Performance Wireless

Performance Wireless

For our tests, I decided to keep things simple, as simple always works. We emulate what you would get performance wise inside a house with a multi-level concrete floor. We'll look at dual-band (5 GHz), single band 2.4 GHz and single band 5 GHz. We first test a single router, not the mesh just yet.  The router (host) is positioned at ground level and we test with a client (laptop) in four stages:

5 Meters	10 Meters	10M Through Concrete Floor 1	15M Through concrete Floor 2 (attic)

Our test is simple, a NAS linked up through 1 Gbit/s Ethernet will have FTP enabled. It's through FTP that we move a large multi-GB file towards my laptop, which has dual-band WIFI. Unfortunately, I do not have triple band WIFI on my laptop. But we're trying to find out if this router can fill all corners of the house with some sort of WIFI signal, dual-band 5 GHz is plenty for that.

For example, if you have your router on the ground floor, and you want to watch Netflix in Ultra HD on your attic where you build your home-cinema setup or gaming crib. For Netflix, that signal would require at least roughly 25 Mbit/s. Netflix Ultra HD - in fact uses just over 15 Mbit/s, but you need a little reserve for caching and error correction. So 25 Mbit's at the worst location would be the absolute total minimum goal to reach. It would also be an acceptable figure for a bit of web browsing of course.

Now before you look at the results, let me state this, results will vary anywhere and everywhere on this globe. There are so many variables that have an effect on WIFI it's staggering. Also, you can have a great WIFI router (HOST) but if your laptop (CLIENT) has poor WIFI, you're not going to achieve great numbers. The same can be said about the building you are in, reinforced concrete/steel walls? You'll get lower performance. Are there more factors involving WIFI performance? Yes, the number of WIFI routers in your neighborhood and even equipment like your refrigerator turning off/on and hey now. Do you know what frequency a microwave uses? Yep, 2400 Mhz, and as such can have an effect on performance. That makes WIFI tests a rather subjective and difficult to replicate thing to test. Our results are pretty much worst case scenario, I am in a building with steel reinforced concrete walls and I just sniffed for wifi hotspots in the neighborhood and stopped counting at 40 (!).

None the less, the ASUS manages to give me proper results throughout the house. Let's have a peek how the router behaves, again we have not enabled mesh just yet:

The numbers you see are measured in Mbit/sec, divided by eight bits, that's your performance in megabyte per second. So yes as you can see, close range dual-band performance kicks ass. Basically, two 5G AC streams at ~5200 MHz and 5500 Mhz are combined. At best we pass 500 Mbit/s, these are real FTP transfer rates, so data was moving at ~65 MB/sec.  If you have 5 GHz single band, you're still at a ~250 Mbit/s, while on our attic through two concrete steel enforced walls at 120 Mbit/s. The 2.4 GHz is the one that disappoints a bit, but some older devices do not support anything else. My laptop locked in at a 150 Mbit/s connection. At best we reached 110 Mbit/s, and all the way in the attic, a poor 32 Mbit/s.  

But hey, we've got two of these puppies right?

The AIMesh performance

So on the previous pages, you have been able to read that we have two of these routers. We configured the Mesh and placed Router 2 in AIMesh node configuration on the first floor, as something in-between good and bad. Here are the results:

So yes, maximum performance numbers remain roughly the same. Actually, the results are fairly close towards each other, until you look a little more in-depth. Dual-band performance overall became a bit stronger and felt more steady. The 5 GHz single band was showing some decent gains. And 2.4 GHz now in the attic reaches 54 Mbit/s. Not super, but fast enough for your Ultra HD Netflix needs of course.

Let's break down the frequency bands and split them up relative to the mesh performance.

Again I like to note, these are actual transfer rate figures, not the reported WIFI connection speed.

In the end, we feel the mesh is not worth the extra money invested (for this particular household with this particular router). In fact, with a 50 USD/EURO repeater, you might get close to similar results. Also, we have ground floor, floor 1 and the floor 2 (attic). With one router in the middle at floor one, your performance would easily match the mesh results.

So was the mesh not working properly? Oh no Sir, it's working fine. It just that one single AC5300 can quite easily fill the house all by itself, as it is a very strong signal router. And that is the lesson learned today.