Performance

For performance testing, I knew right away how I wanted to do my testing. Putting it all together in an easy to read format, on the other hand, was a bigger puzzle. So as I mentioned in the opening, in my home I already have a rather robust network configuration setup. I have a switch in the office for all of the PCs that end up in there and a switch upstairs with POE for an HTPC upstairs and to power some of my security cameras. Then in the middle of the house, I have a network closet with our two printers, a large switch, three NAS, a cache server, the modem, a small server, and our PFSense Router. In that same closet, I’m running a Ubiquiti UniFi UAP AC Long Range access point. This replaced a basic router being used as an access point about 9 months ago.

Now in an ideal situation, I think the Deco would be best compared against a standard home router as I do think it is focused on the average user who would be upgrading from that. In my case, though I’m considering upgrading to a mesh setup over our current access point, so the Deco is getting a trial by fire, being put up against a proper commercial access point. I should also point out that my UniFi AP could actually be run in a mesh mode as well if I added more devices but I only have the one. So this is more of a look at how a mesh network compares to a single good access point. For my testing I used my Razer Blade Stealth ultra book, it has a Killer Wireless-AC NIC that has proven to be solid.

To put both to the test, a few basic tests wouldn’t really be fair. I decided to test the performance of both networks in every single room in our house as well as out front, out back, in the breezeway between the garage and the house, and the garage. I originally upgraded to the UniFi because I had really poor wireless performance, our house is old and I couldn’t even pick up wireless half way into our bathroom. For perspective, the bathroom is right next what we call the Warehouse where all of the LanOC hardware is stored when not in use. So getting wireless upstairs was spotty and you could forget trying to get on it outside or in the garage. To test the network I used Passmark Performance Test 9, its network test has been good for us in the past. I set a computer in the office to be the test server and then went to each room in the most likely location wireless would be used or in the middle in the case of our back room and I ran the test three times on each wireless device.

I documented the average transfer speed and the minimum. The average is obvious but the reason for the minimum was because sometimes wireless can be fast but if it is dropping to nothing it isn’t going to feel fast and it will cause issues in lag sensitive things like games. I took every result that was in the kb’s and marked it as a 1 or less to simplify the averaging and I also cut the ends off the averaged results to make things easier to read. Below is a table of all of those results. It’s a lot of data to go through. I was expecting the Deco network to have better results in areas like outside and in the garage because of the devices now places at the back of the house and upstairs where my original network is placed in the middle of the house but I figured that the mesh network would be slower overall. There are a few situations where the UniFi is faster like out in front of our house, in the bathroom, and also in the backroom, oddly enough exactly where the third Deco was placed. But there were a lot of areas where the Deco was faster including in the Warehouse where both of access points were located. The UniFi couldn’t be reached in the garage at all and was very bad outside as a whole. It seems that the Deco network as a whole is faster but this wall of numbers made it impossible to see.

To figure that out I made yet another table, this time averaging out the total results. This result put the Deco way ahead but I was curious how things would look if we took out the garage results where the UniFi had zeros. This helped and it even raised the Deco’s average numbers up as well. Then for the last one, I removed every result that was outside of the basic home, so the porch, out front, out back, breezeway, and the garage and did one more average to see what would be a better option if I didn’t want my WiFi reaching outside of the house. This brought the two closer again and raised the Deco all the way up to 221 MB/s average throughout my entire home. That is a number I can really get behind, that is 1768 Mbps, I don’t think you have to worry about your new mesh network slowing down your fat pipe anytime soon.