Mikrotik Btest, but testing with CCR's on either side of the Dynadish.

TCP, 20 connections, testing downlink direction only.


On 9/11/2020 10:29 AM, Matt Corcoran wrote:

Are you using the built in microtik speed test or are you testing through from an external device?   If using the microtik speed test what are  TCP,  UDP, Up-only,  Down-only test results?          Externally I use iperf so I can try with multiple streams and know how exactly many streams the test is using.  What is latency?   Maybe we can get a better ideas whats going on.

*From: *AF <af-boun...@af.afmug.com> on behalf of Adam Moffett <dmmoff...@gmail.com>
*Reply-To: *AnimalFarm Microwave Users Group <af@af.afmug.com>
*Date: *Thursday, September 10, 2020 at 2:01 PM
*To: *"af@af.afmug.com" <af@af.afmug.com>
*Subject: *Re: [AFMUG] MT DynaDish throughput

This is in the back woods. SNR is reported as 39db.  Whether we believe that's accurate is possibly up for debate, but the physical rate stays pretty consistently at 390Mbps and CCQ stays above 90%.

I'm not sure what wall I'm hitting on real throughput though.  CPU only gets to 25% or so during the test.  The Ethernet interfaces are gig.  I can get 200Mbps+ testing to the near end of the link, but 90-97 testing through it one-way.

On 9/10/2020 1:55 PM, Jaime Solorza wrote:

    80Mhz wide? No interference? What's your SNR?

    On Thu, Sep 10, 2020, 11:52 AM Adam Moffett <dmmoff...@gmail.com
    <mailto:dmmoff...@gmail.com>> wrote:

        I don't normally do Mikrotik Wireless, but I inherited a
        DynaDish 5
        link.  It's at -66 RSSI on each end, 80mhz channel size.

        The physical rate is shown as 390Mbps in both directions, but
        actual
        downstream throughput seems to vary from 90-97Mbps.  Is this
        normal for
        the product?  Is there some known optimal config for a
        Mikrotik PTP?

        I fiddled with using 802.11 vs nstream vs nv2, and I got that
        90+Mbps
        result on nstream.  The other modes seemed to be worse. I'd
        hate to
        spend any more time fiddling if this is the expected outcome
        with this
        product.



-- AF mailing list
        AF@af.afmug.com <mailto:AF@af.afmug.com>
        http://af.afmug.com/mailman/listinfo/af_af.afmug.com




-- 
AF mailing list
AF@af.afmug.com
http://af.afmug.com/mailman/listinfo/af_af.afmug.com

Reply via email to