The Natural Philosopher  wrote:
> On 06/07/2019 20:34, Michael J. Mahon wrote:
>> The Natural Philosopher  wrote:
>>> On 05/07/2019 22:35, mark lewis wrote:
>>>> On 2019 Jul 06 13:39:00, you wrote to Ahem A Rivet's Shot:
>>>>
>>>>>> or you have a lot of interference.
>>>>
>>>>> No interference at all. Nothing within 300 meters of my house.
>>>>
>>>>> Just metal walls and ceilings :0(
>>>>
>>>> flourscent lights?
>>>>
>>>
>>> Nope. Not even fluorescent ones!
>>>
>>> Even the LV are toroidal. Not SMPS
>>
>> I suspect that your problem with Wi-Fi is not interference in the
>> conventional sense, but interference in the sense of constructive and
>> destructive interference.
>>
>> The metal walls and ceilings create a cavity excited by Wi-Fi transmitters,
>> creating myriad regions of high and low signal strength separated by
>> inches.
>>
>> Since Orthogonal Frequency Division Multiplexing uses multiple frequencies,
>> each frequency will have a different spatial distribution of high and low
>> signal, making coordinated use of multiple sub bands unreliable.
>>
>> In a building with smaller and less speculative reflections, such
>> interference is not problematic, but in your case it is rendering OFDM
>> almost unusable. The MacBook exception is probably a consequence of its use
>> of MIMO antennas which a much more difficult in a smaller form factor.
>>
> That is the first half believeable scenario anyone has come up with.
>
> However it doesn't explain massive fluctuations from two totally
> stationary systems with any movement - me - in another room altogether.

Actually, it could. It is unlikely that the metal in the walls makes
contact with adjoining walls everywhere. The result is “slot coupling”
which would allow the maxima and minima in an adjoining room to move some
in response to changes in the environment of the first room.

> Or why dumping wifi and re-enabling it takes  the laptop from 1Mbps to
> 58Mbps

This could be the result of the laptop “sticking” to the first channel that
permits good communication, which deteriorates, but remains “stuck” until a
restart of its Wi-Fi results in locking onto the new “best” channel.

A complex algorithm operating in a complex, changing communication
environment can lead to unexpected behaviors.

--
-michael - NadaNet 3.1 and AppleCrate II:  http://michaeljmahon.com

--- SoupGate-Win32 v1.05