James G. Sack (jim) wrote:
Lan Barnes wrote:
Now that Josh has suggested a way forward on my MythTV angst, I can return
to my other CB HW challenge -- the wireless on Alex's new laptop.
I could inundate you with all the experiments and research so far done,
but it really comes down to two more steps. The machine is now in Fedora 8
("F8"? sounds like a function key). I need to hand load the driver and see
is I can get it up. If I succeed, I need to test whether it, too, sufferes
from the original problem, inability to pick up a signal from the router
unless it's in the same room.
Any hints on testing to see signal stength received?
What's the device name?
/sbin/ifconfig -a
(is it wlan0, maybe?)
/sbin/iwlist wlan0 scan
If this gives
wlan0 No scan results
try
sudo /sbin/ifconfig wlan0 up
(this should make the word "UP" appear in the output of
/sbin/ifconfig wlan0
I'm told that the signal strengths reported by wireless devices is
somewhat (largely) uncalibrated -- possibly they should be only taken as
relative strengths, and only meaningful for _that_ particular hardware.
Maybe others have more scoop on that?
Anyways,
iwlist [device] scan
is the short answer.
Regards,
..jim (never a short answer)
Yes. That's been a long ongoing topic on both the hostap/wpa_supplicant
and networkmanager dev lists. How signal strength is measured,
calculated and reported varies from chipset to chipset, with no
consensus as to the degree of accuracy of any of them. It's seems to
fall into the full vs. empty category: Either you get a usable signal
our you don't. There is no empirical basis behind the meaning of bars.
In practice, not too different from a cell phone.
--
Best Regards,
~DJA.
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
