AC, who cares? I am KEEPING my Asus RT-N66R Router :-)

[Klueless]

I believe this is possible with the Android app. https://play.google.com/store/apps/details?id=com.asus.aihome

Thank you very much!

 

[Klueless]

Thank you!

I kinda have something like this as well but still not so many. I am curious about your scenario.

For example I have two guest networks. All have passwords btw - in my country, open networks are jumped on faster than <imagine I said something funny here>. 1 Guest network has basically the same access as my main network and the other has access only to the internet. The first one was created to be given to trusted guests or for guests who need access to my local network. The difference between the first guest network and my main is I can switch out the password or take down that network without affecting my truly trusted local users.

I would be very interested in your reasoning for having more than 2. How is it configured? I am merely curious btw. Maybe it's something I could or should be doing on my end as well.

My two guest nets are very similar to yours.

"BYOD" is for employees who bring their own devices in for work related activities. It is password protected.

"Guest" is for customers and visitors. It is "open". (We're in a rural area so interlopers aren't currently a problem.)

The other two are for company owned machines. Originally I has one SSID for both radios but devices kept jumping onto the "wrong" radio so I had to split it into two SSIDs, one for each radio. I had desktops 3 meters away from the router that kept dropping down to the 2.4Ghz radio and I had devices in the next building that were getting crappy performance on the 5Ghz radio.

The nearby desktops also use our internal servers. Locking them at 5Ghz stabilized their performance.

I later ran performance benches to confirm that I had made the right mistake. I guess sometimes ya just gotta to what ya gotta do.

That's a simple one. You simply set your DHCP to avoid handing out IPs that could conflict with your statics (assuming you can't change them). I have a mix of statics and DHCP addresses on my home network. In tomato it's labelled "IP Range" and in asuswrt it's labelled "IP Pool Starting Address" and "IP Pool Ending Address". You can even do DHCP reservation which means that clients of your choosing will always get the same IP, even if using DHCP.

Thanks. Our original ISP provisioned a modem only and our router was set to do exactly as you suggested. Our new (two years now) ISP provided a modem/router so we cascaded our old router off of their router.

Recently we had a month and a half of non-stop trouble (much better over the past month) so double-NAT came up. My lack of knowledge and legacy address space are keeping me from moving forward but, thankfully, the perceived need has lessened now that we're running ok again.

 

[KevTech]

It has been shown that an AC router can improve N performance.

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance

 

[Klueless]

It has been shown that an AC router can improve N performance.
http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance

Thank you. You did it. You swayed me!

I even ran my own set of tests. Wee hours of the morning. Just me and the security cameras. I calibrated by running PC to PC over Gigabit Ethernet through my N router. 67 Mbps. Symmetrical, send and receive were virtually the same. Unless I have something seriously screwed up the limiting factor must be my antiquated (we hear Windows 8 is coming out : ) low end PCs?

I left one PC on Ethernet and used the other PC to walk through the various wireless modes.

• N @ 5Ghz
  • 30 feet - 66 mbps (Ethernet speed for me)
  • 60 feet, in the garage (on the other side of a fire grade door and wall) - 30 Mbps
  • 100 plus feet, in the bldg. next door w/ window to window line-of-site - 30 Mbps
• N @ 2.4Ghz
  • 30 feet - 44 Mbps
  • 60 feet (in the garage again) - 28 Mbps
  • 100 plus feet (next door again) - 28 Mbps
• N @ 2.4Ghz
  • 100 plus feet (next door again) - 20 Mbps

The last test was to see how much my N range extender degraded performance. Hmm ... about a 25% hit. But, save for the office by that window, we had virtually no connectivity to/from that house next door thus virtually no connectivity to the Internet. The range extender allows the rest of the PCs in bldg. 2 to get acceptable performance.

Actually stellar performance. Internet is expensive where I am. We only have 15 Mbps x 1.5 Mbps. Internet speed tests show us hitting our subscribed service on all PCs, even the ones buried in back of bldg. 2

We run our local client/server apps only in the main bldg. All that stuff is on Ethernet or 5Ghz so, I think, we're as good as we're gonna get until we upgrade PCs.

What I learned from you is that I've opportunity to significantly improve performance between the two buildings by running N over an AC router. What I don't have is a business case : (

Thanks again!

 

[dlandiss]

I calibrated by running PC to PC over Gigabit Ethernet through my N router. 67 mbps.

Hoping that is really 67 Mbps

 

[Klueless]

Hoping that is really 67 Mbps

Number sounds suspect to you? Too high? Too low?

Tool reported 8,400 K Bytes per Sec. and I converted to mbps because it's more meaningful to me. Did I calculate wrong? (Possibly a 2% error because I never know whether to use 1,000 or 1,024 in my math so ... I just use 1,000 because it's easier : )

Thank you.

 

[dlandiss]

No, I was just commenting on the use of "m" (0.001 prefix) where "M" (1000 prefix) would be more appropriate. But what's a difference of a million or so<G>?

 

[Klueless]

No, I was just commenting on the use of "m" (0.001 prefix) where "M" (1000 prefix) would be more appropriate. But what's a difference of a million or so<G>?

<LOL> I gets it now ... you scared me!

I'm not sure where I picked it up, local vernacular maybe, but they use MBps to signify MegaBytes per sec vs. mbps to signify megabits per second around here. Sorry for the confusion and thanks for straightening me out.

Sincerely,

Klueless : - )

 

[dlandiss]

It is often not a big deal, and I noted that you already used B for Bytes and b for bits. As an engineer I am especially sensitive to naming and labeling errors, so I also use M for Mega and m for milli. But hey! One is only a billion times bigger than the other<G>.

 

[Nullity]

I just spell it out (Mbyte Mbit). I love when acronyms and initialisms are less efficient than the original phrase...

 

[Klueless]

As an engineer ...

Ha! Now that I know you are an engineer ... Does getting only 67 Mbps over Ethernet sound reasonable for two lowish end PCs?

 

[dlandiss]

Ha! Now that I know you are an engineer ... Does getting only 67 Mbps over Ethernet sound reasonable for two lowish end PCs?

By "over Ethernet" do you mean between the two PCs? That sounds a bit slow unless you have some glitches in the physical Ethernet cabling. Antivirus software can also affect the speed since it will be examining all the incoming files. I sometimes see >90Mbps over 100BASE-T and about 900Mbps with 1000BASE-T.

It can vary a lot with the speed of the HDs sending and receiving the content, also. When I try to send a large file to my wife's computer it starts at about 900Mbps, but after a few seconds when her incoming memory buffer fills up that degrades to about 300Mbps.

It will definitely be slower if you are transferring a lot of small files instead of one big one. Big overhead with small files.

 

[Klueless]

Thank you. That was helpful.

I was going PC to PC. At first I panicked because I plugged them both into Gigabit Ethernet ports. I settled down some when I realized my low end PCs were only 100 Mbps Ethernet. I did some reading that suggested 2/3 of wire speed was reasonable due to overhead.

I'm using something called NetStress to send noise back and forth. I suspect they burst an endless series of short files which would match with what you said.

 

[Klueless]

It has been shown that an AC router can improve N performance.

http://www.smallnetbuilder.com/wire...oes-an-ac-router-improve-n-device-performance

Hey KevTech!

I had to buy a new range extender. Remembering your comment/post I paid a couple extra bucks and got an AC1200. What the heck, if N runs faster on an AC router maybe the same would be true with an AC range extender? Sure enough, I got a crazy inexplicable boost in performance.

• I placed my PC in an optimum location in bldg. 2 and connected to the router in bldg. 1. 38 Mbps.
• I then placed my new range extender in approximately the same location and connected it to the same N router in bldg. 1 using the same 5 Ghz radio and the same SSID.
• I then connected my PC wirelessly to the new range extender. I expected something more that 19 Mbps (half of the 38) but something less than 38 Mbps. What I got was a phenomenal 50 Mbps.

I'm thinking I owe you a big thank you? (But it still confuses the heck out of me!)