> I agree, but you should admit that OpenBSD is clearly a looser in > regard to "pure performances" (e.g. I/O, compression, encryption, > etc.) > > > Nick. > > Bye, > Ektor >
They should have also ran tests on multiple hardware, single core and 32bit. 32 bit, out performs 64bit on OpenBSD, atleast in my experience (my hardware). And as was mentioned with the firewalls earlier. Run real world tests, where speed matters, and then we may care. Firewalling is one task, where speed really matters. You can always use multiple systems behind PF to increase speed too and still keep the security and stability and management time savings.
