[MVMA] iperfspeed: It measures throughput, not latency

[Chuck Gelm]

and while I did not have iperfspeed installed,
I loaded iperfspeed_0.5.1_all.ipk on several nodes (64 MB RAM)
and checked a few paths.  Of note...
*The Dayton Mall to Xenia RF path of ~33 miles has a latency of ~ 6 
milliseconds.*
Excellent! Likely better than the average 'tunnel' path.

  Throughput:

  Most of our backhaul (a) links have speeds of 10 to 100Mbps.
However, some of our downlinks to users are using single polarity 1/2 
bandwidth devices
(pre-2017 there were fewer(b) available devices and we didn't know any 
better ).
So, throughput on downstream user nodes on 2.4 GHz is in the Megabit 
range yet, still,
with very low latency. Worst case; a single 1920x1080 video at 5 fps or 
20 audio(c) circuits.
Please, let us eschew statements that indicate that 2.4 GHz is for user 
access.

(a) Using 'backhaul' opposed to 'backbone'.
In my mind, backbone networks have every node on the same channel.
I abhor node hopping on the same channel.

(b) There were about a dozen available devices pre-2017. Station cost 
was $80-$400.
There are now over 80 (AREDNable) devices and the majority are < $130.

(c) low latency permits flicker free video and dropout free audio.


-- 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/mvma/attachments/20201203/8f623f90/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: sig.gif
Type: image/gif
Size: 4699 bytes
Desc: not available
URL: <http://mailman.qth.net/pipermail/mvma/attachments/20201203/8f623f90/attachment-0001.gif>