Greetings!
I am getting my first RB333 /OS Version 3.x, and I want to verify that if using a SR2/XR2 the tx-power setting is now accurate. Or do you still need to compensate for the 10db offset?
Thanks!
EDIT: I know about the “leave the default power setting”. I need to turn it DOWN, not up.
In the New ROS it is automatically done, it will show you aswell, it will be like
18 | 28
14 | 24
12 | 22
11 | 21
And so on and so forth,. this will be seen when you set card to default it will tell you the output power pending on data rate. Hope this answers your question. -Jordan
Thanks, Jordan! That is the settings display in Winbox?
Do you know if the same applies to the CLI version? Does it show both power settings? Or do I need to compensate there? It looks above like Winbox is doing the compensation, not the OS.
Winbox it works, Not sure about CLI.
IM not sure if the compensation is done with router os, winbox has pretty much stayed the same.. Who knows Cough Mikrotik team Cough…
What makes me think there is a conversion prior to the O.S. entry is the list of power settings you posted has the -10db value next to it.
In the CLI, I don’t think I am supposed to enter
/interface wireless set wlan1 tx-power=14|24
am I?
Hopefully one of the MikroTik Team will come along and help out.
And just my two cents worth:
I design electronic devices, from DC to 10GHz. Is it not possible for the card designers to put a voltage limiter (zener diode or the like) on the power input (collector) to the final transistor stage that stops the voltage at the max for that transistor? I have not examined the details of the final amp stage, but that is what this setting appears to do.
The higher the tx-power setting, the higher the voltage on the final stage amp section.
If the transistor can take it, ok.
If not, the magic smoke comes out. You can never get it back in. From years of examination, it is this “magic smoke” that makes these things work. 
Or maybe Atheros and the other chip manufacturers should give the designer a way to set the max value for that setting so the user can’t set the value higher than that.
Im not sure about the CLI, I do not use it much unless im dealing with user-manager or hard time connecting to the routerboard. Have to ask the Mikrotik team about it. good luck 
-Jordan
BUMP!
I would really like an answer on this one. No response is not always “yes”.
Someone on the MikroTik Team…just a little help please?
FYU: And while I am waiting for that, my RB333 is here. Fast and smooth! No problems, mate!
Wouldn’t a zener cause clipping of the signal causing ringinging (noise) across the spectrum?
Would that not be a great way to get the fcc/crtc (or what ever other governing body) out to shut you down.
I have no papers, just common sense.
It shouldn’t if placed in the circuit correctly. This is in the supply voltage to the transistor, not the signal voltage. That would cause clipping, distortion, maybe a lot of spurious radiation, and the FCC visit. Normally, there is a current limiting device (resistor) in series with the transistor that limits max current. The voltage input to this device should be a filtered, stabilized DC supply. Limiting a smooth DC supply should cause no ill effects. But, like I said, I have not evaluated the circuitry yet.
I like the chipset solution better. Letting the designer set a limit on the value the user can enter for that input in the Atheros or ? chip would be optimum to me. They let you choose the amp. Why not the setting?
I understand that in the past there may not have been a need for this, because there were not people like us around then wanting to, and having the capability of, changing this setting. But now…here we are!
ADD: I had not considered this before but, now that you mentioned it, overpowering that transistor is just as likely to draw the attention of the FCC. That can’t be good for frequency stability!