User Tag List

Page 2 of 2 FirstFirst 12
Results 11 to 15 of 15
  1. #11
    Whicked Sick Chamberly's Avatar
    Join Date
    Jul 2012
    Location
    Vandora Temple
    Posts
    5,488
    Country:
    Quote Originally Posted by SAM View Post
    We havent had any complaints since....about the server being unplayable.
    Ones that was using VPN to get out of the high end ping are heard to have troubles, they just say and not report...


    http://irc.lc/globalgamers/uscript for uscript discussion.

  2. #12
    Administrator SAM's Avatar
    Join Date
    Jan 2011
    Posts
    8,296
    Country:
    Well. I'm not telepathic. If people don't post. I have no way of knowing whether things are better or worse. Which is why I'm reluctant to change anything. 5 posts saying the same is not the same as the full server load saying nothing. Nothing generally means all is fine.

  3. #13
    Whicked Sick Higor's Avatar
    Join Date
    Apr 2012
    Location
    Full sail ahead!
    Posts
    3,676
    Country:
    Quote Originally Posted by SAM View Post
    Why is tickrate dropping though when in the siege ultimate build it was always pretty steady? What's changed. Has siege become tickrate dependant? Are timers being used less?
    Fragnet server was fully dedicated (CPU speed was 3600 or 3500?) and had 5-10ms less ping due to location.
    Antifun leeches are popular, they add an additional 30 checks per player.
    XC_Engine's relevancy makes builds visible through windows, and variable updates take more CPU than a line check (for non-visible stuff, also, line checks are SSE optimized).
    Also, since the profile system's inception the category info rules aren't fixed but sent from server to client so that's an additional check (of around 250 variables... SHEEEIT)

    0026 reduces the check rate from 10hz to 0.5hz on the category info, and sgBuilding will scale down check count based on server tickrate instead of building %HP.
    Combined with a 0026 SiegeNative (if i ever make it) we would see a tremendous reduction in the CPU usage.
    ------------------------------------- * -------------------------------------

  4. #14
    Administrator SAM's Avatar
    Join Date
    Jan 2011
    Posts
    8,296
    Country:
    Even if it was a fully dedicated CPU. UT is a single threaded processes. Therefore it would use only one core.....so the performance difference should in theory be negligible. Likewise with the ~10 ms ping increase or decrease.

  5. #15
    Whicked Sick Higor's Avatar
    Join Date
    Apr 2012
    Location
    Full sail ahead!
    Posts
    3,676
    Country:
    Well, higher latency means more data in the WinSock buffer due to UT's own bandwidth control
    /smartass off
    ------------------------------------- * -------------------------------------

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Tags for this Thread

Posting Permissions

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