• Join Us |
  • |
  • Sign in with:

Timescale and v1.08

by Steve Gibson, Aug 03, 1999 5:06pm PDT
Related Topics – Quake 3 Arena, John Carmack

"Not Given" over at the Quake3 messageboard had a bit of correspondance with John Carmack on the timescale issue that people seem to think still exists in v1.08. Well here is the email from the Carmack-meister himself:

>Why does this still work in 1.08? It DOESN'T. It acts all screwed up, but you don't cover more distance or fire more shots. John Carmack




Comments

32 Threads | 32 Comments






  • here is my observation of timescale and what is working and what isn\'t. ]R[Gauss and myself set up a listen server to test everything this evening and the first thing i noticed was that unless i set my timescale to something close to Gauss\' things were awfully lagged and choppy and overall just crap. the closer i set mine to that of the listen server the smoother it became. when i sync\'d my timescale at .6 with the server things were wonderfully smooth and slow :)

    so.... i am only left to assume that if you are adjusting your time scale, the further you deviate from that of the server, the worst the choppiness JohnC describes becomes. I suppose this means someone could set their timescale to 1.05 and gain a slight advantage over anyone using a timescale of 1, suffering minimal choppiness at the huge advatage of shooting and moving faster than his opponent. it is also a little difficult to tell who is doing it if you aren\'t paying close attention.

    bleh.. if anyone has any further info please post it so that carmack can fix the problem before the final release. this is not something that should be a problem in the final. don\'t forget guys... this is just a BETA TEST. try not to go nuts over little things that are being left in for animation testing and other productive things for the id employees. thanks to john for being so responsive as well.








  • I didn\'t notice much difference, and it is jumpy as hell to try it. I think what alot of people are doing to test it is loading up a map with the normal \"map\" command, and then turning on the cheats. Since you are the server, it will work then. To test it by yourself, you need to set up a dedicated server, then connect to it. If you try q3test2 with timescale set to 100, one with a listen server, the other with a dedicated, you tell me how easy it is to stay on the platform on a listen server. I believe the \"clamped\" theory, as I tried timescale 10 and timescale 100000 and couldn\'t tell much difference ohter than the world objects spinning muuuuch faster. g_sync didn\'t seem to make a difference (read the default, 0, was used in my testing).





  • >From our experiments it seems if g_synchronousclients is set to \"1\" it
    >will not work. With it set to \"0\", it will. Thank you for the response
    >
    >Jon S

    I looked into it some more, and there is still a problem, but it is a
    different one... The server clamps the speed, so timescale 4 isn\'t any
    different than timescale 2, but it looks like there is still about a 30%
    slack that can be pushed.

    I don\'t understand yet how this is happeneing, but I will look into it.

    The game play is so messed up with timescale running now that I don\'t think
    it is an effective cheat, but I will still try and fix it.

    John Carmack