What's new

Forgo high-end router totally and just go with APs?

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

reddevil0728

Regular Contributor
  • 3 storeys house wired up with cat6 cables
  • all will be terminated at the same location
  • ISP providing D-Link DIR-868L

Original plan:
  • Disable wireless feature on DIR-868L and use it as a wired router
  • Termination end of the CAT6 cables will be connected to DIR-868L
  • Purchase a high-end router (either AC2400 or AC3200) and place it on the 2nd floor in the centre of the house (centre of the 3 storeys and centre of both ends of the house)
  • connect the the router to the RJ45 wall socket and use it as an AP

Thinking of:
  • Getting a cheap wired/wireless (disable wireless function) gigabit router
  • Termination end of the CAT6 cables will be connected to this router instead
  • Is it true that as long as the wireless function is not being used and using it as a wired router, as long as it is a gigabit router, the wired performance will be the same regardless of whether if it is a high end or a low end router (so doesn't matter whether if it is a N or AC router?)
  • place the DIR-868L on the 2nd level
  • connect the the router to the RJ45 wall socket and use it as an AP
  • buy 2 cheap AC AP (d-link dap-1665) or cheaper AC router and use it as an AP and place one of them on the 1st floor and another on the 3rd floor
  • connect the the AP/router to the RJ45 wall socket and use it as an AP

Would this be better good idea? Maybe sacrificing wireless throughput a little by not buying such a high end router but at least getting better wireless coverage since now there will be a wireless AP on each floor?
 
Last edited:
Your second option will give you the most complete coverage with an AP on each floor. I would recommend that if the third floor AP is in the left wing of the house, put the second floor AP on the right wing of the house and then back to the left wing for the first floor AP.

However, managing multiple devices over the years for what is supposed to be a simple home network is not what brings me pleasure.

You don't state the size of your home, but I would think that unless you live in a mansion, your first suggestion will work the best overall.

Though you might consider using the RT-AC87U as the main router and the ISP supplied DLink as the AP instead. Again; opposite wings of the house and on different floors if you can't move the main router position to the second floor, of course. If you can have both on the second floor and at opposite ends of the house, that should provide almost as much performance as having three routers / AP's will. And having to manage all those different ssids (recommended) and channels too.
 
Your second option will give you the most complete coverage with an AP on each floor. I would recommend that if the third floor AP is in the left wing of the house, put the second floor AP on the right wing of the house and then back to the left wing for the first floor AP.

However, managing multiple devices over the years for what is supposed to be a simple home network is not what brings me pleasure.

You don't state the size of your home, but I would think that unless you live in a mansion, your first suggestion will work the best overall.

Though you might consider using the RT-AC87U as the main router and the ISP supplied DLink as the AP instead. Again; opposite wings of the house and on different floors if you can't move the main router position to the second floor, of course. If you can have both on the second floor and at opposite ends of the house, that should provide almost as much performance as having three routers / AP's will. And having to manage all those different ssids (recommended) and channels too.
Erm..

Imagine the house is like a rectangle, so it's more of a back to front thing rather than a right to left thing.

original plan
zbGtJap.jpg

thinking of
OgofT60.jpg


pardon the drawings and scribbling

So you are saying it's not a good idea to put all of them in the centre for 1st, 2nd and 3rd storey but alternate it like front, back front?

Why so actually?

It's about 3000sqft built-in over 3 storeys. btw the house is all concrete and bricks. for the wall.

i think i might not have been clear.

so the termination of all the cat 6 wires will be at this cabinet outside the house, and the cabinet is metal. so 1) the location is not ideal 2) the metal cabinet will act like a faraday cage?, so basically it will render any wireless router placed in there pretty much useless? but will still need to have a wired/wireless router there to connect the termination of all the cat 6 cables to the router?

the original plan is to have the ISP provided router to work as a wired router for the termination of cat6 cables, and get a RT-AC87U place it on 2nd floor.

or the new plan which might be cheaper and more thorough coverage (at least better coverage for 5ghz) but wireless throughput might not be as high as having RT-AC87U.
 
Add APs to get better coverage and better speeds in those covered areas.

A 3 story 3,000 sq. ft. home with wallboard walls (not plaster or masonry) would need about 2 APs plus the router's WiFi. Double that for plaster or masonry interior walls.
 
Add APs to get better coverage and better speeds in those covered areas.

A 3 story 3,000 sq. ft. home with wallboard walls (not plaster or masonry) would need about 2 APs plus the router's WiFi. Double that for plaster or masonry interior walls.
Is there a difference in "wired performance" between a Gigabit Wireless-N router and a Gigabit Wireless-AC router with wireless feature turned off on both routers.
 
If there is, it is a design or hardware fault if we're talking about just the switching side of the internal LAN network.

Of course, there can be huge differences between the routing performance between the internal LAN and the WAN / internet between routers of the same class and especially of different classes like your question asks about.


The following link shows these differences.

http://www.smallnetbuilder.com/tools/charts/router/view


When set to WAN to LAN throughput and is measured for wired connections only.
 
If there is, it is a design or hardware fault if we're talking about just the switching side of the internal LAN network.

Of course, there can be huge differences between the routing performance between the internal LAN and the WAN / internet between routers of the same class and especially of different classes like your question asks about.


The following link shows these differences.

http://www.smallnetbuilder.com/tools/charts/router/view


When set to WAN to LAN throughput and is measured for wired connections only.
"Of course, there can be huge differences between the routing performance between the internal LAN and the WAN / internet between routers of the same class and especially of different classes like your question asks about."

Don't particularly understand this part of your post.

Say for example I got a 1GBPS internet, i got a Gigabit Wireless-N router and a Gigabit Wireless-AC router with wireless feature turned off, the primary purpose is for me to connect the other end of the CAT6 cables so that I can have a RJ45 point in each room.

Are you saying, theoretically speaking, I may not be able to get the FULL 1gbps speed via wire if I am using a WIRELESS N router, but I will be able to get the FULL 1gbps speed via wire if I am using a WIRELESS G router?
 
No, I am saying that depending on the specific routers you are comparing, one may be significantly better than the other.
 
No, I am saying that depending on the specific routers you are comparing, one may be significantly better than the other.
Oh so even if using it for wired connection, different gigabit wireless routers will still have different performance level for wired connection? Is my understanding right? And looking at WAN to LAN throughput will give a good idea the performance as it measures the download speed you can obtain using wired from the Internet while lan to wan is the upload speed you can upload to the Internet via wired connection?
 
Oh so even if using it for wired connection, different gigabit wireless routers will still have different performance level for wired connection? Is my understanding right? And looking at WAN to LAN throughput will give a good idea the performance as it measures the download speed you can obtain using wired from the Internet while lan to wan is the upload speed you can upload to the Internet via wired connection?

Yes and No.

The performance of wired devices to other wired devices on the same internal / local network will (or should) be identical. A switch is easy to do today even in $6 examples, though I wouldn't buy those for my networks. This internal switching should be at the GB speeds of the ports.

The performance will vary when it is used as a router. Specifically when routing between the WAN (internet) port and the LAN ports. That is the raison d'etre for a router: to route packets between remote and local networks.

WAN to LAN and LAN to WAN measure as you indicated. Keep in mind though that this may require having hardware acceleration on and active. Depending on the features you enable, this may not be the case for each specific use case.
 
Oh so even if using it for wired connection, different gigabit wireless routers will still have different performance level for wired connection? Is my understanding right? And looking at WAN to LAN throughput will give a good idea the performance as it measures the download speed you can obtain using wired from the Internet while lan to wan is the upload speed you can upload to the Internet via wired connection?
Gigabit switch technology has long been at the point where there is no significant difference in performance from product to product. All gigabit switches move packets at wire speed among ports.

You'll have a Gigabit Ethernet connection to the router turned AP and to any of the LAN ports on its switch.
 
Yes and No.

The performance of wired devices to other wired devices on the same internal / local network will (or should) be identical. A switch is easy to do today even in $6 examples, though I wouldn't buy those for my networks. This internal switching should be at the GB speeds of the ports.

The performance will vary when it is used as a router. Specifically when routing between the WAN (internet) port and the LAN ports. That is the raison d'etre for a router: to route packets between remote and local networks.

WAN to LAN and LAN to WAN measure as you indicated. Keep in mind though that this may require having hardware acceleration on and active. Depending on the features you enable, this may not be the case for each specific use case.

Gigabit switch technology has long been at the point where there is no significant difference in performance from product to product. All gigabit switches move packets at wire speed among ports.

You'll have a Gigabit Ethernet connection to the router turned AP and to any of the LAN ports on its switch.
Ok pardon me for my confusion, but I'm seeing mixed information from you and L&LD.

As you can see from the initial question about the best networking setup, a router would be needed to connect the termination of the cat6 cables.

So would the routing performance of the wired/wireless (with wireless feature disabled) router be obtained from the WAN to LAN, LAN to WAN and Total Simualtaneous benchmark where supposedly the higher the number the better? Or will the difference in routing performance be negligible for any not that old Wireless-N gigabit router and a Wireless-AC gigabit router?
 
Last edited:
You can call this the packets-forwarded per second (PPS), LAN to WAN and vice-versa.
LAN to LAN of course is at switch speed.

For any consumer WiFi router from a name brand, made in the last few years, the PPS rate without QoS rules is most certainly faster than the WAN's speed, for residential DSL or Cable. Network Address translation (NAT) is a key contributor to speed and overhead for PPS.

Some QoS rules slow the PPS.

I prefer to keep high packet rate connections in the LAN connected to a known-good switch to maximize LAN-to-LAN speeds. Some routers' internal switches may be as good. But I tend to use WiFi router ports only for 10/100 connections and the uplink to the first switch.
 
Ok pardon me for my confusion, but I'm seeing mixed information from you and L&LD.

As you can see from the initial question about the best networking setup, a router would be needed to connect the termination of the cat6 cables.
The speed of your connections via wired (your LAN) will not be any different no matter what router you buy since gigabit switches all run at the same speed.

Your LAN to WAN speed will vary based on the router since that is the piece that is 'routing' your traffic to your ISP.

Hope this helps.
 
The speed of your connections via wired (your LAN) will not be any different no matter what router you buy since gigabit switches all run at the same speed.

Your LAN to WAN speed will vary based on the router since that is the piece that is 'routing' your traffic to your ISP.

Hope this helps.
So LAN to WAN is upload speed?

What about WAN to LAN?

When I look at the benchmark, I should look at the WAN to LAN numbers, LAN to WAN numbers, Total Simultaneous numbers where the higher the numbers the better it is?
 
So LAN to WAN is upload speed?

What about WAN to LAN?

When I look at the benchmark, I should look at the WAN to LAN numbers, LAN to WAN numbers, Total Simultaneous numbers where the higher the numbers the better it is?


You've asked and have been answered a few times on the same points. Please re-read this thread from the beginning again completely and slowly. :)


Yes, LAN to WAN is upload speed from your local / internal network to anywhere on the internet.

WAN to LAN is from anywhere on the internet to your local network device.

In the benchmarks, your assumptions are basically correct but in the Total Simultaneous anything above around 30,000 points is essentially equal as it is running into the limits of the testing procedure, so don't base your purchase decisions on anything above that number.
 
Last edited:
Again! LAN-to-WAN and WAN-to-LAN use the router's NAT and other routing functions. The speed of the router's CPU matters, and the size of the buffer memory. But any decent brand router newer than about 5 years can keep up with WAN speeds from the ISP DSL/Cable modem.

LAN-to-LAN is merely the switch. Does not rely on the router's CPU.

And they lived happily ever-after.
 
So LAN to WAN is upload speed?

What about WAN to LAN?

When I look at the benchmark, I should look at the WAN to LAN numbers, LAN to WAN numbers, Total Simultaneous numbers where the higher the numbers the better it is?
Yes, LAN to WAN is upload.

WAN to LAN is download.

You should look at both numbers and higher is better. Your ISP will let you know what your upload and download should be and these should match your numbers in testing.
 

Latest threads

Support SNBForums w/ Amazon

If you'd like to support SNBForums, just use this link and buy anything on Amazon. Thanks!

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top