Ethernet over powerline is an older technology than the newer mesh wifi router approach. As you've described, it uses the home's electrical wiring to carry an ethernet signal, and then an wifi access point (WAP) device is connected at the far end to emit the wifi signal. There can be at least a few problems. First, powerline variability (and possibility of just one poor condition of electrical wiring at one transit point on the signal path) can cause disruption of the signal. Second, the WAPs can interfere with one another. Third, in some houses, if the powerline devices are connected to sockets of different electrical phases in the home's system, the signal may not carry at all. I don't know if this approach produces the same major known problem for repeaters, which is with a repeater the wifi capacity (which some mistakenly call speed) is cut in half, but all the same, it's not the preferred approach. If using WAPs, running ethernet cable is better.

So-called mesh systems know which device is connected to which antenna and send the signal appropriately. I have one, the Netgear Orbi system, and it's fantastic for speed of connections and consistency of connections. A family member uses the Google product and has had similarly positive results. That's the recommended solution of today when having a strong wifi signal throughout a home is a challenge.

"That's the recommended solution of today" - whose recommendation?

Glad the mesh is working for you. And of course, every tech has its advantages and disadvantages. I'm just sharing that the 'older' technology of powerline works well and is pretty fast, and may be viable solution for many. (Note anything made in the last five years or so should be up to date, no real need to chase the latest and greatest).

And it's notably cheaper for the most part than most mesh systems. A simple powerline + wifi pair costs less than fifty bucks.

Again, I'm not saying mesh is no good or doesn't have its own advantages, but I think powerline is underrated as a solution and may be helpful to many. My own experience is unambiguously positive - cheap, functional, fast, stable.

That's a fair point. I should have added 'or were considerably more expensive.' I looked at some mesh network kit in particular that didn't have the 'backhaul' technology (separate frequency specifically for in-house data) or had less up-to-date wifi speed tech (the various AC1200 etc). Some had dependencies on a central network (they need to be able to phone home). I can't say I covered all the brands, either (market availability issues). I actually think mesh networks and better wifi are likely the better technology in the long term (still evolving), but as a stable, functional solution, I think powerline are underrated and likely helpful to many.

Anyway, I'm not selling anything, just found that it worked very well for me.

The so-called backhaul feature is part of what makes the mesh approach superior. My understanding is limited but basically what's up is that the return signal path to the router doesn't consume Wifi bandwidth and keeps the airwaves free of the additional traffic. There may be a similar affect with WAPs that are wired (by Ethernet wire or powerline) but you're still going to get some signal interference and ambiguity in the mid points where signals from more than one antenna are received.

There are a small number of companies that produce the enabling semiconductor devices and many competing router products from different manufacturers use the same chips. All that differs from one to another is the firmware (internal software that runs the product) and the user interfaces. Firmware matters greatly and it's not something that's perfunctory, I just wanted to mention that many share the same hardware parts, in part.

If what you have works for you, that's great. Same for me. For the reasons mentioned before, I think you're selling these new devices short. They're not "evolving" any more than any other technology product types evolve. They're not unproven, not unreliable, and not in a "wait until next year when they stabilize" phase. They work very well today, they're ready for prime time, and they're selling well because of it.

I don't think wireline is underappreciated, it's just an old solution for which today there are better choices. And wireline is a sometimes problematic approach for a number of reasons as I mentioned. It is a less expensive approach, typical of technology that's not the latest and greatest. Running an ethernet wire is an even cheaper way to go with a WAP, you don't need to buy the wall warts.

Not all mesh products have a separate and dedicated backhaul, running dedicated ethernet wire is not always feasible nor cheaper, and as noted, the difference in price can be considerable. I've seen far more differences in offerings/features than you've indicated, but perhaps at the high-end they're more comparable.

I'm not trying to argue superior - powerline is not the 'lastest and greatest.' But still a very viable solution at a reasonable price. Speeds are quite good in my experience, but I'm not a measurebator.

I presume for many who wish to try and are concerned it may not work on their electrical system, buying from someplace with a return policy is an option.

"buying from someplace with a return policy is an option"

YES, especially with things that are technical/electronic.

Just last week I wanted to buy and set up a cheap wifi laser printer/scanner device. Being skeptical, I chose a store with a good return policy but not the best price and bought a Brother. I took it home and it wouldn't connect to wifi. I tried everything, it just wasn't going to happen. I returned it, full refund, and bought a Canon at a different store. Success with the Canon, the setup was as easy as the Brother should have been.

Ideally, if you want a reliable home network that covers the entire house in a metropolitan area, you want one low-powered access point per one or two rooms. Where we live, the noise floor is really high. I can see at least 200 other WiFi networks; and that's not even counting microwaves and other equipment that operates in this frequency range.

Cranking up the power of the access point doesn't really help much, as the mobile device simply won't have enough power to reply back. In the end, you just end up contributing to the global problem without making things better for own local network. Instead, having small point-of-use access points improves the situation dramatically. This is what successful commercial installations in office buildings, event centers, or hotels do as well.

The WiFi standard supports seamless roaming between multiple access points, and has done so pretty much from day one. But many consumer access point have buggy implementations. If you buy an access point that is meant for multi-point deployment, things will work much more reliably. So, mesh access points are great, WiFi repeaters are potentially questionable.

Also, roaming is initiated by the client not by the access point. So, it works much better, if the power of each individual access point is dialed down. If your mesh access points support a separate backhaul, that's a great option. An even better option is hardwiring with dedicated ethernet wiring to a central location.

Powerline networks are somewhere in-between all of this. If it works, it is slower than ethernet but possibly fast enough for WiFi. But if it doesn't work, then there really isn't any good way to debug it. There are lots of things that can interfere with it. Also, it causes crazy bad interference for HAM radios.

"For example, a wall about 5 meters thick at an angle of 45 degrees represents a 1 meter thick obstacle to the Wi-Fi signal."

15 foot thick wall?

Pythagoras and the discoverers of trigonometry would roll in their graves with this display.

I don't know what is being referred to but if the person is saying the hypotenuse of a right triangle is 15 (the path through the wall at a 45 deg angle), that would make give the wall a thickness of about 10. If saying the wall is 15 feet thick, the angled path through it at a 45 degrees (the hypotenuse) would be about 22.