Try running a Tactical Ops server with TOP3 and check a nice 95% CPU usage for a single server, no matter if there's 4 or 30 players on the server... Compared to TO + TOP3, HL is not that bad, is it ?
nelson
Stefan Huszics wrote:
Sindre wrote:
Numbers lie, the p2 cache (even p3 katmai, not coppermine) is 512kb, but it's sloooooow, so the new cache is extremly, by far, more superior.
It's better yes, but that is not always relevant. You have to also factor in where your bottleneck is. If the problem is that the data simply doesn't fit inside the cashe it makes no relevance how fast your cashe is. What matters then is how fast you can update your cashe from main memory and a P4 has a lot higher "clockcycles-to-get-data-from-main-memory" then a P2.
Of cource, hlds_l being inefficient with cashe is only a theory, but all observations does point at this to at least be a big part of the problem, eg * P4 with just 256k L2 sucking really really bad * Low MHz CPUs with lots of cashe preforming relativly good (P2 512k, K6-2/3 with 1MB L2/L3 etc) * Bartons performing a lot better then 256k & especially 64k L2 AMDs
-- /Stefan
Software never has bugs. It just develops random features. =)
_______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
_______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlds_linux
