<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<p>Cameron Donaghey wrote:
<blockquote TYPE=CITE> Hello All,
<br> Ok got stuck into it and did some various testing last night.
It seems that I did overlook that
<br>fact that transferring client->AP->client halves the bandwidth because
the network is shared.
<br>The maximum I can achieve with client->AP->client is 282kB/s which
allowed me to transfer
<br>a 20meg file in 1min11secs. This from what I now understand is quite
good.
<p>I then decided to take the access point out of line and try the two
clients peer to peer ( adhoc )
<br>or client->client mode and see what they could achieve by themselves.
<br>The maximum I could get with the same 20meg file was 569.05kB/s which
allowed me to
<br>transfer the file in 35.5secs.
<p>I then also tried running one client to the access point connected to
the second machine
<br>via its ethernet connection so client->AP->wired.
<br>The maximum I achieved from this was 546.6kB/s and I transferred the
20meg file in 37.9sec.
<p>So from all of this I think my setup is running how it should be although
I still don't entirely
<br>understand when going client->AP->client the transfer speed is halved
when compared to
<br>the client->client, perhaps someone could explain or perhaps this is
not correct.
<br> </blockquote>
client->client is one rf transmission of the data,
<br>client->AP->client requires two rf transmissions of the data, the total
rf bandwidth is
<br>still the same (assuming a single channel AP) so you get half the throughput.
<br>
<p>Regards
<br>Geoff
<br>
<br>
<blockquote TYPE=CITE>
<br>Thanks for everyone's replies,
<br> Cameron
<p>At 10:35 AM 9/17/2002 +1000, Geoff Smith wrote:
<blockquote type=cite class=cite cite>I may have mis-read the original
post, I thought they were
<br>attempting client->client transfers not client->AP->client transfers.
<br>My apologies if my post mislead anybody.
<p>Here is another link:
<br><a href="http://support.countryday.net/Wireless/Performance/Wireless%20Efficiency.htm" eudora="autourl">http://support.countryday.net/Wireless/Performance/Wireless%20Efficiency.htm</a>
<br>it shows the lucent wavelan card getting 4.93Mbps, (616kB/s), for
<br>client->AP->wired transfers
<p>client->AP->client transfers with an AP with a single channel radio
will be
<br>sharing
<br>the same bandwidth between the two clients (i.e. half again) but more
<br>importantly
<br>its very sensitive to the capabilities of the AP, embedded os, cpu
speed, memory
<br>size
<br>for queueing and buffering etc etc; i.e. implementation issues. In
addition the
<br>people who
<br>designed the AP probably tuned it for client->AP->wired connections
<br>not client->AP->client connections.
<p>If the orginal post meant kB, then he was getting 2.4Mbps which is pretty
<br>impressive
<br>and would represent 21% of the total bandwith
<br>If the original post meant kbits then he was only getting 2.6% of the
total
<br>bandwidth
<br>for a client->AP->client test compared to 44.8% achieved by the lucent
card in a
<p>client->AP->wire test, or 27% achieved by the wl100 card in client->AP->wire
<br>tests.
<p>So if Cameron is doing client->AP->client, perhaps he can tell us what
<br>he is getting for a client->AP->wire test?
<p>Regards
<br>Geoff
<br>
<br>
<p>Craig Mead wrote:
<p>> | Do you mean kB, k bytes or kb, k bits
<br>> | 310kbytes -> 2.4Mbps
<br>> |
<br>> | Theoretically you should be able to get about half of the 11Mb
<br>> | bandwidth.
<br>> | ~ 5.5Mbps or ~680 kbytes/s
<br>>
<br>> Doing client to client transfers, this is extremely innacurate. The
11b
<br>> protocol has some massive overheads which consume a high percentage
of the
<br>> possible thruput. The highest I've ever seen a link running personally
is
<br>> 570k/sec (client to wired server behind AP). Though theoretically
from this
<br>> you should then be able to get ~ 285k/sec client to client thru AP,
which is
<br>> a lot higher than the 50KB/sec max I've obtained.
<br>>
<br>> I had a talk to some of the boys in Melbourne about this issue and
they said
<br>> it also depends on the placement of the end nodes in relation to
each other
<br>> and the AP for some reason. Not sure about the technicalities of
how this
<br>> would effect it, but they said it did. As soon as I get a few more
links on
<br>> I'll be able to do some more testing.</blockquote>
<br>
<br> </blockquote>
</html>