Hello, good job, and something about balance.

IllumiIllumi Join Date: 2013-03-07 Member: 183776Members
Hello UWE and NS2 community. I've seen so much complaining on these boards that I feel a strong desire to throw in my 2 cents (it's such a shame that these boards have so much negativity, but that's the internet for you: anonymity + audience = **ckwad). Natural Selection is awesome. NS1 was great, it stands out in my memories as the best HL mod. However this game is even more impressive. You (the developers) have managed to retain the excellent core gameplay of NS1, at the same time grown a larger community of players, and modernized the engine/graphics (seriously this game is beautiful). You've done it as an underdog in the game development world. I have deep respect for the risks some of you took for this project, and from my perspective it is a huge success. Please continue to work hard to improve the game, but also enjoy your well earned $ :)

I want to look closer at the balance numbers for this game. Recently it was announced that the winrate for 240 was ~50% for both teams. This looks good at a glance, but I believe that this statistic does not tell the full story. I know that ns2stats is a terrible source, for well-documented reasons (mods, sample size, competitive vs. pub, limited server representation, &c). However I noticed some trends on ns2stats that seem to confirm a gut feeling I've had about the game since 240. Let me explain. A huge portion of games last from 10 to 20 minutes. Marines are winning a majority of the games in that category. If a game is shorter than 10 minutes, it was probably a rush, or maybe an all-in strategy like shade hive. Marines are winning a huge portion of these >10 minute "rush" or "all in" matches. The alien winrate seems to creep up as the match length goes on. I don't know how to fix it, but I want to see a consistent winrate for both sides regardless of match length.

One more thing: as hard as it would be to balance, I would absolutely LOVE to see more upgrade paths for both teams. 3 upgrade abilities per hive? More marine weapons? It makes the game more fun when there are more paths to victory. Again, I know it's hard to balance such things.

Anyways, keep up the good work, and thanks for making NS2! I'm anxious to see your next project.

Comments

  • Mattk50Mattk50 Join Date: 2013-02-04 Member: 182824Members, Reinforced - Shadow
    edited March 2013
    *snip* If you have nothing nice to say, say nothing at all. - Angelusz
  • IllumiIllumi Join Date: 2013-03-07 Member: 183776Members
    Hi! I wasn't being negative, I was offering my opinion on the game balance. In fact, my post is full of positive statements. Maybe you see any talk about balance as negative? If so that is a misconception on your part. If we're going to quibble about semantics, lets go to town and get a real working definition:

    neg·a·tive (ng-tv)
    adj.
    1.
    a. Expressing, containing, or consisting of a negation, refusal, or denial: gave a negative answer to our request.
    b. Indicating opposition or resistance: a negative reaction to the new advertising campaign.
    2. Lacking positive or constructive features, especially:
    a. Unpleasant; disagreeable: had a negative experience on his first job.
    b. Gloomy; pessimistic: a negative outlook.
    c. Unfavorable or detrimental: a negative review; a negative effect on the child's development.
    d. Hostile or disparaging; malicious: ran a negative campaign against her opponent.

    Tell me again how I was being negative? I certainly didn't intend to be!

    Anyways, what do you think about my "gut feeling" on the game balance? It's not helpful at all to simply call me names and insult my intelligence, you should at least respond to my points!
  • bERt0rbERt0r Join Date: 2005-03-23 Member: 46181Members
    edited March 2013
    I can try to explain this trend: If games end <10 minutes, it was most likely a concede marine victory. If the aliens decide to fight it out, the longer the game goes the more likely they win. This is because taking out hives is terribly hard for the marines. If too few marines go in there, the wavespawn will kill the marines and the attack fails. If they focus down the eggs first, it takes them a lot of time that allows the aliens to assemble a defense or a counterattack. If too many marines go in there, the aliens can pownoderush the mainbase. To do a successful assault on a hive, the marines need to defend their bases and clear all the aliens and structures in the hive room. Just rushing and shooting down the hive does not help much (if it's not the last one), 40 res is not that much of a setback and the aliens keep their upgrades.

    I think we have a design dilemma here: The marines are supposed to have superiour firepower and act on a shock and awe strategy. The aliens are supposed to do guerilla warfare. The marines try to force a decisive battle with the aliens while the aliens try to harrass and to pinpoint attacks all over the marine territory.

    On the other hand you need marine ninjas/rambos harassing rts or killing upgrades. While these players can do serious damage, the marines are not really effective in that role. A harrassment unit for marines would maybe be a jetpack and flamer marine, sadly jp+flamer is damn expensive for a suicide mission like that.
    As mentioned before, losing 2nd hive does not hurt the aliens as much as the marines losing 2nd cc therefore forcing this decisive battle is next to impossible. The aliens decide where to engage.
  • CrazyEddieCrazyEddie Join Date: 2013-01-08 Member: 178196Members
    That wasn't negative, that was constructive.

    I too would love to see more viable paths to victory. The richer UWE can make the strategic options, the better the game will be. And we all want the game to be better.
  • NeokenNeoken Bruges, Belgium Join Date: 2004-03-20 Member: 27447Members, NS2 Playtester, Reinforced - Shadow, WC 2013 - Silver, Subnautica Playtester
    edited March 2013
    @Illumi

    Let's assume that games which end under the 10 min mark did not have well balanced teams to start with. A stacked alien team could end it in the first minute with a skulk rush, but more often they'll just box the marines in until they get some higher lifeforms/upgrades to finish the job. You don't often see skulks ending a marine turtle. A stacked marine team however doesn't have to wait for jetpacks, arcs or exo's. They can just set up a pg outside the hive (not even necessary if it's a close spawn), rush in and shoot all the eggs. They don't really need any upgrades to do this.

    Basically, I think a strong marine side can finish off a weaker alien side faster than a strong alien side can finish off a weaker marine side, which could explain the stats you're referring to.

  • xen32xen32 Join Date: 2012-10-18 Member: 162676Members, Reinforced - Supporter
    Speaking of tech paths, I'd love if it was not possible to fully tech up in any match. Like, 5 hive types for aliens. Even if you are owning marines so much you got 4 hive, you still won't get everything. And then it will be decision to make, not only what hive goes first, but what haves are we actually going to make.
    Lots of abilities should be added for this. So this will never happed. But it'd be awesome.
  • SavantSavant Join Date: 2002-11-30 Member: 10289Members, NS1 Playtester, Contributor
    Illumi wrote: »
    I want to look closer at the balance numbers for this game. Recently it was announced that the winrate for 240 was ~50% for both teams. This looks good at a glance, but I believe that this statistic does not tell the full story. I know that ns2stats is a terrible source..
    That stat didn't come from NS2stats, it came from UWE's internal statistics. The lead developer himself posted the stats here.

    So it's not 'good at a glance', it was taken from 25,000 games from build 240. That's about as good as you can get.

  • TinkerTinker Join Date: 2003-03-11 Member: 14395Members
    The thing is, even in natural selection there was a bell curve for tech. Part of the asymmetry of the game is that at different times the sides are different strengths. I remember distinctly that feeling of "5 minutes? Aliens are about to get SERIOUS!" I have no numbers as to how that bell curve is shaped or even how it SHOULD be shaped but I do disagree that for any given match length the win rate should be 50%. Like classic NS, it should start with kind of a marine advantage (Very early skulks are kinda easy), shift to aliens when they get their big tech bump after 2 hives, then back towards marines as they round out weapon and armor upgrades.

    I think Ideally 7-12 minute games should be marine favored (they have controlled the infection), 12-18 should favor aliens as the infestation surges and anything after as close to 50/50 as possible. These numbers are all arbitrary to illustrate my point.
  • WhiteWeaselWhiteWeasel Join Date: 2012-11-25 Member: 173197Members
    edited March 2013
    Mattk50 wrote: »
    *snip* If you have nothing nice to say, say nothing at all. - Angelusz
    He knows now from your example of it.
  • tarquinbbtarquinbb Join Date: 2012-11-03 Member: 166314Members
    edited March 2013
    Neoken wrote: »
    @Illumi

    Let's assume that games which end under the 10 min mark did not have well balanced teams to start with. A stacked alien team could end it in the first minute with a skulk rush, but more often they'll just box the marines in until they get some higher lifeforms/upgrades to finish the job. You don't often see skulks ending a marine turtle. A stacked marine team however doesn't have to wait for jetpacks, arcs or exo's. They can just set up a pg outside the hive (not even necessary if it's a close spawn), rush in and shoot all the eggs. They don't really need any upgrades to do this.

    Basically, I think a strong marine side can finish off a weaker alien side faster than a strong alien side can finish off a weaker marine side, which could explain the stats you're referring to.

    exactly this... fine explanation for marines winning in 10-15 and aliens winning later.

    otherwise, if 'most games' last 10-20 mins in the marine favoured bracket, why the hell is it still 50% win rate? :P

    personally i wish the snowball effect wasn't so towering in this game. it makes it all a little redundant to play when you know that your team is going to lose and if you fight you're gonna do a lot of dying for naught.
  • IllumiIllumi Join Date: 2013-03-07 Member: 183776Members
    Thanks for the replies. Neoken, I never thought about it that way but I see what you mean. That would explain the larger number of marine early wins. Savant, I didn't mean to say that the official 50% number was from ns2stats. I meant that "50%" doesn't tell the whole story. I looked at ns2stats and noticed a trend: marines were winning a lot more of the short-duration games. I now see that this isn't necessarily unbalanced. Tinker made the point that the winrate can swing back and forth over time as the teams tech up at different times.
  • Mattk50Mattk50 Join Date: 2013-02-04 Member: 182824Members, Reinforced - Shadow
    edited March 2013
    Illumi wrote: »
    Hi! I wasn't being negative, I was offering my opinion on the game balance. In fact, my post is full of positive statements. Maybe you see any talk about balance as negative? If so that is a misconception on your part. If we're going to quibble about semantics, lets go to town and get a real working definition:

    neg·a·tive (ng-tv)
    adj.
    1.
    a. Expressing, containing, or consisting of a negation, refusal, or denial: gave a negative answer to our request.
    b. Indicating opposition or resistance: a negative reaction to the new advertising campaign.
    2. Lacking positive or constructive features, especially:
    a. Unpleasant; disagreeable: had a negative experience on his first job.
    b. Gloomy; pessimistic: a negative outlook.
    c. Unfavorable or detrimental: a negative review; a negative effect on the child's development.
    d. Hostile or disparaging; malicious: ran a negative campaign against her opponent.

    Tell me again how I was being negative? I certainly didn't intend to be!

    Anyways, what do you think about my "gut feeling" on the game balance? It's not helpful at all to simply call me names and insult my intelligence, you should at least respond to my points!

    You were negative about the current game balance, while complaining about negativity. That is, by definition, hypocritical. Im not saying that your post wasnt constructive as well, but it was, unquestionably, hypocritical, you just shouldnt complain about negativity and call people co ckwads without expecting some sort of criticism of your posting skills.

    I refuse to be censored because of a failure to understand this simple fact, containing important advice for better future posting. Maybe his post complaining about other people's so called negativity should be nuked, if this is somehow now something posts get deleted for. Please, if this is now a forum where im expected to shout di ckwad from the rooftops at everyone i disagree with, inform me.
  • _jay_jay Join Date: 2012-11-04 Member: 166951Members
    Illumi wrote: »
    Thanks for the replies. Neoken, I never thought about it that way but I see what you mean. That would explain the larger number of marine early wins. Savant, I didn't mean to say that the official 50% number was from ns2stats. I meant that "50%" doesn't tell the whole story. I looked at ns2stats and noticed a trend: marines were winning a lot more of the short-duration games. I now see that this isn't necessarily unbalanced. Tinker made the point that the winrate can swing back and forth over time as the teams tech up at different times.

    I posted something similar yesterday. Might be worth for you to check out.
  • IllumiIllumi Join Date: 2013-03-07 Member: 183776Members
    _jay, thanks for the link - that is exactly what I was talking about. 239 was skewed for the aliens, but it was a relatively consistent skew. Obviously we want the imbalance to go away, which 240 achieved. However the winrates for 240 are not consistent; they change as the match length changes. Also, I didn't consider that patch 241 may have completely changed this, with the skulk fix.

    I don't want to argue about it with you Mattk50. Let's just agree to disagree about what being "negative about the current game balance" entails. I shouldn't have antagonized you in my first response, sorry. My comment about internet **ckwads was a reference to a penny-arcade comic (beware foul language):
    http://www.penny-arcade.com/comic/2004/03/19
    It was an off-the-cuff remark meant to disparage the negative attitudes on these boards. Being negative about the negativity is hypocritical from your pov, I get it. I disagree (is it intolerant to not tolerate intolerance?), but I understand why you think so.
  • ZekZek Join Date: 2002-11-10 Member: 7962Members, NS1 Playtester, Constellation, Reinforced - Shadow
    If aliens win more as game length goes on, the reason is because the game is often already won by the 10 minute mark, but it takes aliens another 10+ minutes to seal the deal. Marines who are winning decisively on the other hand are more likely to be able to push in and finish it before the game gets that long.
  • ChrisAUSChrisAUS Join Date: 2012-11-17 Member: 172108Members
    Illumi wrote: »
    A huge portion of games last from 10 to 20 minutes. Marines are winning a majority of the games in that category. If a game is shorter than 10 minutes, it was probably a rush, or maybe an all-in strategy like shade hive. Marines are winning a huge portion of these >10 minute "rush" or "all in" matches.

    Alot of this has to do with how your early game works out. I have no stats or graphs to back me up, but basically if you are winning handily in the early game, you go into the mid game with a large advantage.

    Also you will find this happens in basically any game that isn't symmetrical. Different races or classes have builds or strategy that enable strong timing windows.
    eg: When marines get shotguns, when aliens get lerks or fades and when marines get JP.
  • SavantSavant Join Date: 2002-11-30 Member: 10289Members, NS1 Playtester, Contributor
    Illumi wrote: »
    That would explain the larger number of marine early wins. Savant, I didn't mean to say that the official 50% number was from ns2stats. I meant that "50%" doesn't tell the whole story.
    Oh I never meant to suggest that the number of wins for each side was in any way indicative that every possible facet of the game was balanced. The 50% number simply meant that when a person joined a game they were as likely to come out with a win as they were a loss - in the big scheme of things. That doesn't mean the developers' job is done and they can go home now. :)

    Another reason why longer games tend to favor aliens is since the alien lifeform economy is NOT tied to anything the commander does. If you are an alien you can evolve to ANY lifeform if you have the resources to afford it. For the marines they need to research the technology first, and much of that technology is also tied to per-requisites. (Like the EXO needs the prototype lab, which needs two command chairs and an advanced armory before it can be built.) So the marines are much more tied to tech and upgrades, and if they do NOT get that tech then they can NOT buy those better units. Marines generally lose by default since the Onos is not tied to anything. Sooner or later an alien player will have his 75 res to buy an Onos. No research needed. Aliens can be stuck with one have and no upgrade chambers and they can still buy the most powerful unit in the game. This is a key reason why marines are facing a race against the clock. Once the Onos train arrives, things tend to go bad for marines very fast.

    Alternately, marines can be more successful in the early game if they can prevent the aliens from disrupting their economy. What usually kills the marine (early) game are lost extractors. If you lose the first three extractors you drop, the game is pretty much over. Nothing short of rushing the alien hive and egg-locking them will get you out of that hole.

    So I wasn't trying to suggest that just because the win/loss ratio was 50/50 that the game was 'perfect'. All I was saying is that over a great many games the chance a person would win or lose was pretty even in 240.
Sign In or Register to comment.