5GHz wifi no interference, yet ping spikes?

WangGang

Reputable
Jan 24, 2016
92
0
4,640
Using Ethernet causes zero latency problems, but upon switching to Wifi ping will shoot up randomly. Wifi Analyzer confirms that I have the only 5Ghz network in the area. How is this happening>
 
5ghz is good for short range or point to point without anything blocking signal, like walls.
Ensure your using an non DFS frequency, low ones normally the best.
Also ensure the router is setup to use Wireless-N or AC and not A, as this is basically Wireless-G performance.
 


Router Is AC enabled, but my card can only go up to N. Also, which channels are non DFS? Currently on 161.
 
You can not set the DFS channels manually only auto set options can use that and many routers do not want to deal with the regulations so they do not use those channels. Used to be you could set routers with third party firmware to russia and then set them but this has been disabled in the later radio chipsets.

Pretty much there are 2 blocks you can use. 36-48 and 149-165. Be aware using 802.11ac uses 80mhz which completely uses the bottom block and all but 20mhz of the top block. The tri-band routers use both the low and high block so a single router can block 2/3 of the 2.4g and 160mhz out of 180mhz of the 5g block.

You can only see the AP announcement you can not see the actual data streams so these tools have limited ability to detect wireless signals. You would need a actual spectrum analyzer to find them. Even then it does little good. All you do is sit near a busy intersection and watch all the new wireless signals pop up because of all the cars that wireless AP in them. All someone does is drive by your house and you will get a spike of interference
 
DFS is channels 50-144.
https://en.wikipedia.org/wiki/List_of_WLAN_channels#5.C2.A0GHz_.28802.11a.2Fh.2Fj.2Fn.2Fac.29.5B17.5D

Both the 2.4 GHz and 5 GHz bands were chosen to be open because there is high environmental attenuation at those frequencies. 2.4 GHz is readily absorbed by water vapor (which is how microwave ovens work at 2.45 GHz). I forget what exactly absorbs 5 GHz. Oxygen? Anyway, this high attenuation meant these frequencies weren't very good for long-range communications, but made them ideal for short-range communications (your short-range comms would interfere less with other people's short-range comms). So the frequencies were made open and unregulated as long as you stay below 1 Watt.

So even if you don't have interference from other 5 GHz signals, you're going to get a lot of interference from the air, the walls of your house, people walking around, etc. When the signal strength fluctuates because of this, the error correction coding has to detect and fix the error, which takes some time, causing the ping spike.

Edit: Please don't deliberately use DFS frequencies with third party firmware. It interferes with newer doppler weather radar at airports - broadcasting on those frequencies could literally cause a plane to crash. That's why the U.S. restricted its use (DFS requires the router to immediately switch to a different frequency if it detects weather radar is in use). Some other countries do not restrict these frequencies because their airports do not yet have these new radars. And many third party firmware authors are ignorant or simply don't care and think they're sticking it to the man when they allow DFS frequencies to be used, when they're just endangering the public.
 
You can't predict it. They put some energy film on the glass at work. Before the wifi worked ok outside the building after you could barely even detect the ap. You could easily see though the film but it almost completely blocked the wifi.

Without very expensive equipment you can not tell much about wifi signals. There are many thing that cause interference and absorb signal. The only options a home user has is to change radio channels and hope to get lucky.

In general random delays in the the packets cause few problems. Even most video streaming services tolerate some random delays. Online games since they use the delay as part of the calculation to sync a user to the server do not tolerate much if any randomness.
 


Have you found a solution to this?