Page 1 of 2 12 LastLast
Results 1 to 20 of 31

Thread: TF2 New Steam Server Scoring

Hybrid View

  1. #1
    Badministrator grom's Avatar
    Join Date
    Nov 2008
    Posts
    4,256
    Cookies Given
    2,297
    Cookies Received
    4,806

    TF2 New Steam Server Scoring

    From the TF2 Blog:
    One of the things we've been thinking about for a while now is how to improve the player experience around finding a server to play on. It's a tricky problem because our master servers need to ask a game server for its details, and that server can lie to us if it wants to. We decided we needed to find a way of scoring servers, with a goal of finding and delisting ones we considered "bad". The scoring system had to penalize lying without penalizing custom game rules, because some players like custom game rules. Best case, the system needed to work entirely from data that didn't come from the servers themselves, so they couldn't lie to us in any way to affect it.

    After kicking around some proposals, we came up with a simple system built around the theory that player time on a server is a useful metric for how happy the player is with that server. It's game rules agnostic, and we can measure it on our steam backend entirely from steam client data, so servers can't interfere with it. We already had this data for all the TF2 servers in the world, allowing us to try several different scoring formulas out before settling on this simple one that successfully identified good & bad servers:

    * New servers start with a score of 0 points
    * Each time a player connects to a server, it loses 15 points
    * For each minute the player stays on the server, it earns 1 point (up to a max of 45 points per player)

    In short, servers that have lots of players joining & leaving rapidly will score badly. Servers that consistently have players join and stay on for long periods of time will score well.

    Here's a graph showing server scores for all the TF2 servers in the world as of last week. For the purpose of scoring, we ignored all password-protected servers, and all servers that had fewer than 200 connections a day. The blue line in the graph represents the scores for all the TF2 servers. The red line is the matching player connection count for each server.



    As you can see, the bulk of servers in the world are doing a pretty good job of providing an experience that's expected by the people joining them. More importantly, it's really easy to see what servers are bad. Overlaying the number of players connecting to the servers illustrates how nasty an effect these bad servers are having on players. The very worst servers attract a large number of connections, mostly because they're lying in ways that make them look like a very attractive server at all times.

    Our first step in improving this part of the player experience has been to delist all the really bad servers. The master server will simply stop giving these to you when you fire up the serverbrowser. After that, we're going to keep improving our ability to measure this kind of problem.
    ---
    (aka JennyCide.ps)
    putting the Rage in Average
    gromblog (inactive at present)
    Twitter

  2. #2
    Badministrator grom's Avatar
    Join Date
    Nov 2008
    Posts
    4,256
    Cookies Given
    2,297
    Cookies Received
    4,806
    Or to put it simply, try not to join and leave the server instantly, leave it 15mins if you can - that should be enough time to message others to come and join you or to hop onto ventrilo and see if anyone is up for a game (It's remarkable how often this works, just pop into Vent and ask if anyone's bored and up for a quick game and the server soon fills up again).
    ---
    (aka JennyCide.ps)
    putting the Rage in Average
    gromblog (inactive at present)
    Twitter

  3. #3
    Playstuff Member .ps Rev's Avatar
    Join Date
    Dec 2008
    Posts
    963
    Cookies Given
    125
    Cookies Received
    178
    Is there a way of seeing a partcular servers score yet? Also does this mean that for those times that many of us play for hours on end we should d/c and re-join roughly every 45 mins to max score? or is it each player can add only so much to the score? as in +30 per day? ect.
    Burning Ice and Biting Flame!

  4. #4
    ლ(ಠ益ಠლ) Logan's Avatar
    Join Date
    Nov 2008
    Posts
    6,351
    Cookies Given
    4,727
    Cookies Received
    3,341
    Blog Entries
    15
    interesting.

    so 32 slot servers score more highly in all likelihood? odd considering they aren't officially supported.

  5. #5
    Administrator
    L4D2 Group Co-ordinator
    Hawk's Avatar
    Join Date
    Nov 2008
    Posts
    6,724
    Cookies Given
    1,982
    Cookies Received
    3,443
    Blog Entries
    2
    I'd be quite upset to see my server marked as a "bad" one just because people don't stay on them very long. People easily come and go on servers with a low amount of players.
    Last.Played


  6. #6
    Playstuff Member .ps Rev's Avatar
    Join Date
    Dec 2008
    Posts
    963
    Cookies Given
    125
    Cookies Received
    178
    Well I personally find 24 the best server size, good number of players without turning into a complete grind spamfest so I would hope others will too and that will make our server score well, as I am sure it will.

    We have a large enough base of regs that we should even out any join leavers I would hope.

    Do we have any otehr info about what constitutes a join? what if someone clicks connect but then doesn't want to download the map? or if it is someone who has been banned but tries to join?(you know who I am talking about) its says they have joined the server but in reality they havn't actually?
    Burning Ice and Biting Flame!

  7. #7
    Badministrator grom's Avatar
    Join Date
    Nov 2008
    Posts
    4,256
    Cookies Given
    2,297
    Cookies Received
    4,806
    The point doesn't really seem to be to actually rank servers, just get rid of the worst offenders from the list - servers that appear on the list will still be sorted by ping, name etc so it shouldn't affect servers like ours.

    What I don't understand is how long it takes to get a server listed and what factors would cause a server's stats to get reset - I assume a new rotation would not cause issues but changing a name, an ip or a config?

    I 'think' a join constitutes getting to the 'sending client info' stage so if someone decides to quit because they don't want to download a map then they won't have joined at that point.

    Again, it does seem to be for the purpose of delisting bad servers, not necessarily ranking good servers in any order but we shall see...
    ---
    (aka JennyCide.ps)
    putting the Rage in Average
    gromblog (inactive at present)
    Twitter

  8. #8
    Playstuff Member .ps Rev's Avatar
    Join Date
    Dec 2008
    Posts
    963
    Cookies Given
    125
    Cookies Received
    178
    I would have assumed there would be some sort of ranking, even if it is in-house at valve and not released except maybe monthly or less frequently, else why have such a specific point system? but then again this is just me assuming and hypothesising, I don't have any real ground to base my thoughts on
    Burning Ice and Biting Flame!

  9. #9
    Windows Only MacNetron's Avatar
    Join Date
    Dec 2008
    Posts
    6,539
    Cookies Given
    4,444
    Cookies Received
    4,017
    Blog Entries
    2
    It does give 'kicking for playstuff member' some new dimensions
    Kicking a person who just connected or in spectate (usual policy) makes the server rating bad, but kicking a long time connected non-regular is bad for the good name of the server.

    But, already stated by Rev, I think the friendliness of the server combined with the big number of regulars means we don't have to worry...

    But team stacking 12 regs. vs 12 nonregs and then steamrolling them might give loads of rage quits within 15 minutes

  10. #10
    Badministrator grom's Avatar
    Join Date
    Nov 2008
    Posts
    4,256
    Cookies Given
    2,297
    Cookies Received
    4,806
    Quote Originally Posted by MacNetron View Post
    It does give 'kicking for playstuff member' some new dimensions
    I actually saw another forum today that has what sounds like a decent system - there are two reserved spaces on the server for ppl who have donated a couple of dollars. If the server is full, they can still take a reserved space without anyone being kicked and when the next player leaves, the reserved space is released back into the pool - the only catch is that you can then have 26 players on a server advertised as 24 but that doesn't seem to matter. Seems like a good, friendly way to do it to me.
    ---
    (aka JennyCide.ps)
    putting the Rage in Average
    gromblog (inactive at present)
    Twitter

  11. #11
    Now even more spooky!
    Spam Prince Vegeta
    awesome-o's Avatar
    Join Date
    Nov 2008
    Posts
    4,117
    Cookies Given
    2,501
    Cookies Received
    1,711
    woops server crash, minus 300 DKP

  12. #12
    (1)Spudd has joined the game. Spudd's Avatar
    Join Date
    Dec 2008
    Posts
    1,936
    Cookies Given
    2,342
    Cookies Received
    626
    [quote"Tf2 blog"]In short, servers that have lots of players joining & leaving rapidly will score badly.[/quote]

    will the reoccuring (1)Spudd will effect it?

  13. #13
    Administrator
    L4D2 Group Co-ordinator
    Hawk's Avatar
    Join Date
    Nov 2008
    Posts
    6,724
    Cookies Given
    1,982
    Cookies Received
    3,443
    Blog Entries
    2
    Quote Originally Posted by Spud View Post
    will the reoccuring (1)Spudd will effect it?
    Damn it Spudd, one day you will doom us all!
    Last.Played


  14. #14
    Has a beard now
    Join Date
    Nov 2008
    Posts
    2,159
    Cookies Given
    230
    Cookies Received
    732
    Blog Entries
    5
    Does it take into account bans/kicking?

  15. #15
    (1)Spudd has joined the game. Spudd's Avatar
    Join Date
    Dec 2008
    Posts
    1,936
    Cookies Given
    2,342
    Cookies Received
    626
    haha Hawk, i do try

  16. #16
    Administrator jim's Avatar
    Join Date
    Nov 2008
    Posts
    2,664
    Cookies Given
    337
    Cookies Received
    2,712
    I set up my own server for my own little experiment with tf2 players, had the max players set to 32 but a 24 visible limit so if I joined it would be 25/24, if one person left it would be 24/24 and people still couldn't join until there was a slot free by another person leaving. The only thing is that the teamplay server is bought and hosted by someone else, meaning you would have to pay more for the extra slots to simply hide from the public. You COULD set the visiblemaxplayers to 22 and have 2 hidden spots but with no regs you're going to be limited to 22 players.

  17. #17
    Has a beard now
    Join Date
    Nov 2008
    Posts
    2,159
    Cookies Given
    230
    Cookies Received
    732
    Blog Entries
    5
    I don't like the idea of reserved slots if I'm honest. We only have to kick when the server is full - which is mainly when lots of regs are on anyway. So kicking a few people won't really make that much difference considering the server is already full, and regs are the people waiting to get in.

  18. #18
    Administrator jim's Avatar
    Join Date
    Nov 2008
    Posts
    2,664
    Cookies Given
    337
    Cookies Received
    2,712
    Quote Originally Posted by Jamesy View Post
    I don't like the idea of reserved slots if I'm honest. We only have to kick when the server is full - which is mainly when lots of regs are on anyway. So kicking a few people won't really make that much difference considering the server is already full, and regs are the people waiting to get in.
    that idea isn't reserved slots though?

  19. #19
    Has a beard now
    Join Date
    Nov 2008
    Posts
    2,159
    Cookies Given
    230
    Cookies Received
    732
    Blog Entries
    5
    No, it's my reasoning against them.

  20. #20
    ლ(ಠ益ಠლ) Logan's Avatar
    Join Date
    Nov 2008
    Posts
    6,351
    Cookies Given
    4,727
    Cookies Received
    3,341
    Blog Entries
    15
    i like the idea of a 26 slot with reserveds. i was going to suggest we try that out on the map lab, which i was also going to suggest making instant respawn and HLstats - to get traffic for custom maps?

Page 1 of 2 12 LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •