perfectly balanced (that is the ideal we should strive for), just as there are players that both "play for fun" AND "play to win" - these players play games with always-fun mechanics that are balanced at the top-level... but a lot of "fun" is a direct result of perceived balance (something "overpowered" is unfun), so we return to the argument... balance at every level of play. And where you cannot or should not balance (e.g. something <b>has to be</b> "overpowered" for whatever reason), you educate (e.g. teach the player that it is not in fact overpowered when considered in terms of the whole): but ideally, you would have built the game from the ground up with a minimal requirement for player education, using either simple or intuitive game mechanics.

If I may generalise, half of "fun" in a gaming activity is often just balance at every level, because half of "fun" is often just performing an activity with the feeling that you have "a fighting chance". (The other half is simply joy in the activity itself.) That is where the distinction between "fun" (the former half, that is) and "balance" lies - "balance" (at the top level) means that everyone has a fighting chance, but <span style="color:red">only at the top level</span>. If the only people having fun are people at the top level of understanding, then either everyone should always somehow be at the top level (maybe your game lacks depth), or there is something wrong with your game as a consumer product.

Tweadle · February 2012

Well in response to 1), I concede that intuition should be considered but I would reiterate my concerns that in this case, intuition counts for diddly squat if its not balanced because the marine/skulk relationship is so overwhelmingly common. In any case, nobody in this thread <b>wants</b> unintuitive mechanics and when I said it was irrelevant, it was more to say that this discussion isn't going to benefit from stating the bleeding obvious. I thought it was very interesting until it got sidetracked by fun/intuition which nobody denied was important in the first place.

Regarding 2), I would never be so silly as to disregard fun. I'm very sceptical as to your extrapolations of the 'play to win' and 'play for fun' model, though. The more its taken to its logical conclusions, the less I see it as a useful representative tool. Players at all levels of the game and on any end of the win/fun spectrum care about whether is something is balanced <i>and</i> whether something is fun.

More generally, I think you and rebirth have just espoused some nice theories of general game design and blindly and broadly applied it to all circumstances of this game. It's not that I support balance over fun or vice versa, nor that I feel the theory is inherently broken, it's merely that I feel that this instance (where number crunching won't provide any answers and only a really solid movement mechanic designed from scratch will suffice) requires a real emphasis on the relative scalability of aim and movement, as I've always said.

Harimau · February 2012

The thing is, a lot of the "scalability" that people are discussing with regards to movement is about making the effectiveness decrease with lower skill, rather than the effectiveness increase with higher skill. It's a subtle, but important, distinction. I would say that something like bunnyhopping is a case of the former rather than the latter, because: On a level playing field, the perfect skulk will always lose to a perfect marine, because the marine can kill the skulk before the skulk can reach him*. Even if the skulk increases in speed, the marine will most likely deal the damage fast enough to kill him. Even if the skulk moves erratically and unpredictably, the marine has perfect aim, so he will kill him. So unless you can close the distance instantly (like the fade can), speed doesn't play much of a factor: with less-than-perfect marines, skulk speed may indeed play a factor. Assuming that we're balancing at the "top"-level ("good" marine aim and maximum skulk speed), what bunnyhopping would do is actually penalise those not good at bunnyhopping with less than maximum speed and therefore a much higher chance of dying before reaching/killing a marine. If we balanced at the "bottom"-level ("good" marine aim and minimum skulk speed) then bunnyhopping to gain speed would in fact be a bonus for those that do it, because they gain an advantage over "good" marines. Or you could just increase the speed wholesale and balance at the "top"-level. Honestly, skulk movement has so little impact on a perfect marine that I don't know what you're hoping for out of the movement mechanics for balance. Tactical and environmental factors play a much larger role. Is it just the feeling of "I've gotten better"? Now <b>that</b> has nothing to do with balance.

*Crunching some numbers:
Perfect marine fires 50 shots in ~3.5 seconds (it's actually faster than this). It takes 1/5 of those shots (it's actually a bit less) to kill a skulk, so 0.7 seconds. The skulk at maximum speed moves, let's say 8.5 metres a second, that means that in 0.7 seconds the skulk moves roughly 6 metres. That means that if the perfect marine engages the skulk at a distance greater than or equal to 6 metres, he can kill the skulk before the skulk can even scratch him.
Now let's consider that both players are at zero engagement range. Again, 0.7 seconds for the marine to kill the skulk. The attack cooldown for skulk bite is, iirc, 0.35 seconds. It takes three bites to kill a marine (75+75+75 damage vs 160 health), and that (bite->0.35->bite->0.35->bite) takes 0.7 seconds in total. Now, you see, <b>that</b> is balanced - but movement speed had nothing to do with it.

Tweadle · February 2012

There's little value in comparing a skulk's movement to a marine with perfect aim because it doesn't exist in the real world. It's another example of where theories start to break down at extreme levels. It becomes equally pointless to theorise the clash at 100% imperfect skill. Your distinction between effectiveness decreasing with lower skill and increasing with higher skill is pretty moot because of this. I actually struggle to see the difference at all and could frame all your points either way. For all realistic purposes, they are one and the same. The fact is, when we say balance at the top-level, what we are saying is that there needs to be balance at the highest level that players have actually reached. Human input, after all, is what makes a game.

The scalability we are discussing takes into account that people are good, but not perfect. I find it amazing that you don't see movement mechanics having a place in the scope of balance because it comes from the wild assumption that marines are perfect aimers and that we should balance for that possibility/inevitability. All evidence suggests that even at a high level, bunnyhopping places an important role. The beauty of it is that it's not the sole option of attack but part of a complement of choices. Skulks can ambush, wigglewalk, wallhop or bunnyhop and different situations call for different strategies. You might even find that you only want to bunnyhop because you need to get somewhere faster, where you then proceed to adopt a combat-effective ambush position.

I think you're condensing balance into numbers without a perfect conception of the maths involved. In reality, you might find that someone with an average accuracy of 80% versus a straight-line skulk could drop to 40% if some less predictable path is taken. A further advantage would be conferred by the shorter time taken to reach the opponent. On top of this, certain players might find that particular movement pattern predictable and not see quite as big a drop in bullets hit. As soon as you start taking percentages of in-game values, you can begin to see why a movement system that affects those percentages has an affect on balance.

Assigning aim 100% and performance drop 0% to your imaginary values is just counter-productive. What if I assumed that a total mastery of skulking could confer infinite movement speed? Your maths would suggest that this would be numerically balanced but the message would be pragmatically useless. You might as well try and divide balance by zero.

I'm not sure if this graph helps at all to illustrate what I'm getting at. It's just a crappy approximation of what I feel we have currently but I think it puts across the idea semi-ok:

<a href="http://imageshack.us/photo/my-images/710/marinevsskulk.jpg/" target="_blank"><img src="http://img710.imageshack.us/img710/3451/marinevsskulk.jpg" border="0" class="linked-image" /></a>

Balance issues arise with this model because skulks have no way of competing with marines who have invested the similar time (insert: skill) at high levels. The higher the skill, the greater the imbalance. That's why I'm arguing that, while any particular movement mechanic isn't inherently imbalanced, it has a direct impact on balance because performance gain differences start to get crazy. I'd like to see a graph that narrows the gap between the two or at least mitigates the snowballing effect of increased skill in the marine/skulk match up.

Bacillus · February 2012

<div class='quotetop'>QUOTE (Tweadle @ Feb 13 2012, 06:42 PM) <a href="index.php?act=findpost&pid=1902686"><{POST_SNAPBACK}></a></div><div class='quotemain'>Assigning aim 100% and performance drop 0% to your imaginary values is just counter-productive. What if I assumed that a total mastery of skulking could confer infinite movement speed? Your maths would suggest that this would be numerically balanced but the message would be pragmatically useless. You might as well try and divide balance by zero.</div>
Just to illustrate this a bit more, some perfect play scenarios of NS1 would include:

A vanilla marine shooting down 7 skulks before forced to reload. The first two skulks would die almost instantly to the pistol.

A vanilla marine soloing a fade without having to reload as long as we add perfect medpacking to the equation.

No marine dying ever due to perfect commander medpacking.

Catpacks being probably the best commander ability ever as the increased DPS would allow the commander to save up in medpacks.

---

As far as I've understood, even the Brood War professional gamers agree that stuff like Queens could be very useful in certain situations. However, nobody has really been able to establish such control that investing in them would consistently pay off apart from some pretty niche scenarios. Devastating units such as reavers and lurkers would also be relatively little value against perfectly microed unit stacks.

fanatic · February 2012

<div class='quotetop'>QUOTE (Jiriki @ Feb 13 2012, 07:02 AM) <a href="index.php?act=findpost&pid=1902526"><{POST_SNAPBACK}></a></div><div class='quotemain'>I've come to trust Fana's opinion on most things related to NS, as I should, but I disagree on this, at least semantically. I don't think devs have the same intuitive understanding as Fana and many other competitive players do, which leads them to different conclusions. I don't know if I'd describe that as a bias though. I'd also like to emphasize, I never thought devs had any intentional ill-bearing towards us (apart towards manners which is</div>
Good post man, on point as usual. I think you've misunderstood what I meant though, or perhaps my use of the word bias muddled my meaning. The idea wasn't that the devs are necessarily consciously picking the "casual option" over the "competitive option", but that that is the end result of their decisions so far. If you compare the common "battlegrounds" between the playerbases, the casual players come out as the overwhelming winners. Whether that is a result of planned development or just a "random" result of many individual decisions is irrelevant as far as I'm concerned.

Amusingly, if you actually do compare the perfect marine vs. the perfect skulk (for the sake of argument let's ignore how ridiculous that is), it actually shows why a deep movement system is required for the skulk. The reason why the marine will destroy the skulk in such a scenario, is because the potential for improvement with the skulk is so much smaller than for the marine. There is actually no absolute cure for this, but a movement system that rewards time spent practicing by increasing gains in combat efficiency helps a lot to close that gap for the humanly obtainable level of skill.

MuYeah · February 2012

Fana, Tweadle, Bacillus, Jiriki <i>and</i> Puzl all posting in the same thread: NS HEAVEN.

Puzl and Jiriki, especially, please post more often. I really enjoy reading your points of view.

On topic:

I'm pretty sure Flayra & the guys love a bit of the old competitive gaming. I wish there was some way for them to get involved more without it seeming partisan. The demo of Flayra shoutcasting the North America vs Europe NSv1.04 game (that EU won with a 2gorge/2skulk 1hive all-in on ns_caged. BAM.) was literally what made me interested in organised play. I think it was the idea that if the creator of the game was directly involved with it, it had to be good. Right? <i>Right?!</i>

I think the best thing anyone can do is LUA mod the crap out of what you want changed, like the fade momentum mod. Props on that one by the way, Yuuki.

edit: VVVVVVVV

<div class='quotetop'>QUOTE (dux @ Feb 14 2012, 12:28 AM) <a href="index.php?act=findpost&pid=1902779"><{POST_SNAPBACK}></a></div><div class='quotemain'>We all remember your brilliant blocking in west access Fana, with the subsequent panic and dying on waste ramp ;) I think the Fade of the match went to Oetel, though. For his 15 seconds (or around about) of epic fade time.

Everything has been pretty well covered already so not much for me to input here.</div>

I got hand grenaded as a fade in an ENSL match once. <i>By a german</i>. B(

Obviously the counter to fades is grenades.

dux · February 2012

<div class='quotetop'>QUOTE (fanatic @ Feb 10 2012, 08:10 PM) <a href="index.php?act=findpost&pid=1901894"><{POST_SNAPBACK}></a></div><div class='quotemain'>I guess you didn't watch the 2004 ClanBase Opencup final between EoD and Knife where some of the best fades in Europe died over ten times to lmgs in a single round.</div>

We all remember your brilliant blocking in west access Fana, with the subsequent panic and dying on waste ramp ;) I think the Fade of the match went to Oetel, though. For his 15 seconds (or around about) of epic fade time.

Everything has been pretty well covered already so not much for me to input here.

fanatic · February 2012

<div class='quotetop'>QUOTE (MuYeah @ Feb 14 2012, 12:51 AM) <a href="index.php?act=findpost&pid=1902768"><{POST_SNAPBACK}></a></div><div class='quotemain'>Fana, Tweadle, Bacillus, Jiriki <i>and</i> Puzl all posting in the same thread: NS HEAVEN.</div>
...and now you and dux have posted as well. This is starting to get real cozy. Definitely want to see puzl post more often, but he doesn't seem to be involved at all.

<div class='quotetop'>QUOTE (MuYeah @ Feb 14 2012, 12:51 AM) <a href="index.php?act=findpost&pid=1902768"><{POST_SNAPBACK}></a></div><div class='quotemain'>I'm pretty sure Flayra & the guys love a bit of the old competitive gaming. I wish there was some way for them to get involved more without it seeming partisan. The demo of Flayra shoutcasting the North America vs Europe NSv1.04 game (that EU won with a 2gorge/2skulk 1hive all-in on ns_caged. BAM.) was literally what made me interested in organised play. I think it was the idea that if the creator of the game was directly involved with it, it had to be good. Right? <i>Right?!</i></div>
That was the best publicity the comp side of the game ever had. I remember following it through an IRC bot because HLTV didn't work at all that version and even then it was a lot of fun.

<div class='quotetop'>QUOTE (MuYeah @ Feb 14 2012, 12:51 AM) <a href="index.php?act=findpost&pid=1902768"><{POST_SNAPBACK}></a></div><div class='quotemain'>I got hand grenaded as a fade in an ENSL match once. <i>By a german</i>. B(</div>
You and pantsu should start a club.

<div class='quotetop'>QUOTE (dux @ Feb 14 2012, 01:28 AM) <a href="index.php?act=findpost&pid=1902779"><{POST_SNAPBACK}></a></div><div class='quotemain'>We all remember your brilliant blocking in west access Fana, with the subsequent panic and dying on waste ramp ;) I think the Fade of the match went to Oetel, though. For his 15 seconds (or around about) of epic fade time.</div>
Mate it was obviously to lull them into a false sense of security.

swalk · February 2012

Still, two of the most major balance issues, that still havent got much, if any, attention by the devs:
<a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=115515" target="_blank">Nano Shield, Beacon, Building energy doesn't scale with resource situation</a>
That results in certain buildings being spammed to gain more use of the abilities.
Why not make the amount of abilities dependant on the resource situation?
It's also currently resulting in the alien commander not having anything to use his personal resources on, which means yet another lifeform on the field later in the game, or more hydras.
<a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=115904" target="_blank">Random spawns</a>
The current implementation of random spawns with both teams being able to start at the same locations in tram and summit is not feasable.
It results in way too many short games, due to short base-to-base distances, and most of the locations heavily disfavours marines(vents, LOS blocking and tight spaces).
When random spawns were implemented, Hives and CC health was nerfed to get shorter games, this needs to be reverted, asap.
Mineshaft have the vent in marine base issues as well. (Watch the video to see an example of gameplay)
But at least both teams are not able to spawn at the same locations in that map.

This is a perfect example of why CC/hive hp nerfs and vents in marine starts are a bad design(go to 3 minutes):
<center><object width="450" height="356"><param name="movie" value="http://www.youtube.com/v/VOhmPGddK1k"></param><embed src="http://www.youtube.com/v/VOhmPGddK1k" type="application/x-shockwave-flash" width="450" height="356"></embed></object></center>
Add that the aliens could have spawned even closer to the marines, and you unbalanced the game even further..
..oh wait, they already can.
The only way I can see to properly balance the base-to-base distances on mineshaft, is to make static spawns in Operations(m) and Cave(a).

MuYeah · February 2012

Neither of those are inherently unbalanced, though.

What if the abilities tied to energy are designed to be used once every X seconds? You can balance them by changing cost, regen rate and max energy per building. If anything it adds a little diversity and makes the game more bearable for a team who just got knocked back a bit in the res game.

Random spawns an unbalanced right now but the capacity is there for mappers to fix them.

IronHorse · February 2012

swalk:
base to base distances are one issue. vents are another.

elodea · February 2012

<div class='quotetop'>QUOTE (swalk @ Feb 14 2012, 12:08 PM) <a href="index.php?act=findpost&pid=1902788"><{POST_SNAPBACK}></a></div><div class='quotemain'>This is a perfect example of why CC/hive hp nerfs and vents in marine starts are a bad design(go to 3 minutes):</div>

Wait till NS2hd posts 2/2 :). That vent in drill repair is certainly annoying but it can be dealt with as marines because its a ground based vent as opposed to a wall/celing vent. (side note: we did that skulk rush through cart tunnel coz i saw you guys do it in an ensl gather and how effective it was :p)

internetexplorer · February 2012

<div class='quotetop'>QUOTE (rebirth @ Feb 12 2012, 09:47 AM) <a href="index.php?act=findpost&pid=1902331"><{POST_SNAPBACK}></a></div><div class='quotemain'>So you are trying to fix an balance issue that results from asymmetric team mechancis by making both teams even more similar?</div>

yes, because in the case of the onos the marines currently have no equivalent

not only that, but the equivalent to the onos (hmg+heavy or exosuit or whatever they end up calling it) is in the works and has been planned for a very long time (195 is not content complete!)

should we fix the incomplete game by making the teams more dissimilar for the sake of being dissimilar? please write a 2000 word essay in response.

<div class='quotetop'>QUOTE (Harimau @ Feb 10 2012, 04:09 AM) <a href="index.php?act=findpost&pid=1901634"><{POST_SNAPBACK}></a></div><div class='quotemain'>I liked the munching RTs example that rebirth wrote about, but then you have to consider that it's exactly the same for both teams and for every unit, so this is, inherently, balanced. Therefore, it's irrelevant in terms of balancing the game, because assuming that teams are equally skilled/knowledgable (a necessary assumption) and have the same motivations, both teams will either munch RTs or not munch RTs.</div>

This is a really good example to revisit several times. Everyone involved is applying their personal skillset (commanders scouting/chatting to team, using nano/umbra, players attacking/defending res, smarter players setting up/baiting recycles...). The 'numbers' involved are easy to balance (skulk damage vs rifle damage, mac repair vs gorge healing, extractor health etc), so what it comes down to is the players.

It simply wouldn't make sense to look at how confused new players react to their resources being attacked (i.e. they fail to defend them in time, if at all). Anyone who has played more than half an hour probably knows that it's important to attack/defend resources. What happens after that, though? People get better at defending and attacking things, and play against each other. The competition drives them to continue learning and developing the same seemingly simple part of the game.

It wouldn't make sense to add new mechanics here when it's already straightforward for players to learn how this part of the game works, and get better at it. When someone eventually makes a thread that says "Today I played a game where I lost a res tower and I was sad, so buff marines!" it is everyone's duty to read it and say 'no, you're wrong' because of the concepts I just explained.

The interesting thing is that you can take just about any concept in the game and present it as a very simple thing for players to develop (because it's a 2 team game with 2 races and no mirror matches, and tech trees that 'match' at certain points). A great example is the Fade. New players who start right now get smacked six ways from sunday by fades, then go on the forums and say the fade is overpowered. Those of us who have been playing for longer can see that the fade is<i> at least</i> much weaker than it used to be. Those of us who combine shotguns and flamethrowers (the fade's natural enemy) understand that fades are killable in a deterministic, simple way. The only remaining step in the process is for people to practice their shotgun/flamethrower against fades, and for other people to practice their fade against shotguns/flamethrowers. If a patch comes out tomorrow that weakens the fade further, it will be an example of a knee-jerk reaction to inexperienced players crying. That's the kind of thing Idra and Artosis are speaking against when they say you should balance a game for a high skill level. It's still possible that an imbalance exists in the fade, but there's no way we have *truly* found it yet - the patch only came out 4 days ago, and 3/4 of the people playing are very, very new to the game.

A better way is to say that you balance for the "pursuit" of a higher skill level. Even if people are bad at the game now, there should be room to grow (rather than fisher price mechanics that play the game for you and don't let you grow).

swalk · February 2012

<div class='quotetop'>QUOTE (MuYeah @ Feb 14 2012, 02:47 AM) <a href="index.php?act=findpost&pid=1902796"><{POST_SNAPBACK}></a></div><div class='quotemain'>Neither of those are inherently unbalanced, though.

What if the abilities tied to energy are designed to be used once every X seconds? You can balance them by changing cost, regen rate and max energy per building. If anything it adds a little diversity and makes the game more bearable for a team who just got knocked back a bit in the res game.

Random spawns an unbalanced right now but the capacity is there for mappers to fix them.</div>
Yes, both of them are large imbalances/flaws in the current gameplay.
Sure, you could balance abilities to be tied around a cooldown/energy whatever, but seeing as the alien commander needs alot to spend personal resources on, I see this as the best solution. It will create alot more trade-offs res-wise, and it will cause less hydras/lifeforms.
If anything, it really imbalances the game not having the alien commander having to use his personal resources.
Also, the game is an RTS/FPS, which means if you can't keep up with the resources, you should on your way to losing.
Energy powered abilities/structures goes directly against that. Changing it to personal res(and teamres for Distress Beacon) fixes it.
<div class='quotetop'>QUOTE (ironhorse @ Feb 14 2012, 02:51 AM) <a href="index.php?act=findpost&pid=1902797"><{POST_SNAPBACK}></a></div><div class='quotemain'>swalk:
base to base distances are one issue. vents are another.</div>
I completely agree?

<div class='quotetop'>QUOTE (elodea @ Feb 14 2012, 03:20 AM) <a href="index.php?act=findpost&pid=1902807"><{POST_SNAPBACK}></a></div><div class='quotemain'>Wait till NS2hd posts 2/2 :). That vent in drill repair is certainly annoying but it can be dealt with as marines because its a ground based vent as opposed to a wall/celing vent. (side note: we did that skulk rush through cart tunnel coz i saw you guys do it in an ensl gather and how effective it was :p)</div>
Yes, you can deal with it by having most of your marines near the base.
But that forces the marines to be even more defensive than they really should. They are fighting against time, resources, and the aliens.
But when you run into the scenario where you have vents in your base, vents around your first extractor points, marine get in trouble.
It is a balance issue that should be dealt with, you didn't see many NS1 maps with vents in marine starts because of this.

Harimau · February 2012

<div class='quotetop'>QUOTE (Tweadle @ Feb 14 2012, 02:42 AM) <a href="index.php?act=findpost&pid=1902686"><{POST_SNAPBACK}></a></div><div class='quotemain'>There's little value in comparing a skulk's movement to a marine with perfect aim because it doesn't exist in the real world. It's another example of where theories start to break down at extreme levels. It becomes equally pointless to theorise the clash at 100% imperfect skill. Your distinction between effectiveness decreasing with lower skill and increasing with higher skill is pretty moot because of this. I actually struggle to see the difference at all and could frame all your points either way. For all realistic purposes, they are one and the same. The fact is, when we say balance at the top-level, what we are saying is that there needs to be balance at the highest level that players have actually reached. Human input, after all, is what makes a game.

The scalability we are discussing takes into account that people are good, but not perfect. I find it amazing that you don't see movement mechanics having a place in the scope of balance because it comes from the wild assumption that marines are perfect aimers and that we should balance for that possibility/inevitability. All evidence suggests that even at a high level, bunnyhopping places an important role. The beauty of it is that it's not the sole option of attack but part of a complement of choices. Skulks can ambush, wigglewalk, wallhop or bunnyhop and different situations call for different strategies. You might even find that you only want to bunnyhop because you need to get somewhere faster, where you then proceed to adopt a combat-effective ambush position.

I think you're condensing balance into numbers without a perfect conception of the maths involved. In reality, you might find that someone with an average accuracy of 80% versus a straight-line skulk could drop to 40% if some less predictable path is taken. A further advantage would be conferred by the shorter time taken to reach the opponent. On top of this, certain players might find that particular movement pattern predictable and not see quite as big a drop in bullets hit. As soon as you start taking percentages of in-game values, you can begin to see why a movement system that affects those percentages has an affect on balance.

Assigning aim 100% and performance drop 0% to your imaginary values is just counter-productive. What if I assumed that a total mastery of skulking could confer infinite movement speed? Your maths would suggest that this would be numerically balanced but the message would be pragmatically useless. You might as well try and divide balance by zero.

I'm not sure if this graph helps at all to illustrate what I'm getting at. It's just a crappy approximation of what I feel we have currently but I think it puts across the idea semi-ok:

<a href="http://imageshack.us/photo/my-images/710/marinevsskulk.jpg/" target="_blank"><img src="http://img710.imageshack.us/img710/3451/marinevsskulk.jpg" border="0" class="linked-image" /></a>

Balance issues arise with this model because skulks have no way of competing with marines who have invested the similar time (insert: skill) at high levels. The higher the skill, the greater the imbalance. That's why I'm arguing that, while any particular movement mechanic isn't inherently imbalanced, it has a direct impact on balance because performance gain differences start to get crazy. I'd like to see a graph that narrows the gap between the two or at least mitigates the snowballing effect of increased skill in the marine/skulk match up.</div>
Perfect aim is entirely attainable. The only thing the player needs to do is move their crosshair a few pixels at a time to stay on target. Infinite speed is not entirely attainable because of engine limitations and designed limitations. With limited speed, an analogy for aim would be if marine turn-speed was limited.

The point I've made that you're missing is that ranged vs melee is inherently skewed against the melee. That's why people don't still use swords in war.
I'm not sure what point Bacillus was making, but his example actually supported my argument very well: that one very good marine can take on many skulks (assuming infinite ammo) very easily. If the skulks had moved faster, his performance would only have dropped a little. If the skulks had moved more erratically, his performance would only have dropped a little. This is irrespective of the skulk's actual skill.

The point is that variable speed with skill has little to no impact on the marine's effectiveness. Increasing the speed drastically across the board may reduce the engagement times and give the skulk a better fighting chance, but that causes a host of other issues. Variable speed with skill doesn't reward high-skill players so much as it penalises low-skill players, because melee is inherently disadvantaged with respect to ranged so skill only decreases the disadvantage. Things that do highly affect the marine-skulk balance are tactical and environmental factors. You gave the example of using increased speed to reach an ambush location - but once again, you are limiting options to low (technical) skill players rather than rewarding high (technical) skill players.

So in a way, what I'm saying is that, with the same optimal parameters (inc. maximum speed), the maximum performance (the flat portion on your graph) of each player never changes. The only thing you really do with a movement system is change the lead-up (the curved portion on your graph) to the skulk's maximum performance. Increasing the optimal parameters (the height of the flat portion of the graph) is a top-level decision and can be considered independent of the movement system (because, for instance, you could implement the movement system which would increase the maximum speed, then you could scale it all back down). So this idea that adding a movement system is going to change the balance (the height of the performance at the optimal parameters) is, basically, an illusion.

<img src="http://i.imgur.com/qZmBZ.png" border="0" class="linked-image" />

Now that doesn't mean you shouldn't add a movement system - I think you should - but using balance as the reason (rather than flavour, fun, or a personal sense of progression) is erroneous.

Something that would actually help aliens against marines would be to add a visible laser to every weapon. It would "lower the skill ceiling" (but not lower the maximum performance) by removing the element of determining a marine's orientation, but would make it so that dodging marine weapons would become far more viable for many more players. This is not a sincere suggestion, however.

<div class='quotetop'>QUOTE (internetexplorer @ Feb 14 2012, 11:35 AM) <a href="index.php?act=findpost&pid=1902829"><{POST_SNAPBACK}></a></div><div class='quotemain'>It simply wouldn't make sense to look at how confused new players react to their resources being attacked (i.e. they fail to defend them in time, if at all).</div>
This is a good point, but I think this is something that can actually be improved upon, so that it is more obvious and intuitive: Put the number of resource towers and your average income on the scoreboard, and in the game world put a +1 PR above RTs every time you are awarded with a resource and synchronise that with the personal resource counter. This in no way affects the skill required for defending RTs, it just helps perfectly new players understand that they are gaining resources from these towers (and therefore should defend them).

Tweadle · February 2012

<div class='quotetop'>QUOTE (Harimau @ Feb 14 2012, 01:44 AM) <a href="index.php?act=findpost&pid=1902889"><{POST_SNAPBACK}></a></div><div class='quotemain'>The point is that variable speed with skill has little to no impact on the marine's effectiveness.</div>
This is so outrageously wrong.

I think, when it comes down to it, we'll never agree because of your assumption of marine effectiveness. I <b>don't</b> think that perfect aim is attainable and I've been playing the game for almost a decade now. Sadly, I'm probably one of the best players NS1 has nowadays and even I suffer from a drop in performance against unpredictable and fast skulk movement. Bacillus wasn't trying to support your theory; he was trying to illustrate how ridiculous it is to follow your argument to its logical conclusion. Your reasoning is correct, but your premise is amazingly incorrect.

Harimau · February 2012

@Tweadle: You may not have seen this because of an edit:
<div class='quotetop'>QUOTE (Harimau @ Feb 14 2012, 02:44 PM) <a href="index.php?act=findpost&pid=1902889"><{POST_SNAPBACK}></a></div><div class='quotemain'>So in a way, what I'm saying is that, with the same optimal parameters (inc. maximum speed), the maximum performance (the flat portion on your graph) of each player never changes. The only thing you really do with a movement system is change the lead-up (the curved portion on your graph) to the skulk's maximum performance. Increasing the optimal parameters (the height of the flat portion of the graph) is a top-level decision and can be considered independent of the movement system (because, for instance, you could implement the movement system which would increase the maximum speed, then you could scale it all back down). So this idea that adding a movement system is going to change the balance (the height of the performance at the optimal parameters) is, basically, an illusion.

<img src="http://i.imgur.com/qZmBZ.png" border="0" class="linked-image" />

Now that doesn't mean you shouldn't add a movement system - I think you should - but using balance as the reason (rather than flavour, fun, or a personal sense of progression) is erroneous.</div>

Whether or not I am wrong for assuming that perfect aim is attainable, the above cannot be disputed.

MuYeah · February 2012

You're falling into the trap of trying to fit reductionist mathematical models to complex systems. To be honest, I think your point has value, Harimau, it's just you're typing a convoluted thesis full of indecipherable jargon every time you discuss it. We're not really interested in maximum performance anyway. It's more the fact that it should never be attainable.

I suggest re-reading Puzl's comments on the NS 3.2 fade changes to everyone.

Wilson · February 2012

Harimau, perfect aim isn't attainable. While you are correct in saying you don't change the maximum performance, it is impossible to perform a mechanic perfectly 100% of the time. Looking at just the maximum theoretical performance isn't a very good idea. In theory in an FPS game if you had perfect aim then you would never die. The point in making the lead up to the potential maximum something that is harder to master is that it rewards players more for their abilities. Aiming as a mechanic is a nice example of this. The more you practice, the better your aim gets, the more you are rewarded by performing better in the game.

By making a mechanic easier to master you are decreasing the difference between the best players and the worst and reducing the effect of mastering the mechanic has on performance. This decreases the depth of the game as you no longer have wide varying skill levels and it turns more into you either can or can't use that particular skill.

The great thing about aiming as a mechanic is that there is a broad spectrum of potential skill levels and you can always improve since no human can ever have 100% perfect aim. If I aim a little bit better than you then I get rewarded by killing you more often than you kill me (of course this is simplifying things as in a real game many different factors are at play and not just aim).

Ideally a deep movement mechanic would scale in the same way so that you could always improve. In theory if you performed it 100% perfectly then you would be untouchable, but it would be so difficult that no one would ever reach that level.

These are the types of mechanics that keep people playing and playing. You can never master them completely and there is always the possibility for you to improve and perform them a bit better than your opponent.

I'm all for trying to make games more accessible to newbies but I think it's a mistake to sacrifice depth for accessibility. You could add a wide cone of fire onto all the weapons so that having good aim wasn't as important, but then you greatly reduce the depth and very quickly players with decent aim would all be performing it 100% perfectly. There would be no room to improve beyond that which would lead to luck (or perhaps another mechanic within the game) deciding the outcome instead of player ability. This would be boring to play and people would quickly move on to something else that was more difficult to master.

internetexplorer · February 2012

<div class='quotetop'>QUOTE (Harimau @ Feb 14 2012, 02:44 AM) <a href="index.php?act=findpost&pid=1902889"><{POST_SNAPBACK}></a></div><div class='quotemain'>This is a good point, but I think this is something that can actually be improved upon, so that it is more obvious and intuitive: Put the number of resource towers and your average income on the scoreboard, and in the game world put a +1 PR above RTs every time you are awarded with a resource and synchronise that with the personal resource counter. This in no way affects the skill required for defending RTs, it just helps perfectly new players understand that they are gaining resources from these towers (and therefore should defend them).</div>

Yeah, more stuff like this would be good. In starcraft, for instance, you judge your income by counting your workers and glancing at how fast your mineral/gas count grows.
In this game, you have less time to focus on the 'speed' of your resource amount, and you can't count your 'workers' at a glance (since the minimap doesn't show power or built/unbuilt state of extractors). I think it would be fine to show this stuff to the player, as long as it can be done in a non-messy way.

Harimau · February 2012

@Wilson: That's fine. My point was the take-home message at the end though:
<div class='quotetop'>QUOTE (Harimau)</div><div class='quotemain'>Now that doesn't mean you shouldn't add a movement system - I think you should - but using balance as the reason (rather than flavour, fun, or a personal sense of progression) is erroneous.</div>
(And this is a thread about balance.)

Wilson · February 2012

<div class='quotetop'>QUOTE (Harimau @ Feb 15 2012, 04:58 PM) <a href="index.php?act=findpost&pid=1903445"><{POST_SNAPBACK}></a></div><div class='quotemain'>Now that doesn't mean you shouldn't add a movement system - I think you should - but using balance as the reason (rather than flavour, fun, or a personal sense of progression) is erroneous.</div>

You can't say it doesn't effect balance just because performing the mechanic perfectly won't make you better than before the mechanic was introduced.

Of course the balance will change when you introduce a new movement mechanic. All of a sudden things that may have been very easy to do before can become a lot more difficult. That can change the balance at different skill levels even when performing the mechanic 100% perfectly (something that ideally is beyond the limits of humans) doesn't make you better than before.

Using the example of aim again, if you had big cones of fire on all the weapons so that aiming accurately wasn't very difficult then it might seem like many of the guns are overpowered. By decreasing the cone of fire you force players to play better and aim more accurately so while the absolute optimal performance hasn't changed, the balance is still affected. The same goes for movement mechanics.

The problem with having simplistic mechanics that are easy to master is they can be very difficult to balance. As many players will be able to perform the task 100% perfectly it then becomes very powerful or they change the numbers and then it's pretty weak even when performed well. The flamethrower is a good example of this. Since it requires very little aiming skill even when you aim perfectly you don't really get rewarded. Yet if they increase the damage too much then because aiming with it is pretty easy it would be overpowered as most players would be able to just spam it in the general direction and deal damage without requiring precise aim.

Harimau · February 2012

From what you said, I can infer that you are concluding that we should balance at all levels of play, not just the top-level.

I was actually indulging this idea of balancing at the top-level by considering balance simplistically and developing my own labels: that "balance" is balance at the top level of play; and (half of) "fun" is balance at all levels of play (the other half is simple enjoyment in the mechanics themselves). That might be where you have misunderstood me.

If it helps, you can consider all mentions of "perfect aim, perfect movement" to be analogies for "maximum power", where the "maximum power" is the "<b>practical</b> maximum" instead of the "<b>theoretical</b> maximum" (i.e. top-level, "pro" level, if you would like to look at it that way). Adding a movement mechanic that increases the practical "maximum power" (player's effectiveness at the top-level) doesn't <b>really</b> change the "balance" (balance at the top-level) because you could achieve the same thing by increasing the parameter; it only changes the level of (half of) "fun" (balance at all levels).

The fact remains that melee is inherently disadvantaged against ranged at any distances except close range - movement mechanics that affect the <b>speed</b> have so little impact on the engagement range, and if you balance at the top-level then in reality these movement mechanics only serve to limit the amount to which "less-skilled" players can mitigate the disadvantage.

Going on a tangent, now. The problem (one problem) with bunnyhopping specifically (at risk of getting this thread locked), is that there is a huge disconnect between pressing the forward button to move forward (intuitive) and rotating the mouse left and right while strafing right and left and jumping to move forward faster. Essentially it is an option. Bunnyhopping is useless to people who are not "good enough" at it, because if they are not "good enough" then just holding forward is the better option: so until you reach the "good enough" level the <b>player's</b> effectiveness doesn't actually scale. So any movement system needs to be universal (per class) and non-optional.

Naturally, you can't find solutions to problems using reductionist concepts, but you can illustrate the problems themselves.

Wilson · February 2012

<div class='quotetop'>QUOTE (Harimau @ Feb 15 2012, 06:19 PM) <a href="index.php?act=findpost&pid=1903468"><{POST_SNAPBACK}></a></div><div class='quotemain'>Adding a movement mechanic that increases the practical "maximum power" (player's effectiveness at the top-level) doesn't <b>really</b> change the "balance" (balance at the top-level) because you could achieve the same thing by increasing the parameter; it only changes the level of (half of) "fun" (balance at all levels).</div>
I disagree. It really all depends on how often the player can use the mechanic to it's maximum. By just changing a parameter you don't effect how often it can be performed well and so it can make things unbalanced. You are again assuming that pros can perform things perfectly all the time, which is untrue and a very simplistic way of looking at it.

Making a certain weapon deal it's maxium damage from a headshot rather than a chest shot can significantly effect the balance at all levels. While pros will obviosuly get headshots much more often, that doesn't mean they always get them in every situation, or that by just changing the value of getting a chest shot you would have the same effect.

I agree with you though about the bunny hopping. I think it probably is difficult for new players to learn and that is one of its problems. I have seen several suggestions for making it easier though which I think would be a good compromise but I don't think the devs are interested.

Harimau · February 2012

Well, okay, but I was only appropriating someone else's simplification (performance vs time), so if there is a problem with the simplification it's not necessarily mine.

The simplification only considers skill, and not choices. It assumes that there is one best way, and measures your performance against that.

Besides, performance doesn't have to be exclusively measured in "how well it can be performed", but can also simultaneously be measured in terms of "how often it can be performed well". Basically it's the average.

And the thing is, you can, in fact, change a parameter and change how easily you can reach maximum performance. If you, for instance, increase the size of the player's head (or at least their hitbox) then you are making it easier to perform headshots.

As a tangent, I think the problem/advantage with headshots in games, though, is that it's an option: You aim at the chest, or you aim at the head. It doesn't actually reward aim (how accurate you are) so much as it awards a good handle on risk / reward (choose a target based on your understanding of your abilities and the situation). Aiming for the head and missing doesn't mean you hit the chest. Compare this to the scoring in an archery tournament - there's only one target, and you are only aiming for that one target, but how close you get to it determines how well you score. So I don't think headshots are a good example to use as they are more a "meta"-skill that is influenced by your aiming skill, rather than a skill unto itself.

slunk_slothrop · February 2012

The answer to this is simple.
Pub players play the game like COD, John-Rambo'ing and not playing as a team.

In NS1 those players played the co_ style games, not the ns_ ones.

Just add the co_ game mode, duh

Balance in NS2

Comments