I sent files using lpr to network printer both on 10BaseT and 100baseT
network. I found that the data transfering speed on 10BaseT network was
almost three times faster than 100base T network. That sound wired,
right?
Later on I used NetXray to trace the packet sizes of both network and to
find 100Base T got smaller packet than 10Base T. My understanding was
that 100BaseT is coming in a lot of short(but fast)packets, thus yield
better performance than 10Base T. But according to what I observed from
the test, it was not that kind of story. I am not sure if it is the
smaller packet which yield undesireable performance or not. If so,
whether the small packets are the result of the host or network card?
Any ideas?
Randy B. Lim