[ENet-discuss] ENet 2.0 most wanted features?

Lee Salzman lsalzman at gmail.com
Tue Apr 30 08:28:40 PDT 2013


int simulatepacketloss(ENetHost *host, ENetEvent *event)
{
     return rand() / (double)RAND_MAX > 0.42 ? 1 : 0;
}

host->intercept = simulatepacketloss;

On 04/30/2013 06:21 PM, Nuno Silva wrote:
> That may be true at some point, however in game development you don't want to hack other people's devices just to simulate lag and packet loss. Valve for instance already implements their own packet loss and lag simulation on their Source engine, so it might not be a bad idea to give this some thought. Otherwise, for the people who would be more interested in having that sort of support than in hacking their OS like me, if Lee decides not to implement it some tips on how we could do it ourselves would be useful for sure!
>
>
> On Tue, Apr 30, 2013 at 4:16 PM, Shaun Reich <sreich at kde.org <mailto:sreich at kde.org>> wrote:
>
>     packet loss simulation and latency are all something your OS (if it
>     isn't completely shitty), should be able to do. in the case of windows
>     you need to go through a lot of extra work to do it. in the case of
>     linux it's 2 short commands. there's no point in duplicating and
>     wasting effort on that, which is the job of the OS itself...
>
>     as for ENET_PACKET_FLAG_SEMI_RELIABLE, i'm wondering in which cases
>     one would actually want to use that?
>
>     NAT hole punching is definitely not something not-so-difficult. And
>     there are other libs out there that are meant for it that can be
>     integrated into your network handling.
>
>     "As a follow up I meant that it would allow us to simulate packet
>     drops and packet sending speed. Also allowing to simulate a % for
>     packet drop and packet sending through some parameter to the Host
>     would be just as good if it's too complicated!"
>
>     again, that's your operating system's job, not a job for a networking library.
>
>
>
>     --
>     Shaun Reich,
>     KDE Software Developer (kde.org <http://kde.org>)
>     _______________________________________________
>     ENet-discuss mailing list
>     ENet-discuss at cubik.org <mailto:ENet-discuss at cubik.org>
>     http://lists.cubik.org/mailman/listinfo/enet-discuss
>
>
>
>
> _______________________________________________
> ENet-discuss mailing list
> ENet-discuss at cubik.org
> http://lists.cubik.org/mailman/listinfo/enet-discuss



More information about the ENet-discuss mailing list