I am a complete block head. I must have scrambled my brain with 2.4Ghz. I maxed the connection of the EDIMax. That is the throughput. I maxed it out! That's why the test results are so poor with it. I'm going to toss it in the trash with the TP-Link. Never buy cheap wifi. NEVER. New test speeds soon.
Needless to say, I have crappy Wifi speeds. It is SO time for eero to get installed!
I've had to pull the speed testing results. I just don't trust my method and embarrassingly enough the device I used to test with. (SurfaceBook) I'll update this soon. Check the forums thread below for conversation about this.
I almost forgot a major issue that put me on the road to replacing Ubiquiti in the first place. Connection troubles. I started to jump into phase 2 of this speed testing and use my android phone to do it with. It won't connect to the UBI. Nexus 6P will not get a connection to the UBI. I remember having issues with it long ago as well. That's why I employed a secondary AP just for me. The first Android test to EDIMax got roughly 30 Mbits/s. Up and down. FYI.
I feel bad for the UBI. I'm sure the higher end products are good but I find them comparable in price if not more than their competitors. See this thread.
Pre eero testing
I can't just throw eero into production without testing my old gear. The problem is, eero is here so I'm in a hurry! I thought I would get a tracking number so I would know how far out it was. Turns out, I got the tracking number 6 hours after it arrived.
Here is the unboxing video.
It’s probably good to fill you in on how I got to this spot in time with Wifi. Quickly. I have have the cheapo 2.4Ghz Ubiquiti product. It comes in a 3 pack from Amazon for $199. I did this several years ago trying to get by on the cheap. Eero is part to blame in my woes because I was supposed to have eero by Summer 2015. Then the Fall of 2015 and finally promised February 2016. I purchased a couple of AP’s to get me by. One TP-Link, one EDIMax.
Here are the links to what I'm testing today.
I usually resell or give these things away but not the TP-Link AP. It went in the trash and it felt good in doing it. The best experience I’ve had with WiFi prior to the EDIMax was with Apple. Crazy as it sounds to my Windows fans but the Airport Extreme router is very capable. It’s extendable like the others and halves the speed of each extension. It worked much better extending than any other product I’ve tried to extend. Granted, this was 5 to 7 years ago too. At one point I used one as a router and another as just an AP. It still doesn’t cover my square footage however.
I now have the Ubiquiti, now referred to as UBI, and the EDIMax deployed. The EDI has way better range but still only covers a single floor. Outlying areas of that floor are weaker however. Upstairs and downstairs coverage is very weak. UBI would be fine as 2.4 but it would need an AP per 1/3 of a floor. With 3 floors this would get expensive. I also wouldn’t go with the 2.4 only product. I would get into the AC product which would drive up the price even more. My guess is if the EDIMax product was roaming product it would cover ½ a floor instead of the UBI 1/3rd. It's not so we won't argue that guesstimate.
I have 25Mbps up and down Internet at my house. The EDIMax seems to be taking full advantage of the speed provided and the Ubiquiti is having trouble with the upload speed but peaking the download.
I have test numbers for you. They were mainly for me to track my testing and get a feel for the gear but if you like that kind of thing, geek out. If not, scroll down for the paragraphs. It's all summed up in the writing.
When UBI is fringe it gets ugly.
My feeling is that if I had a higher speed connection the Ubiquiti system would not be able to keep up. If I had only a small apartment to cover I would purchase the EDIMax again. At $49 and providing 802.11ac it is a steal. My square footage laughs at the EDIMax however.
My other issue is that I need a router that can be separated from the WiFi. My Internet handoff is in the basement. I’m on fiber and have only an Ethernet jack to plug into. (Seriously nice but I am moving soon and will be faced with Xfinity) Eero may present a problem if you do not use it as the router so I will have to work through that. It may just be that I have to purchase an addition access point to cover everything. I may be able to get creative with CAT 5 wiring too.
Power Over Ethernet
I may be a fringe user, or a power user but I like power over Ethernet, or PoE. The power has gone off at my house before while I’ve been working in a sunny room and I never knew it. That’s because all my UBI AP’s are PoE and backed by battery backup. So is the router, servers, NAS, and desktop PC. All on battery backups. I won’t be able to do that easily with the eero system. I don’t think the “Accounting Department” ie, Spouse, will let me put a UPS at each eero point. I’ll research later on how small I can get with battery backup.
I also use Circle with Disney in my house to secure my kid devices. I don’t have any of my devices monitored. The MAC addresses are listed in Circle but not assigned profiles. Just to be clear, I powered the Circle OFF prior to testing. Just for fun, I plugged it back in and retested. Same results. If you are not protected by Circle your speeds are safe.
Lessons Learned and Notes
This is a learning post. I needed to do some testing and simply wrote it up as I went. Good or bad writing, good or bad testing skills, right or wrong. It is what it is. What did I learn?
A $50 EDIMax router/AP is pretty darned good.
It only has a 100mbps port however.
I have dead spots with both EDI and UBI.
UBI is fine if you have an AP in every room you need access in.
Neither of these products as configured in my house today reach outside well.
I thought the EDIMax was faster. It feels faster but I think it’s simply coverage. It covers better.
I like PoE.
If you have ever heard that Ubiquiti products are “noisy,” they are. There are no moving parts but the electronics sing when they are pushing data.
I'm also finding that each UBI AP (i have 3 deployed) is different. Speed wise, and noise wise.
These are single tests. No AP was hurt in this test nor were they under a load during this test. That is an entirely different set of circumstances.
Want to discuss? Here is a comments thread for this post. Here