Spiffy! I always wondered about that...

On Tue, Nov 22, 2016 at 5:10 PM, Dan Sullivan <
[email protected]> wrote:

> Hello,
>
>
>
> ePMP provides Air Fairness.  Each SM is allocated an amount of time.  If
> an SM is performing at a poor MCS, then its throughput suffers because
> additional time is not allocated.  The other SMs get their time allocated
> time and do not have their time taken away at the expense of a poorly
> performing SM.
>
>
>
> Dan Sullivan
>
> ePMP
>
>
>
> *From:* Af [mailto:[email protected]] *On Behalf Of *Adam Moffett
> *Sent:* Tuesday, November 22, 2016 9:09 AM
> *To:* [email protected]
> *Subject:* Re: [AFMUG] ePMP 5ghz issues
>
>
>
> If someone did make that claim then it was probably a
> misunderstanding....The effect is a function of the medium and no product
> can avoid it.
>
> It's possible they reduce the effect by allocating less airtime to the
> weak CPE.  Basically make the slow one slower rather than making everybody
> slower.  *IF* they do, that's not a unique feature to the ePMP.
>
>
>
>
>
> ------ Original Message ------
>
> From: "Kurt Fankhauser" <[email protected]>
>
> To: [email protected]
>
> Sent: 11/21/2016 10:33:42 PM
>
> Subject: Re: [AFMUG] ePMP 5ghz issues
>
>
>
> I dont know how they can make that claim when their "flagship" 450 product
> will suffer from decreased capacity when SM's with weak signals connect to
> it and pass traffic. Same thing with LTE. Burns up more airtime so why
> would EPMP be better than all the expensive gear? Doesnt make sense.
>
> Sent from my iPhone
>
>
>
> Kurt Fankhauser
>
> Wavelinc Communications
>
> P.O. Box 126
>
> Bucyrus, OH 44820
>
> http://www.wavelinc.com
> <https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.wavelinc.com%2F&data=01%7C01%7Cdaniel.sullivan%40cambiumnetworks.com%7C92e61d550557469517a808d412e987a2%7C0e263e36340946228ac818d993e76eb6%7C1&sdata=zWbtXstzzu5o039f3XRx1Wu%2F24g4TSZG8SuU42cnWjQ%3D&reserved=0>
>
> tel. 419-562-6405
>
> fax. 419-617-0110
>
>
> On Nov 21, 2016, at 9:48 PM, CBB - Jay Fuller <[email protected]>
> wrote:
>
>
>
> It is my understanding a poor receive on one SM will not hurt performance
> of the other SMs...
>
> ( i could be wrong but am pretty sure i read this on the forums )
>
>
>
> ----- Original Message -----
>
> *From:* Timothy Steele <[email protected]>
>
> *To:* [email protected]
>
> *Sent:* Monday, November 21, 2016 8:38 PM
>
> *Subject:* Re: [AFMUG] ePMP 5ghz issues
>
>
>
> Sounds like major interference or you are installing right under the tower
> without giving the SM installs uptilt
>
>
>
> On Mon, Nov 21, 2016, 9:03 PM Josh Corson <[email protected]>
> wrote:
>
> We are having some issues with a epmp 2000 AP.
>
> At first there was an install on a customers private tower to get
> service. I noticed the MCS downlink and uplink were sitting around MCS
> 1 with 80% or so of packets. So it affected all other customers on the
> AP and brought those other customers into 20% MCS 1 levels. Now his SM
> has been raised and signals are great, -60dBm, SNR of 37dB, MCS 13/15.
> The SM is also operating at mostly MCS 14-15 with only 1% in MCS 1.
> I changed the management to be MCS 0 to attempt to help with the
> minimal connection to MCS 1 so it will be allowed to pass traffic at
> that rate.
>
> However, this customer still has issues where the internet
> connectivity is random. Sometimes it just completely drops when you
> put traffic to it, but when it is on, speed tests of 37 Mbps down and
> 12 upload.
>
> All other customers are still on MCS 1 with levels ranging from 7% to
> 20%. I can assume they are having the same trouble as the customer
> listed above.
>
> All software is 3.1
>
> My question is, what is the issue? This SM shows nothing but good
> results, yet connectivity is poor at best, and all other customers are
> still having issues. I even disconnected the SM that caused customers
> to have problems and they were still modulating at the 7-20% MCS 1
> level. Tried changing channels, etc. but still no results.
>
> Any help would be appreciated.
>
> Josh
>
>

Reply via email to