I have one PTP link (OS 4.11 + nstreme) and it does not have a clear fresnel zone clearance hence poor CCQ, but what i found there is a hugh difference between the P Throughput reading at the AP Bridge side and Station mode side of the link,(see screen shot)
Ap bridge side = 7359kbps , Station side = 20769kbps, would have thought that a misaligned link or fresnel issues or both would give a lower than max throughput on the link but this would equal when monitoring from each end? 
To see the REAL possible thrhohughthput (xD) you need to make a bandwidth test between the PTP. You can select 3M only and set it to TCP both. With this you will get the real speed.
If it’s different, then tx and rx signal must be different too.. and it’s caused for the bad LOS or interference.
You can select 3M only and set it to TCP both. With this you will get the real speed.
Can you explain howto?
Would i be correct in this alignment method, at locations which do not have fresnel or LOS issues, is to monitor both sides of a link and when P Throughput is equal at both, then alignment must be good?
tool> bandwidth test
mark TCP put the IP of remote AP
fill user/password of remote AP
Select both in the combo of direction
and in TX and RX speed write 3M or 30000000
If you click start you’ll see the 3Mb traffic.. If you see more or less speed, play with the Zeros in TX RX speed ; )
Then the P throughput will be more acurate
One reason for the possible P-throughput value of 20,769 kbps on one radio and P-throughput value of 7,359 kbps could be that the 7,359 kbps radio is connecting at a lower modulation speed rate (6 Mbps) and the 20,769 kbps radio is talking at a faster modulated rate (35 Mbps).
As a test - if you have the time - try these settings:
#1 on both radios set your supported and basic rates to 6 Mbps - then add 24 Mbps supported.
(Basic Rates 6)
(Supported 6 & 24)
(((note - 6 is for best connection under noise and 24 is for fastest rates in both directions without seeing much RF loss due to “faster rates = RF loss in wireless card sensivity & xmit power”
#2 in the Nstream section - place a check in “Disable CSMA”
(We do not want to use CSMA)
#3 make sure DFS mode is not turend on
#4 in Wireless - advanced
Ack-Timeout = dynamic
Periodic Calibration = enabled
Calibration Interval = 00:10:00
HW Retries = 15
Hw prottection mode = none
Adaptive Noise Immunity = none
Preamble Mode = long
Disconnect timeout = 00:00:09
On Fail Retry Time = 900
Give this a try - let it stay connected for about 15 minutes and see what your P-Throughput and your CCQ is. If this is any better - then only make one change at a time to see if you get better or worse on your link & p-throughput in both directions.
please let me know what you got for your results.
One reason for the possible P-throughput value of 20,769 kbps on one radio and P-throughput value of 7,359 kbps could be that the 7,359 kbps radio is connecting at a lower modulation speed rate (6 Mbps) and the 20,769 kbps radio is talking at a faster modulated rate (35 Mbps).
OK but strange i tried two different units on the AP side of the PtP link and both which were a short distance (approx 50 meters ) apart but also had fresnel issues gave lower throughput than from station side?
As i am using parabolic grids i will rotate the diople 180°, just curious to see if fresnel zone lobes are linear in my case horizontal polarity,and checking any changes on CCQ and throughput,
Attached (1) datarates at 24Mbps just to show is my config correct for this, (2) at 24Mbps AP is on the left, (3) at 6Mbps AP is on the left
#4 in Wireless - advanced
Ack-Timeout = dynamic
Periodic Calibration = enabled
Calibration Interval = 00:10:00
HW Retries = 15
Hw prottection mode = none
Adaptive Noise Immunity = none
Preamble Mode = long
Disconnect timeout = 00:00:09
On Fail Retry Time = 900
Applying these tweaks did not improve throughput but did make the wireless association (R) much faster on dude on both 6 and 24Mbps datarates and quicker to log into units at the other end of the link,
Many thanks
