> Turn off polling and what's the big advantage to Karlnet?
> This sucks. I guess it's time to deploy an 802.11b AP and
> start doing the new installs that way.
Turn off polling and you may be screwed even more by self interference from
hidden nodes talking into the AP. We've tried that and it doesn't work. No,
polling is what you need. But if you want to be independent of
manufacturer's policies, have a look at these two independent open source
projects:
http://frottle.sourceforge.net
http://patraswireless.net/software.html
Yes, these projects could benefit from optimizations like superpacketing and
they are not tied closely to the chipset driver - thus being less effective
than TC. But they let you combine a Linux box with a standard AP to build
your own royalty-free replacement for the Karlnet/Proxim devices, some day.
However, the open source polling approaches very likely also show some
latency jitter - this is (to some degree) in the nature of polling. From the
TC 4.43k release notes on the Karlnet website:
"The minimum polling rate in 4.43k is 1 Hz."
Which means you may see max PING latency up to 1000 ms from the CPE to the
base, unless there is other load that causes this CPE to be polled more
frequently. Have some traffic from the CPE to the base while doing PINGs to
see figures improve.
regards,
--
true global communications GmbH
Thomas Giger
In der Au 27, 61440 Oberursel, Germany
fon +49.6171.6381-0, fax +49.6171.6381-19
www.tgnet.de || www.megaspeed-internet.de
|