I got a few dodgy results, but I think i've figured out how to do it.
You'll need to do it all from a terminal - can't do it in winbox. I tested this with a 2511MP in an RB11 MiniPCI bridge on an Intel D815EGEW motherboard with a Celeron 900Mhz processor. The transmitting 2511MP had a 6" pigtail and isotropic "test" antenna.
The receiving unit was an RB532 with a CM9, 6" pigtail, and a 3db omni antenna. Throughout my testing I did not adjust any settings or placement of the CM9.
Anyways, from a terminal, I did this:
[admin@Mikrotik] > /interface wireless
[admin@Mikrotik] interface wireless> set wlan1 frequency-mode=manual-txpower
[admin@Mikrotik] interface wireless> set wlan1 tx-power-mode=all-rates-fixed
[admin@Mikrotik] interface wireless> set wlan1 tx-power=17
Doing this (in theory) reduces the TX power
from 23dBm to 17dBm (specify tx
in dBm). My received signal on my test bench went from -31 to -41 with pretty good consistancy in results. Setting tx
to 10dBm resulted in received signal of -51, tx
=5 resulted in -54 or so and tx
=1 resulted in -50 to -54. It's pretty clear that there isn't much difference at lower TX
powers, but that maintains the rules behind output power
Remember that 20dBm is half the power
of 23dBm, and 17dBm is half the power
of 20dBm. If you're used to talking in mW here's a short table:
26dBm = 400mW
23dBm = 200mW
20dBm = 100mW
17dBm = 50mW
14dBm = 25mW
Please do your own testing to verify if you can duplicate results! This appeared to work in my lab but I have NOT tried anything real world!
Apply salt, your mileage may vary and all that.