[ENet-discuss] Another Newby Question

Lee Salzman lsalzman1 at cox.net
Thu Aug 23 14:58:49 PDT 2007

Packet loss at one point was envisioned to have a greater role in the 
throttling mechanism, but I found this rather useless in the end, and 
RTTs made a better indicator of network conditions. So, the packet loss 
stuff is currently vestigial, but could feasibly play do more in the 
future if I ever figure out what.

As for scale, because limiting the precision to 100 distinct values 
would be rather silly, and using floating point in what is essentially a 
network stack also would be rather superfluous.


Giuseppe Greco wrote:
> I hope not to bore you... but I have another question.
> What's the idea behind peer->packetLoss and peer->packetLossVariance?
> For instance, why isn't peer->packetLoss a percentage? (the actual scale
> is (65536).
> Tx,
> j3d.

More information about the ENet-discuss mailing list