I have the same problem too, but using ubiquiti powerp-n.
The user "black6spdz" wrote this:
"most HP ubiquiti gear adds a +10db offset to the displayed TX level.. so 18db +/- 2db + 10db offset = 28-30dbm.. there is your 1000mW output"
I compared the airos 5.3.5 (@28db) and dd-wrt (@18db) in my home area and it work much better the dd-wrt, and where i have no signal, it happen with both. So I'm wondering if its just an annoying bug for eyes, because with my test it seem so. The user "black6spdz" he has so many ubiquiti device and I think his words are right, and maybe this offset bug happen also with other hardware when overtake 20db.
The power offset was implemented by airos because the airos madwifi driver was capped at reporting 18.
Ath9k? unknown at this point. If the offset was customized into the madwifi driver by ubiquiti, then why would ath9k/ddwrt have the same offset, unless it is hardware implemented.
DDwrt chris tested the output on both airos (modified madwifi) and ddwrt (ath9k) and said they were similar, and they should just release the db measurements to clear things up, and let the numbers speak for themselves.
One thing is for certain, ath9k has the same limitation of reporting at 18 dbm as madwifi.
The power offset was implemented by airos because the airos madwifi driver was capped at reporting 18.
Ath9k? unknown at this point. If the offset was customized into the madwifi driver by ubiquiti, then why would ath9k/ddwrt have the same offset, unless it is hardware implemented.
DDwrt chris tested the output on both airos (modified madwifi) and ddwrt (ath9k) and said they were similar, and they should just release the db measurements to clear things up, and let the numbers speak for themselves.
One thing is for certain, ath9k has the same limitation of reporting at 18 dbm as madwifi.
I'm sorry for my bad english... but i don't get the mean... is it just a bug and the real tx power in the antenna plug is 28db (don't count the 10db offset).. or is real a problem and the antenna tx power is just 18db instead of 28db as with airos?
One thing is for certain, ath9k has the same limitation of reporting at 18 dbm as madwifi.
The output power is set in the Atheros wifi chip but there is an external high-power amp between the wifi chip and the antenna.
A translation table (offset) is therefore needed so that the gui shows the real output power after the power amp. _________________ Kernel panic: Aiee, killing interrupt handler!
One thing is for certain, ath9k has the same limitation of reporting at 18 dbm as madwifi.
The output power is set in the Atheros wifi chip but there is an external high-power amp between the wifi chip and the antenna.
A translation table (offset) is therefore needed so that the gui shows the real output power after the power amp.