sensitivity differences of rt61pci and rt61 legacy in kismet

Live forum: http://rt2x00.serialmonkey.com/viewtopic.php?t=4914

welm

27-07-2008 06:46:41

Hi,

I've used Suse 10.3 with the rt61 legacy driver and now I'm using Suse 11.0 with the rt61pci driver. Both on a workstation with a Conceptronic PCI card (connected to a 5dBi-antenna) as well as on a Laptop with the corresponding PCMCIA card.
After the OS upgrade I noticed a significant decrease in the number of visible networks in kismet on the workstation (1-2 instead of 15-20), but there is no significant effect on the laptop. When I recompile the legacy driver and use it on the workstation again, the number of visible networks is again above 15. Going back to rt61pci leads to 1-2 visible networks again.

Is this a known phenomenon?

W.

lspci Network controller RaLink RT2561/RT61 rev B 802.11g
rt61pci 2.0.14
rt61 rt61-cvs-2008013100
kernel 2.6.25.11-0.1-pae

Zi7

27-07-2008 14:44:11

Hi,

[quote5irxfprg]After the OS upgrade I noticed a significant decrease in the number of visible networks in kismet on the workstation (1-2 instead of 15-20), but there is no significant effect on the laptop.[/quote5irxfprg]
So basically, you're saying a 5dBi antenna is working negatively with rt61pci, right?

I'm not sure what you mean about the laptop though you see as many APs with rt61pci and legacy?
How many, compared to the workstation?

welm

27-07-2008 18:30:27


[quotejtrg5uxm]After the OS upgrade I noticed a significant decrease in the number of visible networks in kismet on the workstation (1-2 instead of 15-20), but there is no significant effect on the laptop.[/quotejtrg5uxm]
So basically, you're saying a 5dBi antenna is working negatively with rt61pci, right?
[/quotejtrg5uxm]

No, the normal antenna is not better.

Before I tried the old driver again, I thought the antenna cable was broken.


I'm not sure what you mean about the laptop though you see as many APs with rt61pci and legacy?
How many, compared to the workstation?[/quotejtrg5uxm]

Around 8 networks (but of course heavily depending on the position of the laptop). The interesting fact is, that changing the driver didn't make any big difference on the laptop while the same transition changed the result on the workstation completely.

Hmmm, if the chip had two antenna pins and if both were connected to the antenna in the PCMCIA card but only one connected to the socket in the PCI card and if the drivers would select a different pin by default, then the things I've seen should occur. Wow, Many "if"s.

Welm

welm

30-07-2008 08:52:10


Around 8 networks...
[/quote39v9zv9f]

I've done the checks again with a constant time and antenna position to get more comparable numbers
Laptop 6 with rt61pci and 5 with the rt61 legacy driver
Workstation 2 with rt61pci and 17 with the rt61 legacy driver


Hmmm, if the chip had two antenna pins and if both were connected to the antenna in the PCMCIA card but only one connected to the socket in the PCI card and if the drivers would select a different pin by default, then the things I've seen should occur. Wow, Many "if"s.
[/quote39v9zv9f]

That seems to be wrong -(
I've fiddled around with code and found that rt61pci_config_antenna_2x / case ANTENNA_A is used. But hacking the code there to that of ANTENNA_B or ANTENNA_HW_DIVERSITY didn't improve anything.

W.
--
There is always an easy solution to every...problem -- neat, plausible, and wrong. (H L Mencken)