On Fri, 9 Dec 2016 03:04:37 +0100 "Enrico Weigelt, metux IT consult" <enrico.weig...@gr13.net> wrote: > I'm just having a look at ServerPlayer::csKill(). IIRC it sends out > CS'es for all the individual changes caused by killing a player, eg. > removing it's settlements, diplomatic relationships, etc. > > Do we really need that traffic at all ? Why not just sending a kill > player notification the clients and let them do all the cleanup on > their own ?
Why pick on csKill? All of the cs*() routines work like that. FreeCol is very much a client-server application --- the client sends requests, and the server tells it what happened (as minimally as reasonably possible). I do not see the point in doing kill processing in the client, because the server has to do it anyway, so you are just duplicating effort. Worse still, back when I joined FreeCol (around 0.7.3?) there *was* more client side processing, and we had a continual run of client-server desynchronization bugs, where the client did something slightly different from the server. Having fixed a lot of those I have no wish to return to that mode. It is a good thing for the FreeCol client to be dumb. Every time I have run any performance profiling, the limiting routines are in the graphics/map display. That is where the client needs to be putting its effort. Cheers, Mike Pope
pgpXjfXj_bLCY.pgp
Description: OpenPGP digital signature
------------------------------------------------------------------------------ Developer Access Program for Intel Xeon Phi Processors Access to Intel Xeon Phi processor-based developer platforms. With one year of Intel Parallel Studio XE. Training and support from Colfax. Order your platform today.http://sdm.link/xeonphi
_______________________________________________ Freecol-developers mailing list Freecol-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freecol-developers
