Thursday 5 March 2015

Armored Warfare will have skill matchmaking.

Some news about the upcoming “Armored Warfare” game which is in Alpha stages just came out, two things that caught my eye was the Skill-based MM and that it will be added a protection to protect new players from being sealclubbed, so far things are looking fine, there is more info Here that I advice you to check out.

If you never saw gameplay I actually made a video of it not long ago:


62 comments:

  1. Thats the first thing they will remove or rework to a ladder system after a while...any bets?

    ReplyDelete
    Replies
    1. No bets. These are preliminary notes and they just detail the objectives but I am certain some of them will be dropped based on difficulty vs. development schedule.

      They are copying/pasting WoT MM but also taking into account the compaints about the same. Let's see if they can do what WG has not been able to do so.

      Delete
    2. It's not as easy as it seems ... it needs a large number of players (around 100 000) or a HUGE queue waiting time (around 2-3 minutes).
      The problem will no tangible in a new game (skill level not that different)

      Delete
  2. Awesome. Completely understand why SS left FTR to work for obsidian ;)
    Thanks for the info rita. You're doing a really good job so far.

    ReplyDelete
  3. I agree, skill MM is a bad idea. One thing a game developer can't do is listen to players. I hope WG never introduces skill MM except at tier 1 or 2 to prevent seal clubbing.

    ReplyDelete
    Replies
    1. Ok it's a bad idea. I think it's a great idea.
      I would love to know why you think it's bad though. Care to explain?

      Delete
    2. Yea, if you put all GREEN players together and they play against each other then 1/3rd will soon be RED and another 1/3rd will soon be PURPLE. If you put all RED players together than half will soon be ORANGE and so on and so on. Also, I am green but soon to be blue and recently i am purple, so how do you rank me for skill MM on my first 10K games or my last 1000? Also, why would I want to play against all greens? The games would be BORING since greens are very conservative, they hide, use camo and vision, go hull down and are hard to kill. Matches would be WAY longer and if U put all purples together the games would be a nightmare. You would have to double the 15 minutes. Skill MM sounds like a good idea, but it's stupid.

      Delete
    3. If you put all purples together you would be 5 minutes in matches and you would not have any penetrating shots.

      Delete
    4. can you imagine if you had 10 purple e100s fighting against each other all angling. It might take an hour for them to kill each other!

      LOL

      Delete
    5. It's been repeated so many times, but in short:

      - increased waiting times for almost no profit (landslide wins and losses and loss streaks still happen, only a marginal percent rarer)
      - artificially narrowing playerbases' WR close to 50%
      - questions about which rating should be used - skill MM by WR or WG's rating will still lead to voices that WN8 shows unbalanced teams.

      Delete
    6. Skill MM negates the meaning of victory.
      Skill MM greatly limits the amount of diversity you can have in the game.
      Skill MM increases the importance of RNG massively.
      Skill MM makes the player irrelevant.
      Skill MM is the difference that turns a multiplayer team game into a solo game with intelligent bots.

      If you want to have a diverse and entertaining game, that you can just play a couple times "on the go", and that allows you to make a real difference, you cannot allow skill based matchmaking.

      That said, it's obviously possible to build a game based on skill MM. In WoT, however, it's simply impossible to implement, as it would force the devs to rebalance all of the 300+ vehicles (and possibly the maps as well) in the game to much tighter degree, if that makes sense, and remove a lot of the differences and diversity the game has to offer currently.

      Delete
    7. If i understood the article correctly they are not planning to put a segregation between good and bad player but instead will try to make sur that each team are balanced skill wise. I don't see what's wrong with that ?

      You will not have battle with a 20% to win and everyone will be able to enjoy the battle

      Delete
    8. @Maxime J: That's a common misconception.
      It will lead to longer games for experienced players (say, 20% of playerbase) and shorter battles for noobies (the rest). Because in balanced battles, noobies usually die first, and without any chance to have stronger pros on their team, they will "always" die as a first player of the team. The team might win, but they'll die within first minutes.

      There's nothing as a "battle enjoyable for everyone", btw. WR-wise, length-wise, balance-wise, someone will always be disappointed.

      Delete
    9. Id they do this, I probably wo'nt even try the game...

      Delete
    10. Wargaming actually never added skilled MM due to not having enough players online apparently.

      Delete
    11. @Nya-chan
      The bad player dying first is true no matter how the matchmaking is ... and even if it sucks, that's how you learn to play (at least that's how i learned).

      The fact to begin with a 50/50 chance to win is for me a good thing and will definitely be taken into consideration if i have to chosse betwen WoT and AW

      ok, the enjoyable for everyone might be an utopy in a pvp environment

      Delete
    12. @Maxime If you have 15:14 battle, most likely about 20 noobie guys died before the stuff got serious. Only then the battle got interesting, with the pros and lucky noobies in it.
      If you have 15:3 battle, you have some 15 noobies dying and 3 pros dying before the end. From the noobie's point, he has a bigger chance of surviving during landslides. Of course, only in half of the cases - but still bigger chance.

      Delete
    13. sorry Nya-chan, i do'nt get the 15:14 and 15:3 battle thing ?

      Delete
    14. Whatever the skill MM method they choose (ladder/balanced number of skillful to tomato players), it will make WR of every players dropping/rising close to 50%. I don't want that, please keep the right of better players to pawn. What everyone would want is removal of super troll platoon and better MM (balanced tanks by tier) overall. That would be enough for me.

      That said, I support AW's move to implement skill MM so WG could learn a thing or two from the experiment. :p

      Delete
    15. Mark West, very obviously you have never seriously dealt with skill based MM and never really understood how it works. The basic principle is that your "rank" rises as long as you are better than your opponents, and will level out once they are equally good. All you need is an independent skill indicator. Just consider chess: As a novice, you will start playing against other novice chess players. The better you get (ELO ranking), the better opponents you will get. As you can see in chess, there is a skill hierarchy, and there are grand masters. It doesn't level out where everybody is the same. As long as you are a better player than those around you, your rank will increase - in AW as much as in chess.

      Delete
    16. InfernalesBrot, you are making a lot of absolute claims, but aren't backing them up with any evidence or proof at all. Like Mark West, you do not really seem to be knowing what you are talking about (see my above reply to Mark). I propose that instead of posting emotional propaganda without any base in reality, you starting giving good reasons for your opinion.

      Delete
    17. MaterielDefender, it seems to me that you're the one who doesn't quite understand what we're talking about.

      Do for example some little research, Wikipedia will probably be enough already, and you will find out that ELO rankings for example work well with games that only have small teams; if it's not 1v1 anyways. The bigger the teams get, however, the worse the system works, so for games like WoT and AW as well, it simply is completely inadequate.

      As for the rest, simply logic is all you need to back my claims up.

      Victory becomes meaningless, because if the skill MM works, you will on the long run end up with a win ratio close to 50%.
      Why?
      Because, since you get balanced, you are replacable. You can't make a difference.
      If you were better, you'd have a better opponent.
      If you were worse, you'd have a worse opponent.
      You, as a player, would not matter the slightest.
      What would matter, however, is RNG, because that will always be, unlike player skill, which you propose should be even on both sides, random, and thus: possibly uneven, therefore make an actual difference.
      Since skill does not distinguish the players in a skill MM'd environment, the only thing left to make a difference is the equipment the players use, the vehicles they drive, the ammunition they use, and if there are pay to win elements, the amount of money they are able to spend on the game.
      Hence the "Skill MM greatly limits the amount of diversity", simply because everything has to be balanced tighter.
      However, none of these things can define player skill.
      Unless of course you say that the driver of a Loltraktor is per se a worse player than a Maus driver.

      But that would be stupid, wouldn't it. Well, let's let this be enough about my absolute claims.

      Though, I'll give you that: my 5th claim was purely emotional, but I'm sure you'll understand that one as well, once you comprehend the other four.




      Delete
    18. @Maxime No problem there. The 15:14 is the result of battle - it will happen slightly more often if you have Skill MM. Many players say, that that is what they want - closer results, longer battles, intense until the last second. The thing, though, is that it's only interesting for the more skilled people who survived until the end. For a newbie/pubbie who got killed in the first minute or two it's not interesting as he moves into garage and it only blocks his tank for longer.

      Delete
    19. @Materiel I mostly agree with Infernales and I have to add - even if ELO did work, it would start to overload the players. So far there's WR, WG rating, EFE, WN8 and now you'd add some ELO rating...

      I don't think statistics are bad per se, but there should be two things about them:
      - Don't overload players with them
      - Don't predict future based on them (win chance, oh, win chance)

      Delete
    20. Oh dear. It seems a lot of people have no idea how skill mm works.
      There will be a mix of good and bad players in the teams even with skill mm. There is no such thing as teams of only greens and teams of only purples.
      This means that there is an even chance to win and it depends on the personal contribution of each player. Not the RNG of mm. Less 'gone games' where a team with all good players stomp a team of bad players. Fun for no one and happens all the time.
      And yes if the team is even then RNG makes a bigger difference but with 15% RNG rather than 25% already makes a huge improvement. Not only that but the effect RNG has will be completely overshadowed by each persons contribution. For eg a bad player making a good contribution will make a much bigger difference than it currently does. Just as a good player doing bad will also have a much bigger impact than it usually does. This means that it is way more reliant on the contributions of the players into the team as a whole than RNG. Each player who does better or worse than the stats would suggest in that particular game will have a much bigger impact on the result.
      Also why is it a bad thing that good players will last longer while bad players wont. That's exactly the point. There is now a reason to learn and do well unlike WoT which breeds retards and rewards them for doing nothing.

      Delete
    21. "This means that there is an even chance to win and it depends on the personal contribution of each player."
      - stopped reading there.

      How can you not understand that your personal contribution is completely irrelevant, if your opponent ALWAYS has a player who is statistically EXACTLY as capable of contributing to his team as you are to yours?
      Personal contribution NEVER matters with skill based matchmaking. It may feel like it, but it will NEVER be the case.

      Delete
    22. Stopped reading there...huh...
      Well allow me to explain what I meant. To be honest I assumed with logic and common sense nearly anyone could understand what I meant but alas, apparently not.
      Your reason that there is someone who is your opposite number who can do the same is exactly the reason why the contributions made by everyone matters.
      Because believe it or not each player doesn't perform exactly as their stats say in every game. In fact it is quite rare to be exactly the same. That's why the stats are an average.
      Therefore each persons contribution is how much and how well he/she has done in comparison to his/her opposite number or 'rival' if you like.
      Example:
      Player x in team 1 does better than the stats would suggest while his rival player y from team 2 does his average then the advantage lies with team 1. Then add up all the differences from all the different 'couples' in each team and voila. The team who did overall better than their 'rivals' win.
      That way not only do the contributions of the good players matter but everyone else as well because it is compared to what their opposite number has accomplished in that particular game.

      Delete
    23. I understand why you don't understand me. Allow me to throw some of your logic and "common sense" back at you:

      Player skill varies from their individual statistics with non-skill MM as well! Surprise!
      Oh, wait. Does that mean, that, according to your very own argument, your personal contribution is even MORE important in a game without skill MM?
      Again, surprise: yes it does!

      And if you'd bother and extend your example to the next game, your player x might as well do worse than his stats suggest, while his 'mirror image' does his average. This time the opponent wins.
      Repeat this a couple thousand times, and you will always end up with a 50% win ratio, as in, with the statistical evidence of player x's irrelevance for the outcome of a battle.

      Again, I repeat: Personal contribution NEVER matters with skill based matchmaking. It may feel like it, but it will NEVER be the case.

      You can not outperform your statistics. Because your statistics are bound to your performance.
      And if you actually can, then your skill MM and player rating system isn't working.


      Allow me to emphasise this with a small comparison:

      Imagine a two teams that only consist of Maus tanks fighting each other. Or one Maus per team, one E 100 per team, 1 AMX 50B per team, 1 T92 per team, one Obj-263 per team, ... you get where we're going, right?
      So, would the armour of the individual tanks matter in this game? No, it wouldn't. Because the other team has the exact same amount of armour.
      It's the same with skill MM. If team skills are even, it simply doesn't matter anymore, and you have to take it out of your equation.

      Delete
    24. I may as well add one more thing to the entire skill MM discussion:
      It is waaaay easier to do much worse than your stats suggest (e.g. when you're drunk, or sleepy, in a derpy mood, just don't give a fuck, or whatever..) than it is to do better than your statistical average.

      What this means is that while you can statistically do very little to be a positive asset for your teams (since, again, you get balanced out), you can very easily fuck your teams up.
      Just as easily as in WoT, of course, however, in WoT it simply does not matter, because the game does not expect you to perform to a certain standard. So skill MM would effectively punish you for for once playing "for fun", so to say.

      Delete
    25. And yet even if it were exactly as you say it would still be better than the current system of WoT and non-skilled mm.
      At least there is a 50% chance to win. But on the other hand where 80% of games in WoT the battle is already decided at the beginning just because of the team setup. Not to mention that xvm (yes not official or used by all) is correct 95% of the time of who will win.
      And I dont see how that is better than the skill mm even if it goes exactly as you say.

      Dont misunderstand me i see your point and I can agree with parts of it. However the current skilless mm means that regardless of what i do the match tends to be decided right at the beginning when pressing the battle button. I don't like to put that faith into chance.

      But to each their own. We have made our points.

      Delete
    26. MaterielDefender, you seem to be missing the point. Games are no fun if you're always playing against people with the same level of skill at you and if your winrate is always going to be 50%. fun is in out skilling your opponents.

      Delete
    27. Havoc199, no game is decided before it ends. The matchmaker is working really well most of the time, so every team always has a fighting chance.
      Also, xvm gives you a win CHANCE PREDICTION. A very good one, there have been numerous tests in the past, but that's exactly the point:
      even if you are fighting a game with a predicted win chance of 29%, you are almost definitelly going to win 29 of 100 games of this same kind.
      No battle is pre-decided.
      Because you always have a fighting chance.
      Depending on how good you are, you will take these chances more often.
      Get better, and you will get better results. It all dependson your personal contribution. On the long run even more than in a single battle.

      And that's what would be impossible with skill MM. You can try to take your chances, to improve, but it would have no effect, because the game would simply rate you higher and level out your improvement immediatelly.

      Delete
    28. InfernalesBrot,

      how come you do not understand that your personal effort only is truly relevant against an evenly matched opponent? Only then every small thing you do better than your opponent counts. Otherwise, your personal performance often will not play a role since your team either dominates or is dominated any way. That's how it is with WoT's MM right now: Personal performance only counts in a small area of your team's performance, and that is when win chances of both teams do not differ dramatically. If your team is a bit weaker, a great personal performance would count more, but that doesn't diminish the point that with non skill MM, only a few of all matches offer you the chance to tip the balance with your personal performance. With skill MM however, every match offers that opportunity to every player: One player playing a bit above his usual level may decide. That is a constant motivation to give your best, and will make every mistake count. A team being dominated and still winning because a few of its players play above their usual level or because the other team's players are making massive mistakes is an extremely rare event. That is what you totally fail to understand.

      What you also do not understand is that for the majority of players a match is most fun if it is a neck to neck race; Unlike one of your bold and unfounded claims that is the result of a survey done by Microsoft, the people who also invented TrueSkill, a system to constantly rate players based on their performance in the game and even giving a useable indicator of player performance when little hard performance data for them is available.

      Your argumentation has holes as big as barn doors.

      Delete
    29. "your personal effort only is truly relevant against an evenly matched opponent?"

      For hell's sake, that's the placebo effect I've mentioned an entire day ago.
      Instead of attempting to point out your superior understanding, I suggest you try and extend your horizon beyond one single battle.

      The better you do, the better your next opponent will do.
      The worse you do, the worse your next opponent will do.
      Yes, you can decide one single battle by playing above your standard.
      But you can do that in the World of Tanks as well, NOTHING hinders you from making a good effort, in any game.
      The difference is that in the WoT, if you are able to reliably perform well, you are going to win more. Your contribution matters.
      In AW, with your glorious skill based matchmaking, if you are able to reliably perform well, you are going to... get matched up against an equal opponent, and NOT win more.
      I could let my guinea pigs play AW and they would end up with a proper win ratio close to 50%, because they would get matched up against an equal opponent. Now tell me my piggies' contribution matters.

      Also, you're mistaking skill MM with neck to neck races. Losing a player leads to a huge disadvantage that is going to cause a snowball effect. Even with skill MM, you will generally have way more 15:8 / 9:15 results than 14:15 games. There has even been a large topic on the WoT forums regarding this issue, where a player recorded win chances and battle results of a very large sample of battles. Even the 50%ers generally don't lead to neck to neck races.

      Besides that, it doesn't take much to figure out that neck to neck races are extremely stressful. You don't play games to get stressed.
      If you REALLY need to have this "unfounded claim" backed up, ask Google. There are literally hundreds, if not thousands of books regarding stress.

      I'm aware of the holes in my barn btw, after all perceiving logic can be a difficult task when you're wearing glasses made out of ignorance.


      Again: personal contribution never matters in a game where your skill gets evened out on the enemy team. The difference you make now will be made by your enemy tomorrow. You can't make a difference when your skill gets evened out. If you can, your skill MM is not working. Because that's what it's supposed to do: make sure that both teams have an equal chance, so that nobody can win more than he is supposed to. All of your efforts get evened out. You are not going to make a difference. Stop claiming your placebo is real.

      Well, anyways. We're on the internet, so you'll either just leave without saying anything, or just continue to repeat your misunderstanding. In any case I'm not the one to teach you thought, and this discussion is starting to upset me for no reason, so let's end it here.
      I will in any case.

      Have a ncie day.

      Delete
    30. Hail my stronk typing skillz. With skill MM, you would statistically have to do the same typo now as well.

      Delete
    31. This comment has been removed by the author.

      Delete
    32. InfernalischDummesBrot,

      re balanced matches: Microsoft survey (and hence a lot of computer gamers) says otherwise. And what do you want - your contribution to count in every match, or not? Make up your mind already.

      Your comments are polemics all the way. You did not present any real arguments, just strong personal opinions.

      As far as personal bias go: I'd rather have five really good WoT matches per day where I lose in the end but got thrilled all the way there, than 20 where I either roflstomped or got roflstomped.

      It says all about you that you feel the need to get personally offensive. You are the typical smartass ("Rechthaber") who just refuses to accept any good reasoning and arguments because the real problem is in your personality.

      To back this up with some competence here: I have a university study of computer science, 25 years of professional experience as sw dev and project leader, communication and negotiation trainings and experience, etc. etc.

      I know your kind of personality more than I would like to, and you are getting on my nerves with your arrogance and insultive language.

      Delete
  4. This comment has been removed by the author.

    ReplyDelete
  5. Hello Rita, I have just arrived from FTR blog as advertised by Silent. I hope you will continue his great work and put some leaks&info from all areas that is interested for us - tankers. Personally Im interested in WoT (still ;)), I am waiting for XOne version too and I did register for AW CBT and hoping to get there soon.
    I wish you succesful blog.
    XOXO.

    ReplyDelete
  6. Also FTR survivor.
    If SS trusted u doing fine... I sure can do too :)

    Nice first post i read here
    Hope u mix it up once in a while.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
  8. All the best for the future with taking up the FTR mantle!! Here's to plenty of WOT leaks.

    ReplyDelete
  9. FTR refugee here too :)

    any RSS feed incoming?

    ReplyDelete
    Replies
    1. There's already an RSS feed ( http://ritastatusreport.blogspot.com/feeds/posts/default?alt=rss ) along with an Atom feed ( http://ritastatusreport.blogspot.com/feeds/posts/default )

      Delete
  10. RITA!!!! All the best for the "new FTR" i have faith in you:)

    ReplyDelete
  11. Was just told about this site so HELLO :D
    Its always cool to see people other people covering the news. Myself, i do it on YouTube and do it in live format hence why all my videos are "world of tanks live news" but again blogs can be nice to read from time to time :)

    ReplyDelete
  12. Just dropping by to say hi and salute you for taking on the burden you're lifting from Silentstalker's shoulders. Looks like you're doing a good job so far. Keep up the good work!

    ReplyDelete
  13. @all who complain about skillmm; Please READ this article from obsidian:

    http://aw.my.com/us/news/general/matchmaking-armored-warfare

    It´s not like many represented it in some comments.

    ReplyDelete
    Replies
    1. "both sides will be balanced to have players of roughly equal skill present in each team. If one team has three extremely good players, the other will have the same (within certain margin of course)."

      It's the same system which we are talking about. Yes, it's "every battles are balanced" MM, or "Everyone will get WR close to 50%" MM.

      Which may be good for some people, but not for me at least. However, "not showing the exact rating calculation or even the rating point" is a good step indeed. The rest are similar to WoT, I think.

      Delete
  14. If I had needed more reasons to switch to AW as soon as it will become available, then this would have been it.

    Greetings to WoT, who so stubbornly refuse to implement skill based MM because of their totally retarded idea that "owning others would be satisfactory". That is such a revisionist, bullshit macho blabber. Most people want balanced matches with tight wins and losses. Neither owning nor being owned is much fun.

    I see another nail in the coffin of WoT here.

    ReplyDelete
  15. Now if only my CBT key would arrive!

    P.S. Boa sorte com os trolls do FTR Rita :)

    ReplyDelete
  16. Ok... now that would probably be a minor deal breaker... :/ or maybe a major one... :/
    best regards,
    Hustodemon

    ReplyDelete
  17. its usually a good Idea to actually read first before you assume anything.. typical trolls then dont want to put in the effort of "clicking"

    They do realize that this is a very hot and important topic for many people interested in our game. Therefore, we are pleased to confirm that yes, the teams will not be matched only by their vehicles, but also by player skill. Please note that this does not mean that good players will be segregated from the average ones. That would mean punishing the good players for being good by giving them tougher opponents and we don’t want that. On the other hand, we do understand that when one team consists of top players and the other of new ones, it’s not fun for either side and therefore, both sides will be balanced to have players of roughly equal skill present in each team. If one team has three extremely good players, the other will have the same (within certain margin of course).

    ReplyDelete
  18. Easy fix for it is to have a balanced number of players of different skill on each team. 3 of each skill level per team for example. Would gladly wait in a 2 min queue for such games, than the shite unmatched mm of wot. All oranges vs blues and greens is the norm. Shite for both teams.

    ReplyDelete
    Replies
    1. That sounds like a very bad idea, since at some tiers it can take way longer than two minutes before someone with certain skill level appears.

      Delete
    2. are you guys even comprehending what was said?

      Delete
  19. - Bigworld engine version upgraded to 2.8.1

    so? whats the deal with that?

    ReplyDelete
    Replies
    1. For the common of us? Nothing, maybe some little optimization but that's all

      Delete
  20. I have to be skeptical of all of that. It's nice on paper but implementation is an entirely different matter. The main reason most MM systems fail is because "equal and fair" matches take more time to form as there wouldn't be enough players the system deems acceptable. And people want to play NOW, so they are designed to "widen the net" and take players outside the criteria just to get games going.

    I'll believe all that said about MM when I see it in action and it actually works.

    ReplyDelete
  21. Skill BASED MM is awful, unless you have an enormous player base. Skill BALANCED MM however is superior. Take your same pool of candidates, say 30 based on tier, vehicle type etc, then take the ranking system and sort the #1 on one side, then the #2,3 on the other, then fill the teams accordingly. Far superior, better games, same wait time.

    ReplyDelete