One powerful/expensive router vs multiple cheaper APs?

[BlobbyBlob2]

Hi there!

I'm trying to figure out if I should get one very powerful wifi router (something like the TP-Link AC5400 was recently on sale for $135) or two cheaper routers (2x ASUS TM AC1900s, which are selling for $55 each on sale), with one set up as an access point. I have a two story 2000sqft single family, wood frame home. Right now coverage is spotty in the garage, but I think that's because of a big brick fireplace which we are removing in a remodel.

I'm getting conflicting advice on this. This thread suggests two APs are always better than one, but my friend who is an IT professional is saying the extra complexity of setup, plus issues with AP hopping and reduced MIMO performance would make a single more powerful router like the TP Link AC5400 a better choice.

Any thoughts are much appreciated! Thanks!

 

[razel]

Hmm... With current routers as long as you aren't buying the cheapest of the cheap they all tend to use amplifiers on the antenna so they're all the same power. And to meet U.S. regulations it maxes out at 1 watt. You can use DD-WRT to bump it up, but you'll need to ensure that the chips don't start to cook themselves too quickly.

The ASUS is powerful, 2 would fine. Don't get too caught up with the TP-Links AC5400 number. I remember seeing that sale. It's got 3 radios 1 2.4 G that's 1Gig/s and 2 5G that's 2.1G/s they add that together to get 5400. In order to enjoy that bandwidth your devices must support it and most of the time need direct line of sight of the router to achieve those speeds. As soon as you turn a corner... realistically you'll get 200-300mb/s... so all of the sudden the puny AC1900 = 600 + 1300 isn't looking so bad anymore. MuMiMo is great, but only applicable if your devices support it and only those devices will enjoy it only when they want attention at the exact same times.

Yes, there's no getting around your situation. Wire it up best you can. To improve your coverage, treat WiFi like light a light that happens to go through walls. Most of the time simply elevating it and away from it's dark tucked away burrow vastly improves coverage.

 

[BlobbyBlob2]

Thanks Razel - that's helpful. Do I read between the lines correctly that you're suggesting either solution should work fine? Which would you choose?

 

[mnewsham]

I personally prefer multiple wired access points. Much greater coverage and much higher overall throughput throughout the house.

If you don't have ethernet runs already, you can always look into MoCA assuming your house is wired for coaxial.

 

[BlobbyBlob2]

Thanks mnewsham. We will be running ethernet cables to all rooms during the remodel, so having multiple APs should be easy, if it is better.

 

[mnewsham]

Thanks mnewsham. We will be running ethernet cables to all rooms during the remodel, so having multiple APs should be easy, if it is better.

Multiple access points is pretty much always better, as razel already noted, the maximum range on consumer wireless equipment is limited by law in the US so as not to allow people to purchase a router/access point with a radio so powerful it disrupts your neighbors wifi or even potentially cause issues with EMS frequencies and similar.


I would recommend looking at the Linksys Velop mesh system, it allows for an ethernet wired backhaul. Most mesh systems use a wireless backhaul which can significantly degrade performance the further you get from the primary router. A wired backhaul means each access point is getting it's own dedicated ethernet link to communicate back to the primary router. This greatly reduces latency and throughput and should give the best performance.

 

[BlobbyBlob2]

Multiple access points is pretty much always better, as razel already noted, the maximum range on consumer wireless equipment is limited by law in the US so as not to allow people to purchase a router/access point with a radio so powerful it disrupts your neighbors wifi or even potentially cause issues with EMS frequencies and similar.

Thanks mnewsham. Should I be worried about AP handoff as an issue with multiple access points? I would set them up with identical SSIDs/passwords and use different channels for the two APs.

 

[Rifter]

multiple will always be better for that big of a space.

 

[mnewsham]

Thanks mnewsham. Should I be worried about AP handoff as an issue with multiple access points? I would set them up with identical SSIDs/passwords and use different channels for the two APs.

That's why i recommended a mesh system with a wired backhaul, the mesh system SHOULD detect when you're in range of one AP over the other and automatically switch you.

With normal access points set up with identical SSID/PW it remains up to the client device to decide which network to connect to. And generally newer client devices (phones are most common exmaple) they should detect and switch networks fairly quickly (within a few minutes). However, this is far less robust than a mesh system. (assuming the mesh system is working as intended of course)


I personally have no experience setting up mesh networks as my house is fairly small and I don't really have a need for it.

 

[BlobbyBlob2]

Thanks mnewsham & rifter. The only downside of Mesh wifi networks is cost - each node is about $200, whereas the Asus tm1900s I can get for $55 each. Although there are obviously major benefits to mesh as you suggested.

 

[mnewsham]

Thanks mnewsham & rifter. The only downside of Mesh wifi networks is cost - each node is about $200, whereas the Asus tm1900s I can get for $55 each. Although there are obviously major benefits to mesh as you suggested.

https://www.amazon.com/dp/B01N2NLNEH

3 nodes for $400. $133.33 each. Though you're still right, the price will be a good chunk higher either way if you go for a mesh system.

 

[JackMDS]

If you can lay a CAT6 to the second floor.

Then a Good Wireless Router at the source would cover the first floor.

A second Good Router configured as Access Point can cover the second.

While every one in the above posts has something that is true, the decision is always according to the specific enviroment and the owner perception of life.

While owning a Lamborghini ($200.000) in NYC can boost the moral of any one. A Smart car ($15,000) would provide the same practical service (and One do not have to worry about parking, or that it will stolen within a day).

 

[BlobbyBlob2]

My IT friend mentions one other benefit of the single router is ease of debugging.

 

[mnewsham]

My IT friend mentions one other benefit of the single router is ease of debugging.

It's so rarely an issue in a residential setting I don't really see that as a real benefit.

Troubleshooting is always easier when you've got less pieces to the puzzle so to speak.

But when it is set up properly and running, as it should be 99% of the time. I'd rather have a more robust network than have an easier to diagnose network that doesn't give me good performance where I need it.

 

[sdifox]

2000 sqft on two story is not that big for one router. As long as it is well situated, at center of house, you should be fine.

 

[razel]

I think I also see that ASUS TM1900 sale. That is the Tmobile one. That's basically the ASUS RT-AC68 which is a benchmark router. It's probably 3-4 years old. You will want to put on the ASUS firmware, but it appears as if it depends which hardware version you get. Once you can though, it's awesome. DD-WRT, Tomato, but I ended up going back to the ASUS Merlin firmware.

In your case the moderator's suggestion is what I'd do. I love onHub/GWiFi myself. The GWiFi 3 pack has been available on sale frequently in the $250 range. $83 each basically. I suspect Google will come out with another cheaper revision new next few months. Maybe even one that is basically Google Home with more radios for WiFi MESH. You'll hate GWiFi though it if you love tinkering with WiFi settings. Love it otherwise for friends and family.

 

[VirtualLarry]

I think I also see that ASUS TM1900 sale. That is the Tmobile one. That's basically the ASUS RT-AC68 which is a benchmark router. It's probably 3-4 years old. You will want to put on the ASUS firmware, but it appears as if it depends which hardware version you get. Once you can though, it's awesome. DD-WRT, Tomato, but I ended up going back to the ASUS Merlin firmware.

Yeah. I haven't messed with the T-Mobile hardware variant, but the AC68U / AC68R is a competent bit o' kit, and with third-party firmware, is insanely powerful and flexible. Even the OEM firmware isn't too bad on these.

 

[MrBill10]

I studied up on how the consumer wifi companies exploit marketing jargon, then ditched my wireless modem in favor of a wired EdgeRouter Lite and access points.

Here's the FCC maximums (https://www.air802.com/fcc-rules-and-regulations.html) for 2.4 and 5gHz. Legal maximum of any 2.4 radio is 30dBm = 1 Watt; with this you can legally use an antenna with 6 dB of boost. I mention legally because its easy to add a hi-gain antenna and really get some distance...

But you have to start with a radio that has balls. My AP's are Engenius ECB350's which put out 29dBm = 800mW. I also have a few directional antennas that work very well.

(Here's an easy cross-reference (https://www.cisco.com/c/en/us/suppo...an-wlan/23231-powervalues-23231.html#backinfo) of dBm to mW, about half way down the page.)

The OP mentioned the Archer C5400; I looked it up (http://www.tp-link.com/us/products/details/cat-5506_Archer-C5400.html#specifications) and it puts out a whopping 100mW on 2.4, and a spectacular 200mW on 5gHz...

So if you want a powerful signal, start with a powerful radio and add a hi-gain antenna.

/edit: Asus does not list the transmission power in the RT_AC68U. I did find reference to it being 500mW here: https://w.wol.ph/2015/08/28/increasing-transmit-power-asus-rt-ac68u/

 

[razel]

The antennas that boost don't create additional energy. They just change the shape of the signal coming out. Instead of what's usually a doughnut where the antenna is at the middle, the 3dbs usually are a bubble where the end of it starts at the antenna and the 6dbs often are a flatter bubble. They claim gain, since it at the further ends there is more WiFi energy. You can create your own 9 or 12db of gain by using aluminum foil. Like these sail reflectors.

The TPLink onHub is one of the few consumer products with an active antenna for 2.4G. Ifixit's Step 7 pic 2 shows the 2.4 high gain design that looks a lot like Mr. McNeil's build. Step 10 shows the heatsink smartly designed as a reflector and I would assume is grounded, which makes it active.

 

[frowertr]

Its not boost, its gain. For every 3dBi of gain, you double output power. So if the transmitter is 100mW and you attach it to an antenna with 3dBi of gain, your EIRP is now 200mW assuming no signal loss (very little in such a short run in an AP) through the cable from TX to antenna.

It depends upon the band, but the FCC regs for wifi may concern EIRP output not IR output. So if its a maximum EIRP of 1W, then you have to feed the antenna less than that.

Only reason you would want to have an EIRP of 200mW or more is if you had a directional antenna attached. If you used it with an omni antenna, the client stations would never be able to talk back to the AP as their distance would be too great.

 

[razel]

So you're saying these gain antennas create energy turning 100mW to 200mW? WOW. Let us know how then we'd solve major world electrical problems.

These antennas just change the shape of how 100mW of energy is radiated. Think of it like a balloon filled with air... say 100mW of air. Instead of a doughnut ballon where signal at it's horizontal farthest ends is weak or none you flatten that ballon, less energy vertically but it gets moved horizontally that you get 3db of signal on the horizontal plane.

 

[JackMDS]

Its not boost, its gain. For every 3dBi of gain, you double output power. So if the transmitter is 100mW and you attach it to an antenna with 3dBi of gain, your EIRP is now 200mW assuming no signal loss (very little in such a short run in an AP) through the cable from TX to antenna.

As I recall from sixth grade.

"In physics, the law of conservation of energy states that the total energy of an isolated system in a given frame of reference remains constant—it is said to be conserved over time. Energy can neither be created nor destroyed; rather, it transforms from one form to another".

ASAIK, this law was never broken and it is Not Fake either.

 

[MrBill10]

For every 3dB of antenna gain you double output power in one direction at the cost of halving power in another direction.

 

[frowertr]

So you're saying these gain antennas create energy turning 100mW to 200mW? WOW. Let us know how then we'd solve major world electrical problems.

These antennas just change the shape of how 100mW of energy is radiated. Think of it like a balloon filled with air... say 100mW of air. Instead of a doughnut ballon where signal at it's horizontal farthest ends is weak or none you flatten that ballon, less energy vertically but it gets moved horizontally that you get 3db of signal on the horizontal plane.

Its called the rule of 10s and 3s. Look it up if you don't believe me (https://www.youtube.com/watch?v=VqoUS-ecw4Y). It's not like it's a big RF secret. I learned this when I was a Ham radio operator 20 years ago. Every 3dBi of passive gain in an antenna doubles the radiated power. Every 3dBi of loss halves its radiated power. For every 10dBi of gain you need to multiply the beginning ouput power by 10. Makes no difference if it's an omni-directional antenna or a directional antenna. The rule is truth for RF.

Not sure why you think this solves the word's problems...

Your describing radiation patterns of higher gain omni-directional antennas which is correct. But the rule of 10s and 3s still exists for those too.

 

[razel]

Awesome. Tweet Musk. You may have superpowers that can help the world.