[189333] in North American Network Operators' Group
Re: B5-Lite
daemon@ATHENA.MIT.EDU (Mike Hammett)
Tue May 17 11:45:20 2016
X-Original-To: nanog@nanog.org
Date: Tue, 17 May 2016 10:37:47 -0500 (CDT)
From: Mike Hammett <nanog@ics-il.net>
To: Jared Mauch <jared@puck.nether.net>
In-Reply-To: <59F38A72-9CA6-4A8C-81DF-400439D1E040@puck.nether.net>
Cc: North American Network Operators' Group <nanog@nanog.org>
Errors-To: nanog-bounces@nanog.org
I know it'll result in the air interface coming down on the M series, but v=
erify your noise with the AirView tool. I've grown to not trust the noise f=
loor measurement. 40 MHz at that supposed amount of SNR should be rocking a=
lmost double what you're getting. With the V and H chains that far apart, a=
lignment might be off. What are your CCQ, AMC and AMQ numbers?=20
-----=20
Mike Hammett=20
Intelligent Computing Solutions=20
http://www.ics-il.com=20
Midwest Internet Exchange=20
http://www.midwest-ix.com=20
----- Original Message -----
From: "Jared Mauch" <jared@puck.nether.net>=20
To: "Mike Hammett" <nanog@ics-il.net>=20
Cc: "North American Network Operators' Group" <nanog@nanog.org>=20
Sent: Tuesday, May 17, 2016 10:29:57 AM=20
Subject: Re: B5-Lite=20
I=E2=80=99m seeing -61 (63/67 V/H) with floor at -101 right now with the XW=
PowerBeam 400 w/ 40mhz. The speeds are =E2=80=9COk=E2=80=9D but getting be=
yond 60Mb/s is hard as the CPU maxes in a bridged setup. Doesn=E2=80=99t se=
em to have any issues with the wireless rate during load, so perhaps it=E2=
=80=99s not doing offload to the chipset right? The goal is to improve capa=
city in the interim while some strategic fiber is deployed for some areas. =
A pair of B5s or AF5X would likely work out but would rather spend that on =
fiber.=20
- Jared=20
> On May 17, 2016, at 11:06 AM, Mike Hammett <nanog@ics-il.net> wrote:=20
>=20
> I think there is some information missing on your longer link. Did you st=
ill have appropriate signal? Was there noise?=20
>=20
> I have a B5 link that's about 2 miles that's rocking full data rate and a=
B5c one that's going about 4 miles at full data rate. My 8 mile B5c link i=
s less than full data rate due to interference.=20
>=20
>=20
>=20
>=20
> -----=20
> Mike Hammett=20
> Intelligent Computing Solutions=20
> http://www.ics-il.com=20
>=20
>=20
>=20
> Midwest Internet Exchange=20
> http://www.midwest-ix.com=20
>=20
>=20
> ----- Original Message -----=20
>=20
> From: "Hal Ponton" <hal@buzcom.net>=20
> To: "Matt Hoppes" <mattlists@rivervalleyinternet.net>=20
> Cc: "North American Network Operators' Group" <nanog@nanog.org>=20
> Sent: Saturday, May 14, 2016 7:31:10 AM=20
> Subject: Re: B5-Lite=20
>=20
> We've deployed 2 B5 links into production, the newer firmware seems to ha=
ve fixed the issues we saw in the links when we first tested them.=20
>=20
> We have a very rural customer where two hops are needed around the site. =
We're lucky in that we had two 80MHz channels free. We see around 350Mbps b=
oth ways actual throughput on both links.=20
>=20
> However, these links are short est. 200mtrs when we had tested these on l=
onger links their performance was awful, on a 40MHz channel we saw 20Mbps.=
=20
>=20
> For our longer links that need a bit more throughput than a Rocket M5 we =
either use Licensed radios or the AF5X which works very well.=20
>=20
> Regards,=20
>=20
> Hal Ponton=20
>=20
> Senior Network Engineer=20
>=20
> Buzcom / FibreWiFi=20
>=20
>> On 14 May 2016, at 11:07, Matt Hoppes <mattlists@rivervalleyinternet.net=
> wrote:=20
>>=20
>> Jared - why not go to Ubiquiti AC gear if you need some more speed and s=
omething more modern?=20
>>=20
>>> On May 14, 2016, at 01:43, Eric C. Miller <eric@ericheather.com> wrote:=
=20
>>>=20
>>> B5c is the only product that I've had much success with from Mimosa.=20
>>>=20
>>> The B5Lite is a cheap plastic shell and, and it performs like it too.=
=20
>>>=20
>>> If you have UBNT gear now, Mimosa is a good next step, but I'd strongly=
recommend that you stear away from the lite and go with the B5c. We use th=
em with rocket dishes. You just need the RP-SMA to N cables.=20
>>>=20
>>>=20
>>> Eric Miller, CCNP=20
>>> Network Engineering Consultant=20
>>>=20
>>>=20
>>>=20
>>> -----Original Message-----=20
>>> From: NANOG [mailto:nanog-bounces@nanog.org] On Behalf Of Jared Mauch=
=20
>>> Sent: Friday, May 13, 2016 7:06 PM=20
>>> To: North American Network Operators' Group <nanog@nanog.org>=20
>>> Subject: B5-Lite=20
>>>=20
>>> Anyone deployed this radio in production in the US? I=E2=80=99m curious=
to hear from people who are using it, looking at replacing some UBNT hardw=
are with it on some PTP links, going from the M-series class devices to som=
ething more modern.=20
>>>=20
>>> Thanks,=20
>>>=20
>>> - Jared=20
>=20
