I realised that I had "Private Address" turned on on the iOS devices - that randomises the MAC address and I'm using IP-MAC binding

I discovered that option on my other half's new Samsung. Who's stupid idea was it?

Piste Basher wrote:
I think I would generally agree with what they say there apart from No 6

Yes, point taken. The fact that your phone's Wifi still works, when you're stood right next to the mighty AP903, suggests it's not an issue :wink:

I also notice when playing around with mine that the settings on the AP903s seem to have a mind of their own - e.g. I set WMM Capable as ON (as per Apple instructions) but when I check again some time later I find that it is OFF again

Yup, I believe I've seen that one.

I have an interesting DNS issue going on...
Sometimes the browser on the phone won't resolve a local name (I have a home-brew IOT device with a web interface). Yet, said DNS server is 'pingable' from phone ... it's almost as though I've got DNS traffic from it, blocked by a firewall somehow. Yet toggle Wifi Off/On and it suddenly decides it's going to work

Going back to these Power Levels...
I realised they are actual 'Power' levels, so a chart like this: http://wireless.ictp.it/school_2002/docs/howto/dBm%20to%20Watts%20Conversion.html , reveals that altering 'Power' from 100% to 10% is a 10dBm drop, so 29dBm to 19dBm (ish).

Then I started thinking - the highest received Power level I've seen is -35dBm ... I've not allowed for all the losses along the way

I wondered if the info on this Huawei web page would help me understand it - but no joy so far ... maybe I just need to keep re-reading it.

I can't get my head around the basic question - "Would halving (say) the TX power, result in me seeing a signal that is half its original strength?" ... on either a linear or a logarithmic scale. As an example that I find easier to visualise; if the house at the end of street doubled the power of their porch light, would the light hitting my house be 'twice as bright'? I know there's an inverse-square law involved in this - I wonder if that's the most pertinent factor?

(I think my A-levels must have expired or been revoked, 'cos they don't appear to work any more )

That Huawei paper gives an interesting summary - except for

For example, a 10 W bulb consumes 10 W of electric power when it is on for 1 hour.

err - no, it consumes 10W immediately and continuously - if on for 1 hour it consumes 10Wh or 0.01 kWh

but getting back to the power question - yes in ideal "free space" conditions there is an inverse square law involved which is based purely on distance, but in practice here on earth there are many other things which can attenuate the signal between the transmit aerial and the receive aerial. Not only walls and windows etc. but also, at these GHz frequencies, gases, particles and vapours in the atmosphere as well. If you also consider reflections from objects and hence the possibility of multiple paths things get even more complicated...

But yes, all other things being equal a doubling of transmitter power (eg 1W to 2W) should double the strength of the received signal (i.e. increase it by 3dB ). I notice in that paper they mention the possibility of automatic power control, but I don't think I've seen any mention of that in the Draytek literature.

Not sure I can help with the DNS problem :?

Oh - and the "Private Address" change has not resolved my "black hole" IP address issue