What Ubiquiti Wireless Products Should We Test?

Avatar
Ethan Ace
Sep 12, 2015

We're planning to test some Ubiquiti gear, since our last test was about 5 years ago now.

We're planning to test the NanoBeam and NanoStation for sure, as they seem to be the most popular of their new models. We'd like to include something with an external antenna as well, so possibly the Rocket with an omni antenna.

Your thoughts? Anything you use more than these options? Anything to watch for with these models?

(2)
Avatar
Robert Baxter
Sep 12, 2015

I look forward to seeing the result on these products. We use the 5ghz models.

Looking at the last report you did, can you include some analysis of delivered and missed packets as a measure of data reliability as a function of an environment where there is variation of other wireless communication? This is the primary reason we had to move from 2.5Ghz.

MM
Michael Miller
Sep 12, 2015

What will be the testing procedures?

Avatar
Dale Bullough
Sep 12, 2015
IPVMU Certified
Would you please test various wavelengths (ghz?...may not know right vocab there but think yall know) and different scenarios with/withoit line of site and through different obstacles...trees to reinforced concrete etc... Thanks
RM
Ryan McGovern
Sep 13, 2015
We use a combination of the Nano Station and Rocket with antenna 120 degrees and omni. All 5Ghz. Can the Nanobeam be used to replace part of this network? Is the Nanobeam a replcement for the Nano Station? How are they different? We have found Ubiquiti to be rock solid over the past 4 years. We run CCTV and access control IP traffic on them. Easy to set up & no warranty call outs.
Avatar
Luis Carmona
Sep 13, 2015
Geutebruck USA • IPVMU Certified

It'd be interesting to know what sustained throughput they could offer, or how many streams they would seem to maintain. We don't use them anymore, but we once replaced a bridge pair (earlier models that seem to work a lot better), we were having issues with. It was a pair of 5Ghz nanonbeams with only about 150 or so between buildings. The data throughput was only maybe around 30mb/s to 40mb/s, but it had around 25 to 30 camera streams going through it. We switched to a wired connection and problems went away. I always wondered if it was the number of stream sessions it was having trouble with. I commented on it before back in the old LinkedIn forum I think.

MM
Michael Miller
Sep 13, 2015

Delete

AS
Andrew Somerville
Sep 13, 2015

The Nanobeam M5-19 and M5-16 are nominally replacements for the older Nanostation and Nanostation Loco (faster processor and tighter antenna pattern). Given they are at the same pricepoints, I see no obvious reason to keep deploying Nanostations.

So I would say the following would be the priority for testing:

* Nanobeam M5-16 point to point (lowest cost solutionfor shorter links)

* Nanobeam AC point to point (newer ac chipset and gigabit Ethernet interface)

* Rocket M5 with sector antenna plus multiple Nanobeam M5-16 / M5-19 to test point to multipoint scenario

* AirFiber5 point to point (high capacity over longer distances)

(1)
JH
John Honovich
Sep 13, 2015
IPVM

Andrew, thanks. Question: Is the Nanobeam 5Ghz only? I ask because Mark is mentioning testing a 900Mhz product which made me think about doing a 900Mhz vs 2.4Ghz vs 5Ghz comparison as part of the test.

We are certainly going to test multiple lines so it's not one or the other. I am just trying to get a sense of how to match them up.

AS
Andrew Somerville
Sep 13, 2015

Hi, the nanobeam /nanobeam ac are 5GHz only. Over here in Europe that's pretty well all we use since 900MHz is a cellular band and 2.4GHz is power restricted to 100mW.

900 MHz should have better performance in non line of sight conditions due to the lower operating frequency and higher power, so certainly worth a comparative test for your North American audience.

My view is that 2.4 GHz is generally to be avoided because of the high probability of noise / interference from other wireless sources (Wi-Fi, bluetooth, microwave ovens etc.)

(1)
JH
John Honovich
Sep 13, 2015
IPVM

"My view is that 2.4 GHz is generally to be avoided because of the high probability of noise / interference from other wireless sources (Wi-Fi, bluetooth, microwave ovens etc.)"

That's reasonable. From a testing perspective, it would be good to try the different frequencies in the same (urban / congested) area to see how much of a difference it makes across them, so we would obtain some numbers / quantification on it.

(1)
MM
Michael Miller
Sep 13, 2015

Designing a relible wireless solutions is all about selecting the least used frequncy and while using the correct antennas to midagate any interfrence. Not sure testing different wireless frequencies in the wrong environments isn't going to prove anything but the obvious.

I think testing streams verse capacity would be a way more intresting test. Example: 10 X 10Mbps streams vs 50 X 2Mbps streams over a PTP to figure out where the link fails. Then do the same test with Fluidmesh/Proxim and see if they have the same results. Comparing UBNT vs Fluidmesh/Proxim IMO is way more intresting.

(2)
(2)
LT
Larry Tracy
Sep 13, 2015

Have had aot of interference on the 900Mhz versions, 5.8 works well in major cities. Don't use 900Mhz any longer.

2.4 Ghz forget it, too many microwave ovens and consumer goods.

UI
Undisclosed Integrator #1
Sep 16, 2015

I like the idea of a comparison. It would be interesting to see performance between them when going through a little foliage or wall.

(1)
BL
Brian Leander
Sep 14, 2015
IPVMU Certified

Hi Andrew,

Good points, however on the NanoStation vs Nanobeam, I would love to see a second Ethernet port on a Nanobeam, this is the reason we still use the Nanostation in many applications. Since we do a lot of solar, adding a second camera to a pole or creating a repeater link with a NS means no need to add a switch, which saves considerably over the power budget. Once we go more than 2 cameras we need a switch and go to the Nanobeam regardles.

Otherwise, great recommendations, looking forward to seeing AirFiber5 results. We have not deployed those yet.

John, a repaeter link is something we have to do at times as well (to avoid physical obstacles). It would be good to quantify the performace loss on the repater links vs. a direct PtP/PtMP link.

Avatar
Mark Jones
Sep 13, 2015

I may be the only one, but I would like to see a 900Mhrtz product, and I would like to see some questionalble line of sight tests. I don't mean block or a relay affair, just something with some trees, but still you have some line of sight. Real test in real world applications.

GH
Greg Haynes
Jan 15, 2016

I have used Ubiquiti 900mhz for years. They will do an awesome job thru trees. I usually deploy them in rural areas and have no interference. Very reliable. Recently set up 1/4mile link thru trees in winter, even with no leaves could not see from end to end because of trees and got 40mbs thru link

(1)
Avatar
Luis Carmona
Jan 15, 2016
Geutebruck USA • IPVMU Certified

That's great info Greg, thanks. I always wondered about how effective 900mhz would be for "light" obstacles like foliage.

Avatar
Rob Hammond
Sep 14, 2015
IPVMU Certified

John,

I look foward to a refresh on wireless. In my market I have been hearing a lot about Comnet's Netwave. I would appreciate an your unbiased view of their product.

JH
John Honovich
Sep 14, 2015
IPVM

Rob, thanks. Anything, in particular, you are hearing about Comnet's wireless? Alternatively put, any major differentiators claimed?

Avatar
Rob Hammond
Sep 14, 2015
IPVMU Certified

Most of what I am hearing about Comnet and Netwave is coming from the mouths of sales people. Their product line seems to be fairly broad and targeted towards surveillance.

JH
John Honovich
Sep 14, 2015
IPVM

Are they saying anything in particular outside of that its wonderful and you should buy it? I am trying to understand what to test or consider.

Avatar
Rob Hammond
Sep 14, 2015
IPVMU Certified

It has been a while since i talked to them, I will reach out to see if they have any difreaciators.

CW
Craig Wilson
Sep 14, 2015

Ethan,

I suggest you use the Bullets, as they are dirt cheap, tied to a small panel antenna. We use that combination all the time and i would be interested to see how they stack up. Just remember to get the 90 degree N type adapter to mount to your antenna in the right direction.

AS
Andrew Somerville
Sep 14, 2015

A Nanostation Loco is cheaper than a Bullet, has better throughput and you don't need an antenna or to worry about weatherproofing connectors. So I don't see the point myself.

MM
Michael Miller
Sep 14, 2015

Not to mention if you use a bullet in a PTMP network with rockets or nanostations you will slow the whole network down to the speed of the bullets.

(1)
Avatar
Luis Carmona
Sep 14, 2015
Geutebruck USA • IPVMU Certified

Good point, Mike. Not many people realize the slowest, worst connection determines the speed for everybody (other nodes) else. One radio, one job.

(1)
UI
Undisclosed Integrator #1
Sep 15, 2015

Almost missed this one. Kind of a sucker for training so:

- Completed the Airmax training last week. Lot of good info on how to speed up/strengthen weak links (I am a novice to this equipment and had been using it just to connect buildings without knowing application of some of the advanced settings). Met a couple people that do larger WISPs so they had some good insight as well.

- Have the Unifi and Edgemax training (one next week and the other 2 weeks later)

Different products from Ubiquiti I'm interested in:

1. While I know this is one of their most expensive products, I'd like to see the airfiber in action if you get the chance. It would be a nice option for customers that want to connect buildings, but don't want to bore for conduit.

2. I would also be interested on your take of their Edgemax routers and EdgeSwitches. I have one of each set up in my office to play with coupled with an older AP. So far so good and I've created a few VLANs to segregate, but I don't have a lot of experience here so if you do, I'd like to see your opinion on how they compare to UBNT's competitors.

3. UBNT just released their new AC APs. Have an order in for a few pro versions, but they're on back order right now. The Ubiquiti forum is packed with complaints on the old unit so it would be nice to compare how the new ones stack up.

4. MPort(-S). It would be nice to see if you could find a great application for this device since you can code it how you like, but most of what I've found is home use. There's got to be something there I can use, but I haven't thought about it a lot either.

Anyways, excited to see your testing. Thanks.

Avatar
Ethan Ace
Sep 15, 2015

We actually have an EdgeSwitch. It's the main switch in our test lab. We haven't released a test on it because we're not really sure how interesting it would be, and some of the higher end functions are more advanced than most people here will understand/not often used in surveillance.

The one thing I will say is it is very quiet. Compared to the D-Link PoE switch we had, it's maybe half as loud.

As far as MPort, we haven't really known what to do with that, either. It's a weird little automation system, and I think it was intended to be sort of an APC NetBotz competitor. Just as an application off the top of my head: The controllable power strips are fairly inexpensive and would be good for cycling power to multiple devices in a rack. I'd prefer they be a 19" or rear rack rail for factor, but for $99 vs. others that typically cost hundreds, you have to give on something.

(1)
UI
Undisclosed Integrator #1
Sep 16, 2015

"We haven't released a test on it because we're not really sure how interesting it would be"

This is the point where I realized how truly boring I am.

Avatar
Ethan Ace
Sep 17, 2015

Equipment started arriving today. Here's what we've got so far:

900 MHz vs. 2.4 GHz vs. 5 GHz

Going to test a pair of NanoStations in these different frequencies to see what tradeoffs there are. Line of sight, non line of sight, throughput, etc., etc.

NanoBeam M vs. ac

The NanoBeam seems to be the current go-to for simple PtP links, plus we can see the difference between the old radios and this new 802.11ac based model.

PtMP Test

Going to test the Rocket ac with a 45° sector antenna as the base station, and use the NanoBeams and/or NanoStations as end points.

UI
Undisclosed Integrator #1
Sep 17, 2015

Great looking forward to the test. If possible, could you look at the difference with one of the Locos behind glass (Low-E or tinted)? They sell suction cup mounts, but it's stated doing so can degrade the signal so it would be interesting to see by how much. Just a thought.

AS
Andrew Somerville
Sep 18, 2015

Hi you should note that currently that ac Access Points (such as Rocket ac) are currently not compatible with the older non-AC AirMax products (such as Nanostation / Nanobeam M5). You can only use the Nanobeam ac's with Rocket ac's

(2)
(1)
Avatar
Luis Carmona
Sep 18, 2015
Geutebruck USA • IPVMU Certified

"you should note that currently that ac Access Points (such as Rocket ac) are currently not compatible with the older non-AC AirMax products "

Do you mean not compatible when using Ubiquiti's proprietary Airmax, or not compatible period even trying to connect using what are supposed to be generic wireless protocol standards..?

AS
Andrew Somerville
Sep 18, 2015

Not compatible using the two flavours of Airmax.

If you were to switch off Airmax and run standard 802.11 I guess they would work together but then you would create a new world of pain due to distance limitations in standard 802.11, hidden node issues etc.

Avatar
Ethan Ace
Sep 18, 2015

I'm setting things up today since most of the radios are here. Here's a short video showing the NanoBeams and NanoStations, and the physical form factors. We'll definitely go over each in more detail later.

MM
Michael Miller
Sep 18, 2015

I would be intrested to see if you get the POE converters to work with the AC Nanobeams. Never been anble to get them to work.

Avatar
Ethan Ace
Sep 22, 2015

They're working fine here. Not sure why they wouldn't work. Do they simply not power up or do they fail after some time?

I just pinged the far side of a link for half an hour and didn't drop a single ping, so the adapters for both radios were obviously stable that whole time:

Ping statistics for 172.20.128.27:
Packets: Sent = 2148, Received = 2148, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 0ms, Maximum = 6ms, Average = 0ms

U
Undisclosed #2
Oct 22, 2015
IPVMU Certified

I have used those several of those POE converters, both the indoor and outdoor models without issue, so I was thinking the same thing as Ethan: Why wouldn't they work, you just plug them in.

Today I received 4 of the outdoor ones, 3 of 4 were totally DOA. Maybe there is a bad batch or two out there?

Avatar
Cosimo Malesci
Oct 07, 2015

John, Ethan,

would you guys be interested in testing some Fluidmesh gear as well to compare? Not sure what the test protocol would look like, but we'll be glad to send you some gear to play with. I see this being particularly valuable if you do this comparison outdoor in a real install scenario.

Let me know!

Thank you,

Cosimo Malesci

Co-Founder/Fluidmesh Networks

(1)
JH
John Honovich
Oct 07, 2015
IPVM

Cosimo, one we'd buy it if we tested it.

Two, I am not opposed to testing Fluidmesh, it's just a secondary priority to Ubiquiti given Ubiquiti's broad use. After Ubiquitit, Fluidmesh would be the next most logical choice to test for wireless video surveillance. We'll consider it but it's not likely going to be for months, since we just started field testing the Ubiquiti products this week.

Avatar
Cosimo Malesci
Oct 07, 2015

John,

thank you for the quick response. Totally understandable and thank you for considering it. I truly believe there would be value in having you guys do a shoot out of different wireless solutions for video and share results. Although this would take some work and carefull planning, there is really nothing out there like this and I believe people will benefit from it.

Particularly if you do it in an outdoor environment testing PTP and PTMP and find a way to add interference to the links and see how they perform.

No pressure and no rush. Just a thought.

Thanks,

Cosimo

(1)
UI
Undisclosed Integrator #1
Jan 14, 2016

Just wondering if I missed the post for this test or if it is still ongoing. Interested in what you guys come up with. Thank you

(1)
JH
John Honovich
Jan 14, 2016
IPVM

We did some prelim testing, then we it to the side because of camera testing and then the holidays. We plan to restart this next month.

UI
Undisclosed Integrator #1
Jan 15, 2016

Great thank you for the update.

New discussion

Ask questions and get answers to your physical security questions from IPVM team members and fellow subscribers.

Newest discussions