> Your example is wrong. Server tries to run game tick as many times as
> time_to_previous_tick / TICK_LENGTH. So if it gets behind at 1 tick
> (which is bad) it'll try to compensate it.
In that case, it would still be bad (as you say), and the reduced FPS
will show it. The measured FPS would still have some relevance in seeing
a performance problem.
> What I'm interested as a player is how many network updates can the
> server reliably transmit when server is running game/mod with 32 or 64
> players on it. Frame count per second is totally irrelevant because
> ticks use most CPU not frames (frames during which there were no ticks
> executed).
What I am saying is that is at least sometimes relevant, because it is
as close as we can come to seeing the realized tickrate -- the number
that we'd prefer to see.
> BTW I hate email quoting style used here.
Everyone has an opinion on that one.
-John
_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please
visit:
http://list.valvesoftware.com/mailman/listinfo/hlds_linux
