NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Inconsistency with Orbi Speed
Hi I've recently bought the twin Orbi router system and I have mixed feelings about it. I have seen the fastest speed I've ever experienced using it purely by speed testing it on my iPhone 7 i.e. 3...
Should I be disconnecting my satellite and doing some tests with just the router? Haven't tried this ! Hoping the guys at Netgear are reading these posts and a magical firmware update is in the works that solves these issues !!
I would definitely recommend taking steps to isolate whether the problem occurs when connected through just the satellite, just the router, or both, as well as trying different placement... process of elimination (which is what I THOUGHT I had done thoroughly enough, but the red herring of a good client to router speed and a good satellite to router speed caused me to overlook the router's placement as the likely culprit).
Disconnecting the satellite is one step you could take. My process was to prompt a connection to the main or sat by being near the desired unit and turning my WiFi off and on, because if you just walk over to one, if your device still has a usable connection to the other it will not switch, but turning WiFi off and on will result in it then connecting to nearby unit. Run a WiFi scan using Apple's Airport Utility (you need to enable this feature in the settings in order to see it), and determine the MAC address for each of the 2.4/5ghz bands of the satellite and the router, so that you can confirm which one you are actually connected to. I used an app called WiFi Sweet Spots to see the MAC address of the radio I was connected to (this app also supposedly measures link speed, but I found it to be quite inaccurate most of the time).
You may have a situation similar to mine, where your internet terminates at a specific location, restricting the placement of the router, but as I mentioned above, temporary power and data cables will allow you to experiment with moving it around just to see if that's a factor.
One other thing I discovered that could have played a role with my problem, behind a wall of this closet (about 1.5ft. away from the router) there's a flexduct run. As with the wire shelving, it's not in the path of the signal (at least not in the direction of the router), but presumaby the metallic outer layer of the flexduct behaves as a large piece of metal, and reflects signals around. Plus there's also the metal cases of an ethernet switch and a sercurity camera NVR. Again, these metal objects did not cause noticeable problems with a connection to a router in this location (the Orbi, Velop, and the R7800), but for some reason connecting through a satellite node was unreliable, which I can now only attribute to SOMETHING in or near this closet.
That would isolate things, yes.
You should test using two devices on your LAN, one wired and one wireless to the Orbi. You should not test your wireless network using a connection to the Internet. The Internet speed can vary, especially at the high data rates you mentioned (250 Mbps.)
You should test with a tool like iperf. You should use iperf on laptops or desktops which are powerful enough to handle the data rates, and which will report the detailed results of the test, including the average throughput.
We all want to make sure we have our gear setup to work at maximum speeed, but it's easy to draw the wrong conclusion with these tests.
UGGGGGHHHHH!!! Ok, the bad news is that my sporadic speed problem is back... I don't think the wire shelving in the closet was the main factor. The good news is that I am getting closer, but this demon is still proving difficult to exorcise.
Trying to figure out what was different between yesterday when my speed was good, to today when it is suddenly inconsistent again, I noticed that sometimes the Velop connects clients to channel 36, and other times to channel 153. Yesterday, I observed that my devices were connecting to a particular node on channel 153 (and speed was good), but right now it's using channel 36 (and speed is intermittent). That sparked a memory from when I got my first AC router a year ago. I recall that for some reason, I got poor throughput when using the lower 5ghz channels (36 - 48). The high channels (149 - 161) were fine, much faster.
While I can't directly control which channels the Velop selects, some trial and error this evening has revealed that when the client happens to connect to a node on channel 36, I get intermittently bad speed, while if the client is connected to channel 153, speed is excellent and consistent. That mostly corroborates with my Orbi experience, since the Orbi always uses the high channels for satellite communication and the low channels for clients... a laptop wired to the satellite gave good speed because that wireless link from the sat to the router was on channel 153, while a wireless client would be on 36 and would get intermittent speed. What doesn't totally add up is why I got good speed connected wirelessly directly to the router (again, on channel 36), the only explanation for which I can come up with was perhaps I was just lucky and happened to test during a good period.
Another aspect of this theory that is difficult to explain is, with the Velop, if I happen to end up in the "good" configuration where the client is communicating with the wireless node on the channel 153 radio, my assumption is that the backhaul to the wired node would then be on channel 36. Yet in this configuration, I get consistently good speed. Perhaps the nodes are better able to cope with interference than the presumably weaker radios in the client devices?
Here's a screen shot that shows the channels in use. The 2.4ghz channels occasionally register a little noise, 153 is clean, but 36 is showing considerably more noise (varies from -90 to -80, occasionally as high as -74).
I don't know if that's enough noise to cause problems, but still, it's hard to ignore as a potential factor. There are 2 other weak networks in my vicinity on channel 36, which maybe could be the source of the interference reading, but on the other hand, there are 3 other networks on channel 153 that are not causing an issue.
Does anyone know of any other potential sources of household non-WiFi intereference on the 5ghz band besides certain cordless phones? peteytesting, help!
Ok, after scouring my house I've identified a likely culprit...
A Vizio soundbar, which is in the living room (where I encountered the most difficulty) and also happens to be right below my office (where I also had problems during testing). This soundbar transmits wirelessly to the subwoofer (which also feeds the rear surround speakers), and operates on the 5ghz band. While I'm not going to make the mistake of once again prematurely pronouncing this mystery solved, I am very hopeful that this will prove to be the case. Sitting in the living room with this soundbar turned off, I can run test after test getting 250-350mbs+. Turn the soundbar on, speed becomes more sporadic and frequently drops to around 50mbs. Turn the soundbar off, speed goes back up to normal. Turn it on, speed goes down. I repeated this cycle perhaps a dozen or so times with identical results.
So, it is without a doubt *A* problem, but what remains to be seen is whether or not it is *THE* problem. What makes me hesitant is that in the WiFi scanner app, it still shows channel 36 as having more interference than 153, so I'm not sure what to make of that.
