Wireless Ubiquti AP Sgle or Dual Band?

[Lewis]

Hi, I think I'm going to try and deploy a couple of Ubiquti APs to get stronger and further signal than I'm seeing with a wireless router. There is Ethernet from the router to where each of the APs will go. I'd been considering the Wireless Long Range AP, which looks to be only 2.4ghz band - I'm wondering if I should be looking at an AP that is dual band? I know the 2.4 band goes further and better through walls. Thanks - Lewis

 

[stevech]

Hi, I think I'm going to try and deploy a couple of Ubiquti APs to get stronger and further signal than I'm seeing with a wireless router. There is Ethernet from the router to where each of the APs will go. I'd been considering the Wireless Long Range AP, which looks to be only 2.4ghz band - I'm wondering if I should be looking at an AP that is dual band? I know the 2.4 band goes further and better through walls. Thanks - Lewis

Long range AP is a bit of a misnomer. The AP can run a million watts but it won't help the inbound from client device signal. WiFi is two-way.

I would not spend the premium for dual band AP. You can probably get any dual band WiFi router (cheap) and config. it as an AP.

 

[abailey]

I agree with stevech that the Unifi AP-LR is not really any better than the standard Unify AP 2.4Ghz. I have the AP-LR in my house and I love it but if I had it to do again I would just purchase the Unify AP 2.4ghz. Anyway you need to decide if you want a pro level AP or consumer level AP. If you have chosen pro level then I would say go for the Unify AP-Pro 2.4/5ghz AP. I will say they are rock solid. I put mine in about 10 months ago and have not had to reboot it yet. It just works.

 

[crazyea]

The question isnt that simple.

Are you deploying in a home? OR a wharehouse type setting where walls wont interfere?

In a home a unifi ap-lr is not going to make any difference over the regular version. I have both and there is really zero difference for me. Also the LR versions are known to click when data is being transmitted.

Also, having more than one AP in a home, for me anyway, doesnt always get the most desirable results. On many occasions my devices will roam to the further away access point. This is especially true if you decide to use Zero Handoff.

Now bear in mind most of the devices in my home are hand held (ie phones, tablets, etc.) or gaming systems (xbox, ps3, xb1, etc) so I never really get any better than -40db on any of them. I'm not sure if this is based on being a
good neighbour signal or not.

If you have dual band laptops, etc then the higher priced ap's could make sense for you. For me personally I'm happy getting 30-40mbps on anything other than my pc, which is hard wired.

Now if you are in a large wherehouse with fewer walls, then the LR version could make a difference.

 

[azazel1024]

The question isnt that simple.

Are you deploying in a home? OR a wharehouse type setting where walls wont interfere?

In a home a unifi ap-lr is not going to make any difference over the regular version. I have both and there is really zero difference for me. Also the LR versions are known to click when data is being transmitted.

Also, having more than one AP in a home, for me anyway, doesnt always get the most desirable results. On many occasions my devices will roam to the further away access point. This is especially true if you decide to use Zero Handoff.

Now bear in mind most of the devices in my home are hand held (ie phones, tablets, etc.) or gaming systems (xbox, ps3, xb1, etc) so I never really get any better than -40db on any of them. I'm not sure if this is based on being a
good neighbour signal or not.

If you have dual band laptops, etc then the higher priced ap's could make sense for you. For me personally I'm happy getting 30-40mbps on anything other than my pc, which is hard wired.

Now if you are in a large wherehouse with fewer walls, then the LR version could make a difference.

That is very odd. I've never noticed that behavior with my 3 APs. My devices always associate with the closest (in terms of signal) Access Point.

For any kind of high power or long ranged AP/Router, that is generally true. About the only time you'll see a real difference is if you are bridging two together, or sometimes in edge cases when you are near the limits of the signal strength you'll tend to get somewhat higher receive bandwidth on the client because the signal is stronger, but the transmit bandwidth will be just as crappy.

Other downside there is that because of the higher receive signal strength, it can throw off a client in roaming. The client (if it is client initiated roaming, which is most, except for the zero handoff stuff like Ubiquity does) is ONLY looking at receive signal strength. So their transmit ability to the router/AP can be total crap because of signal degredation, but the router is still pumping out a nice strong signal because of the much more powerful amps/radios in it...so the client won't switch to a nearer and better AP because it thinks its connection is good enough.

Its highly driver dependent, but most devices will look to roam in the -60 to -70dB range and generally require another router to be at least +10dB up to around +20dB stronger than the current one before they'll associate with the new access point.

If you have a LR/High Power router, that can mean 3-6dB higher signal strength from the router...but no corresponding increase in transmit signal strength on the clients end. That means potentially you are sticking with the current AP for 3-6dB "longer" even if you would have otherwise been close enough to another AP to have associated with that one...which would have given you better Rx and Tx.

Anyway, high powered stuff is really only desirable if it is going to be the ONLY unit, or if you are going to set up a pair of them in bridge mode (and then you might as well get a couple of dedicated bridges). If you need higher signal strength you are 99 times out of 100 better off with higher gain antennas as that'll benefit Rx and Tx equally.

 

[crazyea]

In my home my access points are both around -30db to -50db, on average -40db.

This is a typical Event log. It might be different in another home, but it seems to me I probably dont need two ap's in mine, but I do it because these are limited to about 35mbps each and I have a lot of devices on my 100mbps line and I'm hoping for the devices to use the AP closer to them. I have zero handoff enabled as well so both AP's are on channel 11.

2014/06/16 10:06:08 User[xxxx-nexus5] roams from AP[uap2] to AP[uap1] on "channel 11(ng)"
2014/06/16 10:05:24 User[xxxx-nexus5] roams from AP[uap1] to AP[uap2] on "channel 11(ng)"
2014/06/16 10:04:28 User[xxxx-nexus5] roams from AP[uap2] to AP[uap1] on "channel 11(ng)"
2014/06/16 10:03:31 User[xxxx-nexus5] roams from AP[uap1] to AP[uap2] on "channel 11(ng)"
2014/06/16 10:02:09 User[xxxx-nexus5] roams from AP[uap2] to AP[uap1] on "channel 11(ng)"
2014/06/16 10:01:24 User[xxxx-nexus5] roams from AP[uap1] to AP[uap2] on "channel 11(ng)"
2014/06/16 09:59:25 User[xxxx-nexus5] roams from AP[uap2] to AP[uap1] on "channel 11(ng)"
2014/06/16 09:58:40 User[xxxx-nexus5] roams from AP[uap1] to AP[uap2] on "channel 11(ng)"

 

[azazel1024]

If they are both limited to 35mbps...having them on the same channel means sharing the same bandwidth. Makes no sense to have two APs, competing for the same bandwidth. Either use only a single one, or else set them on different channels. Otherwise you don't have 35+35, you have 35mbps that is shared.

Its odd that things are transfering, but I guess it depends on how you have zero hand-off setup. Normal client initiated doesn't happen to generally at least -60dB, which is normally the edge from very good to good signal strength (and often times were you'll start seeing a real drop in transfer speeds...depending on background noise).

Just addressing my two indoor APs, signal strength on them runs from -25dBm to -80dBm inside the house depending on where you stand. As for association, the worst connection is maybe around -65dBm with average more in the -50dBm range or better. There is a pretty clear break point between them.

That is 2.4GHz of course. 5GHz its more like -35dBm down to off the charts low (<-92dBm according to InSSIDer) with average around -65dBm.

 

[crazyea]

Are you saying that if one device is pulling 35mbps on one AP, then another device would not be able to do the same on another AP at the same time?

 

[crazyea]

Also, Even if I sit right next to the AP (within a foot) the signal averages -40db. I have no idea how people get -25.

 

[abailey]

Are you saying that if one device is pulling 35mbps on one AP, then another device would not be able to do the same on another AP at the same time?

Not if the second AP is on the same channel. Only one radio at a time can talk on the channel. This is true whether it is a second AP in your house or an AP in a neighbor’s house that is on the same channel (within range). If you set your second radio to the same SSID and passcode, with a different channel then you should be able to roam between them fine and you will not be cutting your bandwidth by using the same channel.

 

[azazel1024]

Not if the second AP is on the same channel. Only one radio at a time can talk on the channel. This is true whether it is a second AP in your house or an AP in a neighbor’s house that is on the same channel (within range). If you set your second radio to the same SSID and passcode, with a different channel then you should be able to roam between them fine and you will not be cutting your bandwidth by using the same channel.

Correct.

Same channel, shared medium. Different channel, not shared medium.

You also have the additional issue if both are talking at the same time, you are going to have contention. So actually, you are likely to run in to a situation with two APs where you might of have 35Mbps with one AP and multiple clients connected to it, where you'll actually have reduce bandwidth with two APs and clients connected to both and trying to receive/transmit at the same time.

Because of contention. If the APs are physically seperated enough that the transmissions can't step on each other, no issues. For example in my setup (just talking about 2.4GHz), I have 3 APs, with AP 1 and 3 sharing the same channel (11). However, they and their clients are always going to be physically highly seperated. The closest overlap has a signal difference over >>>30dBm. In testing, file transfers on both APs at the same time from different clients have zero impact on each other (when the clients were connected to the appropriate APs).

My AP2 in on Channels 1+5, because its coverage DOES physically overlap with APs 1 and 3. If they shared channels there, there would be/is contention. Being on seperate channels though, there is no overlap or contention with the other APs or clients on those other APs.

It also means that instead of having to deal with hidden node issues and contention and all of that kind of stuff reducing bandwidth, I have a nice clean low interference setup with the benefit that I have 600Mbps of through put on AP2 (300Mbps 2.4GHz, 300Mbps 5GHz), 130Mbps on AP3 (2.4GHz only and 20MHz...outdoor AP) and 430Mbps on AP1 (130Mbps 2.4GHz 20MHz only, 300Mbps 5GHz) that I can potentially utilize if I had clients dispersed around my house/property, or 1160Mbps.

If I had them all on the same channel I'd have at best 600Mbps, minus contention issues.

 

[crazyea]

You lost me. There are at least 20 SSID's on channel 11 and the same on 1 and 6. If this were the case no one around here would have any bandwidth at all.

 

[azazel1024]

Oh as to why your RSSI is so low. Dunno, what are you using to determine the signal and what are the APs?

My old Netgear 3500L with internal antennas on 2.4GHz would hit around -30dBm sitting 5ft away on the arm of my couch on my laptop. With my TP-Link WDR3600 it is -25dBm on 2.4GHz and around -35dBm on 5GHz at best in the same location. Dropping off quickly on both of course.

On my son's android tablet it measures about 5dBm lower. My windows tablet amazingly enough actually records about 2dBm higher on average than my laptop at a bit of distance (same -25dBm max)...but it could be because the antennas in my laptop are in the bezel of the lid...which is aluminum (plastic bezels, aluminum backing) and my tablet (asus t100) is all plastic.

 

[crazyea]

Also, then what is the point of zero handoff if it forces both AP's to be on the same channel?

 

[azazel1024]

You lost me. There are at least 20 SSID's on channel 11 and the same on 1 and 6. If this were the case no one around here would have any bandwidth at all.

Are they all transmitting at maximum all the time and are they all within just a few dBm of your own network? Signal strength plays a big roll with contention. If you are getting within a few dBm of your network, say 20dBm, you are going to have issues where parts of your stuff is going to get lost because the AP can't hear it.

As for bandwidth, you stated you have 35Mbps. Unless you are doing 11g...you have no bandwidth as far as I am concerned. On 2.4GHz 40MHz (because I have NO networks around me, so I don't worry about being a good neighbor) I get 180Mbps (2:2). On 5GHz I get 200Mbps (same 2:2 40MHz). Even on 20MHz 2.4GHz I can hit 90Mbps with a bit of a tailwind. Heck, my 1:1 tablet on 2.4GHz can hit around 45Mbps or so on 20MHz....standing 80ft from my outdoor access point.

So you ARE seeing a bunch of the contention from the surrounding networks (and I'd highly recommend a 5GHz network).

 

[azazel1024]

Also, then what is the point of zero handoff if it forces both AP's to be on the same channel?

Marketing fluff combined with seamless roaming*?

*Seamless roaming in my experience is generally highly overrated unless you have VOIP...and in that case, maybe it is beneficial. I see completely seameless roaming with NON-zero handoff on my APs running video streaming, video chat (skype and facetime), file transfers, web page loads, streaming podcasts, streaming music pretty much everything I have tried. The most noticable impact is in file transfers where sometimes I can see the file transfer speed plumet for a second or two before picking back up as it transfers between APs. Everything else is completely seamless.

 

[crazyea]

I'm testing on my nexus 5 using speedtest.net. I get 35mbps. This could be a limitation of my phone though. The iPad gets substantially less. I have not tested on a laptop as I do not own one. My xbox one built in test shows about 40mbps.

All of this aside. I have zero issues with multiple xbox playing at the same time, nor do I have any issues streaming 1080p youtube videos while streaming netflix on at least 5 devices at one time either.

 

[azazel1024]

Okay. That's great that it is working really well for you. I never said it couldn't be working well. If you need more wireless bandwidth, which it doesn't sound like you do or care about needing then your setup is perfect.

The only time you need more than what you have is if what you have isn't sufficient (or you like overkill). I have a tendancy to need more bandwidth. It isn't constantly, but I am shuffling pretty large files wirelessly at least a few times a week either to my tablet or laptop. When its a 1-5GB file(s), every drop of wireless bandwidth is nice.

If it was chugging along at 35Mbps I'd just get out a network cable or my USB3 to GbE adapter and plug in my laptop/tablet for something like a 5GB file. At 80Mbps its bareable on my tablet and at 200Mbps on my laptop its not bad at all. I don't want to be giving up any of that just because my wife is pulling some pictures down from the server to her iPad on the other side of the house, or I am downloading some windows updates on my tablet sitting in my office and I am in my living room trying to copy a game install to my laptop.