Hi, Some of us are still here :)
I have not the time at the moment to fully read the article as it's quite long. From a quick read through I get the impression this is not as simple as just performing GeoIP on a server and instead tries to find closeness via hops/speed?
Isn't this still flawed in that the "user" must still perform these queries on all 36,000 servers?
On a side note: Regarding the "find a server via co-ordinates" system, HLSW uses the lat-long of the player to find servers that are physically closest.
Regards, Andrew On 03/02/2011 07:10, grenville armitage wrote:
Hi, Not sure if anyone's still on this list -- I haven't seen any emails since late 2009. Anyhow. I recently released a paper describing a a possible evolution of server discovery mechanism to vastly speed up the ability of clients to discovery a hundred or so 'closest' game servers. I'm interested in feedback on the practicality of the scheme... and perhaps inspiring someone to try and implement it :) G. Armitage, A. Heyde,"REED: Optimising First Person Shooter Game Server Discovery using Network Coordinates," ACM Transactions on Multimedia Computing, Communications and Applications (TOMCCAP), In press. (accepted January 2011) It might take a year or so before being physically printed in the ACM journal, so I've placed a copy online for now. http://caia.swin.edu.au/pubs/inpress/REED-Armitage-TOMCCAP-preprint-18jan11.pdf cheers, gja _______________________________________________ hlds_apps mailing list hlds_apps@list.valvesoftware.com http://list.valvesoftware.com/mailman/listinfo/hlds_apps
_______________________________________________ hlds_apps mailing list hlds_apps@list.valvesoftware.com http://list.valvesoftware.com/mailman/listinfo/hlds_apps
