If you don’t need a wide channel, it’s not clear you should use a wide channel. 
 People tend to forget that both the external interference and the receiver 
noise floor go up as the channel width increases.  I see this all the time with 
home routers.  People get talked into a $200 AC router and setting them up for 
the widest channel possible to get the highest speed and range.  And of course 
using 5 GHz and trying to go through multiple walls.  And they end up with 
terrible performance, and would be better off with a 20 MHz channel, and 
putting their portable devices on 2.4 GHz and reserving 5 GHz for video devices 
in the same room with the router.

256QAM and 80 or 160 MHz channels can work against each other.  In many cases 
it’s one or the other, not both.

But in a perfect world ...


From: Eric Kuhnke via Af 
Sent: Sunday, November 02, 2014 4:39 PM
To: [email protected] 
Subject: Re: [AFMUG] rocket ac lite performance

Relative to a Rocket M5AC PTP setup, or Mikrotik 802.11ac boards & good quality 
dishes (Jirous 32dB high performance), it is more expensive.


However, as a 256QAM radio when compared to a PTP600 or PTP650 connectorized, 
it's much less expensive. 

I guess the question cannot really be answered yet, since the people who are 
beta testing MIMOSA radios are still under NDA. When they're available to 
purchase we'll know how they perform in a fixed TDD channel vs. much less 
expensive competing radios.



On Sun, Nov 2, 2014 at 2:27 PM, Mike Hammett via Af <[email protected]> wrote:

  I operate in suburban Chicago and touch at least 12 WISPs across my 
footprint. I've seen noise. One site has a -75 noise floor on my sectors (it 
actually even goes up to -50 once we go above 5850).

  I'd imagine doing anything other than their auto-everything will cripple your 
experience, especially the simultaneous DFS channels.

  Might as well get Mikrotik or Ubiquiti at that point.

  Might want to step up the dish size or quality. Check out Jirous dishes. 
They've given me some excellent SNR.




  -----
  Mike Hammett
  Intelligent Computing Solutions
  http://www.ics-il.com





------------------------------------------------------------------------------

  From: "Eric Kuhnke via Af" <[email protected]>
  To: [email protected]
  Sent: Sunday, November 2, 2014 4:01:42 PM
  Subject: Re: [AFMUG] rocket ac lite performance


  Because in an urban environment frequently the widest available "clean" 
channel you can find is a single 20 MHz of spectrum between two locations. All 
of the FDD-like features of the Mimosa radio are useless if one of the channels 
you want to use has a -82 noise floor.


  Very different use scenario for city vs. rural.




  On Sun, Nov 2, 2014 at 1:43 PM, Mike Hammett via Af <[email protected]> wrote:

    Why would you pull half of the spark plugs on a Ferrari?




    -----
    Mike Hammett
    Intelligent Computing Solutions
    http://www.ics-il.com





----------------------------------------------------------------------------

    From: "Eric Kuhnke via Af" <[email protected]>
    To: [email protected]
    Sent: Wednesday, September 24, 2014 9:03:18 AM
    Subject: Re: [AFMUG] rocket ac lite performance


    I wonder how the performance of the Mimosa 256QAM product will compare in a 
20 MHz wide, TDD channel when used with the same antennas, head-to-head against 
the Mikrotik board.  In a scenario not using any of the special frequency auto 
selecting features of the Mimosa/Quantenna chipset.  


    I have in mind a setup with a pair of the Jirous 32dB high performance type 
antennas with a metal enclosure on the rear.


    On Wed, Sep 24, 2014 at 6:53 AM, Stefan Englhardt via Af <[email protected]> 
wrote:

      The Sextant did not work for us at all. The QRT-5 has a quite good 
Antenna for it’s size.

      But you cant upgrade/repair them as they use special screws you cant open 
without damaging

      them.



      The Mars-Antennas have a metal housing so there is some shielding to the 
back. We don’t like

      the outdoor-pigtails/connectors as with the rocket / ePMP-Force / ePMP 
Sectors. Just one

      Ethernet leaving the housing to the bottom is the solution we see the 
least problems.





      Von: Af [mailto:[email protected]] Im Auftrag von Mike 
Hammett via Af
      Gesendet: Mittwoch, 24. September 2014 15:40
      An: [email protected]
      Betreff: Re: [AFMUG] rocket ac lite performance



      I remember when they announced their new "high gain CPE" the Sextant...  
which was still smaller than the smallest CPE I used.



      -----
      Mike Hammett
      Intelligent Computing Solutions
      http://www.ics-il.com




--------------------------------------------------------------------------

      From: "Glen Waldrop via Af" <[email protected]>
      To: [email protected]
      Sent: Wednesday, September 24, 2014 8:37:54 AM


      Subject: Re: [AFMUG] rocket ac lite performance

       
      I can't get over the small gain MT uses.

      It also bugs me that the FCC seems to be all about massive amps and small 
antennas rather than the reverse.

      If it was actually about interference PTP shots with narrow beamwidth is 
preferred. I suppose it is too much to ask for those in our government that set 
the regulations to actually understand the tech they're regulating.

      I suppose it could be the manufacturers are going with big amps and small 
antennas, but it seems that would cost more.





        ----- Original Message ----- 

        From: Mike Hammett via Af 

        To: [email protected] 

        Sent: Wednesday, September 24, 2014 8:33 AM

        Subject: Re: [AFMUG] rocket ac lite performance



        The SXT antenna were always too weak. Give me 25 dBi or give me death. 
Well, okay, I don't feel that strongly. I just won't buy it if not.



        -----
        Mike Hammett
        Intelligent Computing Solutions
        http://www.ics-il.com




------------------------------------------------------------------------

        From: "Stefan Englhardt via Af" <[email protected]>
        To: [email protected]
        Sent: Wednesday, September 24, 2014 8:30:07 AM
        Subject: Re: [AFMUG] rocket ac lite performance

        We’re starting the move to .ac with MT now. The 922UAGS-5HPacD has the 
same dimensions as

        the 911/411 Boards so we just replace boards.



        We get good results with Mars Antennas with housing for P2P. The 
SXT-Antennas are to weak for .ac.

        The 19db Mars Antennas give good CPE/short range PTP with a small 
footprint. We use these as

        Sector-Antennas where we have to cover small areas.



        The .ac firmware adaption is quite new but we see stable results in the 
300-400 Mbit/s range for short links.

        The .ac boards have faster CPUs so they may increase 11n-Speeds/NAT 
Performance.

        The 922-Board has a SFP. Ethernetport has moved. Due to this the 
RFElements Stationbox XL do not fit.



        MT .ac does PTP and PTMP and is downward compatible to older boards 
with 11n/a.



        SXTs with .ac are not stable with the latest sw-release. So as always 
with MT you’ve to betatest

        HW/FW-combination to get it running smooth.







        Von: Af [mailto:[email protected]] Im Auftrag von 
Rory Conaway via Af
        Gesendet: Mittwoch, 24. September 2014 14:26
        An: [email protected]
        Betreff: Re: [AFMUG] rocket ac lite performance



        Yeeeeaaaaa, I don’t’ think so.  If you use Ubiquiti you pretty much 
know what works and what doesn’t.  in reality, you use as few  custom features 
as possible outside 802.11 compatibility and limit the radios to Layer 2 
bridges or nothing more than NAT whenever possible.  Try not to use any of the 
customer features although AirMax seems to be working pretty well.   You just 
don’t want to add anything that adds to processor overhead on an AP for Rocket 
M5’s if you have a higher density.  When the Titanium’s tanked a couple years 
ago, there was a huge hole in any kind of AP product with the ability to handle 
density due to the processor but nobody filled it.  It still hasn’t been filled 
to my satisfaction meaning we aren’t replacing Rocket 5M’s any time soon and we 
are keeping them at 50 users or less for another few months.  However, NetFlix 
and video streaming means we are pushing that down to a planned 30 over the 
next couple of months.  That numbers are just estimates but somewhere between 
30-50 under heavy video streaming usage and AirMax will start causing issues.



        If you allow torrents or anything that opens up a massive number of 
connections, then the number of users per AP drops significantly which is why 
we run filtering on the back end to reduce that.  I’ve seen APs with less than 
30 go apoplectic with a couple of wild torrent users.



        Rory  



        From: Af [mailto:[email protected]] On Behalf 
Of That One Guy via Af
        Sent: Tuesday, September 23, 2014 11:47 PM
        To: [email protected]
        Subject: Re: [AFMUG] rocket ac lite performance



        :-) the ubntboys tend to not be all that informed they blindly swear by 
whatever the spec sheets and feature notices tell them



        On Wed, Sep 24, 2014 at 12:44 AM, Josh Reynolds via Af <[email protected]> 
wrote:

        For the first question, I have no idea, since the only released radios 
are PtP at this time. Pretty hard to tell.

        For the second one, "ubntboys" (at least the informed ones) don't run 
airsync, because they know it doesn't work well. :)

        Josh Reynolds, Chief Information Officer
        SPITwSPOTS, www.spitwspots.com

        On 09/23/2014 09:06 PM, That One Guy via Af wrote:

          does it work? or does it work in ubntboys eyes like airsync?



          On Tue, Sep 23, 2014 at 11:12 PM, Josh Reynolds via Af <[email protected]> 
wrote:

          For clarification for EVERYBODY reading this post, ALL AIRMAX AC 
PRODUCTS HAVE AIRMAX OFFLOADING :)

          Josh Reynolds, Chief Information Officer
          SPITwSPOTS, www.spitwspots.com

          On 09/23/2014 07:33 PM, Mathew Howard via Af wrote:

            The ptp-lites do have airmax offloading? I thought they didn't...

            We've had a link running for awhile, but it's got some noise 
issues, so it hasn't really been the best place to test these... that said, 
they've been running fine without any real problems that I've noticed.
            �


--------------------------------------------------------------------

            From: Af [[email protected]] on behalf of 
Josh Reynolds via Af [[email protected]]
            Sent: Tuesday, September 23, 2014 9:49 PM
            To: [email protected]
            Subject: Re: [AFMUG] rocket ac lite performance

            ptp lites are missing the internal shielding (i think?) as well as 
airprism tech

            they do have airmax offloading, but it's also a brand new product 
line... some bugs are showing in the software

            Josh Reynolds, Chief Information Officer
            SPITwSPOTS, www.spitwspots.com

            On 09/23/2014 06:48 PM, That One Guy via Af wrote:

              These are basically beta release hardware? its missing some guts?



              On Mon, Sep 22, 2014 at 9:04 PM, Rory Conaway via Af 
<[email protected]> wrote:

              I don�t have one but from what I�m reading, it�s not quite 
ready for primetime.� I�m waiting.

              �

              Rory

              �

              From: Af [mailto:[email protected]] On 
Behalf Of TJ Trout via Af
              Sent: Monday, September 22, 2014 6:39 PM
              To: [email protected]
              Subject: [AFMUG] rocket ac lite performance

              �

              How are the rocket AC's performing for you guys? Throughput? Bugs?







              -- 

              All parts should go together without forcing. You must remember 
that the parts you are reassembling were disassembled by you. Therefore, if you 
can't get them together again, there must be a reason. By all means, do not use 
a hammer. -- IBM maintenance manual, 1925











          -- 

          All parts should go together without forcing. You must remember that 
the parts you are reassembling were disassembled by you. Therefore, if you 
can't get them together again, there must be a reason. By all means, do not use 
a hammer. -- IBM maintenance manual, 1925









        -- 

        All parts should go together without forcing. You must remember that 
the parts you are reassembling were disassembled by you. Therefore, if you 
can't get them together again, there must be a reason. By all means, do not use 
a hammer. -- IBM maintenance manual, 1925









Reply via email to