On Fri, Feb 24, 2012 at 11:00 AM, Kuba <[email protected]> wrote:

> Dear Ian,
>
> Indeed, adding a pause of ~350ms after the context termination makes the
> loop stable, however, it is too long to call it "low latency" :)
>
> Is there any way to fix it without the delay? Or perhaps for my simple
> scheme of communication I should consider using windows named pipes?
>
> Kuba
>

The issue is really with the kernel so it's a bit tricky. Are you doing the
bind() in a fast loop then, is it possible you could connect() in the loop,
and bind() on the other side?

Ian
_______________________________________________
zeromq-dev mailing list
[email protected]
http://lists.zeromq.org/mailman/listinfo/zeromq-dev

Reply via email to