Linksys Announces Velop Mesh Wi-Fi System

[Pericynthion]

Awesome - thanks for the info!

p.s. If anyone is interested, they have published a semi-official method to run the Velop in bridge mode

http://community.linksys.com/t5/Vel...a-semi-unofficial-official-guide/td-p/1110302

 

[NPMan]

Couple of thoughts after installing a 2 unit Velop setup. Wasn't as quick as I had hoped, but part of that was my fault. First, if you're planning to use ethernet on node 2 as the backhaul, it has to be wired directly to node 1. Ie. not through a router or hub. I'm not a network pro, so this was not obvious to me. (Even though it was mentioned how Velop NATs everything.)
On top of that, when installing my Devialet Phantom Gold speakers through the Dialog box, they didn't work. Again, NAT stuff. (The Dialog could not sense the speakers.) After talking to support (which were very helpful), I moved the ethernet cable to the Dialog from a hub to Node 1. Then it worked. Of course, now I can't use ethernet for the backhaul. Great.
Luckily, the nodes see each other and the wifi backhaul seems to work great. I now have great wifi coverage and speed in my whole house. After years of trusted service, I've turned off my Apple Airport Extreme.

 

[Pericynthion]

if you're planning to use ethernet on node 2 as the backhaul, it has to be wired directly to node 1. Ie. not through a router or hub. L, I've turned off my Apple Airport Extreme.

Not sure why this would matter - also it's impossible in a 3 node scenario as you would need 2 uplink ports and apart from the internet connection you only have 1 spare on the master node. So it has to go through some form of switch/hub. (Mine 2 remote nodes are, and they work fine)



Sent from my iPhone using Tapatalk

 

[NPMan]

Interesting. You could imagine node 1 to node 2, then node 2 to node 3. That is what support told me, but this gives me hope, I'll try again. Thanks.

 

[thiggins]

First, if you're planning to use ethernet on node 2 as the backhaul, it has to be wired directly to node 1. Ie. not through a router or hub. I

I don't think this is true.

Linksys has advised that setup must first be done with all nodes connected via Wi-Fi. After setup is complete, you can move any node to Ethernet. They said they are working on incorporating Ethernet onnected nodes into the setup wizard.

 

[thiggins]

You know how sometimes you want to ask a question, but don't want to look like and idiot? - hopefully Tim and y'all are feeling benevolent ;-)

I often look like an idiot, so you're in good company...

To accurately answer your question, I'd have to know how inSSIDer determines the PHY mode, or do a packet trace to see what the 2.4 GHz radio is advertising in its header.

As SFX noted, 11ac routers are now usually built with the same radio SoC used for both bands. In the Velop's case, the 2.4 and low-band (I think) 5 GHz radio are in the IPQ4019 SoC.

Since Linksys specs Velop as having 400 Mbps max link rate in 2.4 GHz, it means 256-QAM is available and is probably advertised in the 2.4 GHz header. This is not really kosher (802.11 compliant) since that modulation scheme isn't part of 802.11n, which is what the 2.4 GHz radio in an "AC" router really is / should be.

But ever since Broadcom introduced 256 QAM in 2.4 GHz to give products using its chipset a bigger number on the box, other chipmakers followed suit. Specs have never stopped the consumer Wi-Fi industry from shipping product!

 

[pete y testing]

Specs have never stopped the consumer Wi-Fi industry from shipping product!

yep , never let truth get in the way of a good sales pitch

btw does anyone with the velop kit have any issues with heat if they have been running the for a while

 

[michael814]

First, if you're planning to use ethernet on node 2 as the backhaul, it has to be wired directly to node 1. Ie. not through a router or hub. I'm not a network pro, so this was not obvious to me. (Even though it was mentioned how Velop NATs everything.)

Setting aside possible issues with double-NAT if you're using a router as your ethernet switch (which sounds like may be the case), I can say with absolute certainty that ethernet backhaul works perfectly through a regular ethernet switch. In fact, my setup has TWO switches between the main unit and the hardwired remote node. One port of the main Velop is taken up by the cable modem of course, while the other one goes to a 16-port switching in the closet. One of the ports on that switch goes to a small switch in another room, to which the remote Velop is connected.

 

[gowg]

Setting aside possible issues with double-NAT if you're using a router as your ethernet switch (which sounds like may be the case), I can say with absolute certainty that ethernet backhaul works perfectly through a regular ethernet switch. In fact, my setup has TWO switches between the main unit and the hardwired remote node. One port of the main Velop is taken up by the cable modem of course, while the other one goes to a 16-port switching in the closet. One of the ports on that switch goes to a small switch in another room, to which the remote Velop is connected.

Have you tested speeds with wired backhaul vs wireless?

 

[michael814]

Have you tested speeds with wired backhaul vs wireless?

Yes, not scientifically though. On a wireless node with a pretty good signal (about 25ft away, but down a floor and through a few walls), I get about 350mb/s when connected through that node. Speed through a hardwired node is virtually identical to that of main node (about 500mb/s).

 

[Chairman007]

A bit strange as this does not mirror Tim's results ... obviously a different selection method.

http://www.prnewswire.com/news-rele...mesh-networking-wi-fi-solution-300410936.html

 

[michael814]

A bit strange as this does not mirror Tim's results ... obviously a different selection method.

http://www.prnewswire.com/news-rele...mesh-networking-wi-fi-solution-300410936.html

Yeah, I mentioned in a post in this thread a few weeks ago that my Velop system is delivering MUCH faster speeds than Tim's test indicated, despite a similar arrangement of nodes. The Linksys people said they were trying to work with him to figure out the issue, as other reviews also noted much higher speeds.

 

[pete y testing]

The Linksys people said they were trying to work with him to figure out the issue, as other reviews also noted much higher speeds.

read the test procedure and hardware used , i believe tim's testing is done on 2 x 2 adapter

 

[michael814]

read the test procedure and hardware used , i believe tim's testing is done on 2 x 2 adapter

Sorry, I should have been more specific... I was referring to the extremely poor speeds seen when connected to remote nodes in his real world "open air" test. Even with a laptop hardwired to a remote node (so strictly testing backhaul speed), he only saw around 20mb/s. The distance and wall/floor obstructions between his main node and "Living Room" remote node are very similar to mine, and in the same test configuration (laptop hardwired to the remote node), I'm seeing around 400mb/s.

Other reviews have noted comparable speeds in their "real world" test environments, so for some reason Tim's test is the outlier. It sounded like Linksys support initially brushed him off when he inquired about it, as they just told him he had to move the nodes closer. But after the review was posted it caught the attention of a higher-up Linksys guy, who said he wanted to work with Tim to figure out what the issue was (defective hardware perhaps?). So, hopefully Tim will revisit this.

 

[thiggins]

I've had multiple discussions with Linksys about this. The logical explanation for the low throughput was that the backhaul between root and second "Living room" node was 2.4 GHz, not 5. Most likely with 20 MHz bandwidth.

I made a brief attempt to repeat the measurement but ran into trouble getting all the nodes reset. This apparently requires uninstalling the app and doing a factory reset on nodes. Haven't had the time to get back to it.

All DWS products will be retested after I get a more robust test process in place in the next month or so. Linksys, along with everyone else, will get their chance then.

 

[pete y testing]

The logical explanation for the low throughput was that the backhaul between root and second "Living room" node was 2.4 GHz, not 5. Most likely with 20 MHz bandwidth.

im not sure why they would set the velop to have the ability to swap its backhaul from 5 gig to 2.4 gig certainly makes the setup more confusing esp placement wise

All DWS products will be retested after I get a more robust test process in place in the next month or so. Linksys, along with everyone else, will get their chance then.

look forward to reading the results and the debate that will follow

 

[thiggins]

Mesh systems should use all available backhaul channels as necessary to balance bandwidth use.

 

[pete y testing]

Mesh systems should use all available backhaul channels as necessary to balance bandwidth use.

but in your testing case would it not have been a decision made on distance from the other node that set it to 2.4 gig rather than any bandwidth usage

 

[thiggins]

but in your testing case would it not have been a decision made on distance from the other node that set it to 2.4 gig rather than any bandwidth usage

I don't know what the decision basis was. The second node location has a good signal for both 2.4 and 5 GHz. The third node is out of 5 GHz range and barely in 2.4 GHz range from the root node to ensure it connects to the second node.

 

[pege63]

Can these be used like this?