This is a problem for me too. My solution is to re-use seqs as often as
possible.
I have this (in my Ubuntu14 VM):
10 * test: 2.598e-05
10 * tx: 2.049e-06
10 * t0: 2.965e-06
But in your benchmark, you need to use `i` so the
I've pushed code to close _all_ unused fds in each new fork, and to end
communication gracefully. I do not understand how the failing fds are known to
**select**.
I think this is a bug in a Nim library. But if no-one can help me figure this
out, I'll try switching from pipes to sockets (maybe
Of course not, there is much allocation work involved. But about 500 ns (a few
thousand CPU cycles) is indeed more as my expectations. I discovered that when
tuning my chess engine -- using one more tiny seq noticeable reduced
performance. Of course seq's are convenient, because of existing
I just saw on the Manning web site that the book has been delayed again until
May this time. It's like we are asymptotically approaching the publication date.
