[Statistics] Insight into what pasta is best

  • Masters

    Also, is it just me or is voting in a map into the voting pool not working for anyone else? Ill vote one in and it will never show up in map battle for selection...

  • @Øwl said in [Statistics] Insight into what pasta is best:

    Are these maps really the most popular/favorite? Or is it because the algorithms used to put maps into the rotation is broken and thats all we have a chance to play?

    If no maps were suggested, the server will randomly select two maps out of the whole map pool and have you vote between them - not too much of a complicated algorithm. You could argue that maybe the last 3 maps or so could be excluded fom the map pool so they can't be randomly selected which i think is a nice idea, but the only other reason that comes to mind is people suggesting those popular maps all over again, which I think is in the nature of the game. I'm not a fan of akroseum but if people suggest it 300 times a day and it is also voted for, there's nothing you can do about that other than suggesting other, better maps yourself.

    Before mapbattles became popular, there has also been a time in which the popular public servers would just play a fixed map rotation and not even enable anyone to influence that. This would guarantee variety and might be another solution to your concerns.

  • Masters

    I could not agree more with Øwl. Is it possible to vote on 4 maps instead of 2? This way, we will have much more diversity in maps played. Also, if I try to suggest a map it doesn’t work unless I’m a master. So any random (normal, non master) player is not allowed to suggest a map I suppose? Maybe another bug?

  • Masters

    @benzomatic I totally agree. If that’s what people keep voting into the map pool its valified. But I’m having the same problem as @Morty , unless i claim auth none of my map votes even make it to the map battle pool, so I’m not sure anyone else’s voted maps are making it either. In addition most players i dont think are even voting maps into the pool, and most agree its a very repetitive map sequence, Hence why its not the players but the algorithm or computer generated map rotation at fault. Did we have this problem in pastaland, demophobia, nooblounge? Not that i remember. If it is what it is, so be it, oh well. Just a minor frustration i found a lot of people shared with me, so i posted it on here to see if anything could be done. As always i appreciate the work you do @Obstriegel , this may only be a problem to players who generally play regularly and at long intervals. Nonetheless it is an issue.

  • I'm not aware of the problem you mentioned, I could always suggest any map just fine without auth, since when has it been like this? And does that happen on riga or on that other server, "classic rugby party" or something?

    The maps might also seem repetitive because a while ago the behaviour of the mapbattle was altered in such a way that only maps fitting in size (depending on the amount of players) were auto-suggested, to avoid situations with 30 people on alloy and 7 on asgard. Afaik this measure was only very limited though to remove the "extremes" on both ends while keeping variety. I don't know anymore how many maps were in which pool or if maps were removed at all, i think obst can tell you more.
    Nevertheless I think any map from any pool can still be manually suggested.

  • Masters

    @benzomatic Just recenlty within the past couple weeks I’ve noticed this problem, figured it was just me though until Morty said something. That makes sense and I like that idea to take into account number of players and all, but sometimes (most of the time), I’d rather play Asgard 4v4 than dust2 for the fourth time that day. Just my preference though.

  • Masters

    @Morty go see a doctor. you are wrong and your brain is malfunctioning.

    @miu when do we eat some pasta?

    @Øwl hehehehe, fuck you!

    First of all I would like to thank you for the questions about randomness because it is something I am working on for a science project!

    Randomness in sauer where we do not roll the virtual dice very often also means we can get many times the same result. Like in a dice game we can be a loser or a winner many times in a row. And our brain always tries to find a pattern in this even if you use perfect dices so there is no pattern.
    Both the pseudo randomness in lua for rigatoni and the pseudo randomness for lets say shotgun shots are kinda good overall.

    Back to sauerbraten. For example on the Trouble in sauertown / Traitor server the sauer mersenne twister is used and while it actually does have a good distribution of results you can only see this when like playing 1000 games for example. Since this is not good I do something different there.
    Imagine a stack of cards, and when you join you pick two empty cards and write your name on them.
    Now the pseudo random number generator is used only to shuffle the cards once.

    For each round a card is removed from the stack and that is the traitor.
    So We have a stack of cards and you can be sure every player will become the traitor twice until the stack is empty and the process is started again.
    In theory it is also possible, that when you have like the last two cards in the stack and a new stack is generated you become the traitor three or four times in a row.

    This works really great I think for this game because everyone wants to become the traitor sooner or later and at the same time have the feeling it is random.
    But it is the traitor game.

    So lets get back to the public server.
    I prefer choice. Someone suggested that we could also just use the old linear map rotation. So no randomness, no voting, all maps are played the same amount of times.
    Well I think it is an achievement that you can choose what to play next.
    And if you think about it... in vanilla sauer it was already the case that you could vote with /map for what you want.
    You just actually have to do it. One fucking command and it seems too much.

    The map voting stuff has been altered in three minor things months ago:

    1. The server does not suggest the same map immediately again
    2. The lists of what is a big and small map have been changed. The code for it is like over a year old already.
    3. A list of questionable maps was added that the server will not suggest itself such as tortuga, see my other thread about it.

    You are free to suggest any ctf map for the mapbattle. It is the same code and works the same as it has for years now.
    If your shit is not accepted you either typed it wrongly or someone else was faster than you and filled the voting pool already.

    The next step for you fat lazy Americans will be the ultimate map picking algorithm. An AI that takes the data up there into account and picks always the map that will keep the player count at a maximum. The ultimate optimization function that takes control away from you. I have a layout for it already. But once we are there actually you as humans are also not needed anymore because we can also calculate the result of any player/map combination. So you can finally stop playing and put that Mc Donald's needle into your arm and inject the mayonnaise.

    I hope we do not go this path and wish you a happy day and good (manual) voting and fragging!

  • @Obstriegel said in [Statistics] Insight into what pasta is best:

    @miu when do we eat some pasta?

    any time :ramen:

    This works really great I think for this game because everyone wants to become the traitor sooner or later and at the same time have the feeling it is random.
    But it is the traitor game.

    that's really sad. I've had like 10+ traitor rounds and NEVER WAS A TRAITOR. I feel abandoned in this mode!

  • Masters

    @Obstriegel I’m YOUR doctor, don’t forget obst, an apple a day.

Log in to reply