> -----Original Message-----
> From: tclug-list-bounces at mn-linux.org
> [mailto:tclug-list-bounces at mn-linux.org]On Behalf Of Mike Miller
> Sent: Monday, April 04, 2011 5:56 PM
>
> On Tue, 29 Mar 2011, Chuck Cole wrote:
>
> >> I may have happened to read the question wrong also, so I will answer
> >> the alternate (Can you achieve the same /internet/ speed with WiFi and
> >> wired).
> >
> > That *IS* the question.
> >>
> >> If you receive ~15Mbps from Charter but you only see 7.5 via WiFi you
> >> are most likely on a busy WiFi channel. Try different channels, or
> >> perform a site survey and make sure to not choose a neighbors channel.
> >
> > Data given above stated "clear channel with high S/N"
> >
> > Are you saying that I should get the same ~15Mbps on WiFi as LAN per
> > Charter's speed test?  I think I should, unless there's some detectable
> > overhead. that I have not been able to identify or detect.
>
>
> I am curious about this.  I haven't done any careful testing, but I have
> always had the impression that my wifi internet connection was
> slower than
> my wired connection.  It would be nice to hear that Chuck figured this
> out.  Are we really out of ideas for him?
>
> Mike

I did further testing and verified that the problem doesn't seem to be a
limit or fixed characteristic of my local hardware.

Found a freeware utility for windows that does bidirectional tests pairwise
between computers on the local network.  This tool generates a large dummy
file, and times 2-way transfers to/from a disk, then cleans up its mess.
This test tool shows nearly 100mbps between wired pairs, and nearly 54mbps
between a wired and a WiFi pair.  Doesn't seem to be much difference between
having only one of my WiFi boxes active and having two active.  These
results seem to show that I should get about the same download speed on my
WiFi as on my LAN connections, since the WiFi links have more than about
twice the internet download speed.

Wild guess: some protocols have short and long transfer modes that differ by
block lengths and thus the packetizing overheads.  MAYBE the internet test
from Charter I was using makes the WiFi link use a short transfer mode and
not a more suited long transfer mode.

Quite a puzzle..


Chuck