Larry Finger <(E-Mail Removed)> hath wroth:
>Eric wrote:
>Exactly. A good rule of thumb is that if your interface is set at M Bps, you will be able to
>transmit at about .5 to .6 M. My actual number, measured with Iperf, is 6.5 Mbs with a nominal rate
>of 11 Mps.
>Larry
You have 2 thumbs. Therefore two rules will work. The first rule is
50% of connection speed or less (not more). The 2nd is that all
statistics and benchmarks lie in the optimistic direction.
My calcs for an 11Mbit/sec association show 5.9Mbit/sec maximum:
<http://wireless.wikia.com/wiki/Wi-Fi#Performance_and_Speed>
for TCP traffic. You can go up to 7Mbit/sec using UDP, but I've never
seen it using Iperf. Most of my typical field tests (with reflections
and interference) yield about 3.5Mbit/sec with an 11Mbit/sec
association.
Did you really get 6.5Mbit/sec with Iperf? If so, do you recall the
command line incantation or parameters (packet size, TCP/UDP, etc)?
No points for setting up 802.11b/g and just noticing it was running at
11Mbits/sec. I might easily switch momentarily to a higher speed and
wreck the benchmark. You gotta fix the speed in the access point if
you want a meaningful test.
--
Jeff Liebermann
(E-Mail Removed)
150 Felker St #D
http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558