Asuswrt-Merlin 3.0.0.4.374.33 is out

[bluepoint]

There is a 4 channel "break" between channels in 5ghz (so APs in the same area should be able to use adjacent channels, unlike 2.4ghz) so there actually isn't a channel 159, maybe it's using that to tell you it's taking all the airspace between 157:161 = 40 MHz, like how 40 MHz channels on 2.4ghz will sometimes identify as the median channel that they are occupying. I dunno I will have to double check mine.

I just thought to report it but if my memory serves me right, when in SDK5 it reports equals to inSSIDer. Anyone using the SDK5 version please check if that's the case?

 

[hungarianhc]

Merlin, does this work on the 68U? I'd love to give it a test this evening if it does!

 

[RMerlin]

Merlin, does this work on the 68U? I'd love to give it a test this evening if it does!

There's no build for the RT-AC68U yet, as I don't have a router to test things on and ajust parts of the code that will require so (such as Stealth mode).

 

[RMerlin]

inSSIDer and the N66U's GUI(hover over wireless icon) does not agree with the channels being used in 5GHz band set for auto channel. The GUI reports I'm using 157+ 159 while inSSIDer says 157+161. I don't know which to believe.

Note: Speed is still great even with the confusion.

What does it show on the Wireless Log page?

I'm not 100% sure that the icon report is accurate, because I don't fully trust InSSIDer either. I need to find a third method to validate which channels are really used - Asus's own utility for the USB-AC53 might help me when I find some time to look again at this.

 

[bluepoint]

What does it show on the Wireless Log page?

I'm not 100% sure that the icon report is accurate, because I don't fully trust InSSIDer either. I need to find a third method to validate which channels are really used - Asus's own utility for the USB-AC53 might help me when I find some time to look again at this.

The wireless log agrees with the icon. I'll be waiting for your test result with the AC53. Thanks.

RSSI: 0 dBm	SNR: 0 dB	noise: -84 dBm	Channel: 157l
BSSID: 30:85:A9:xx:xx:xx	Capability: ESS 
Supported Rates: [ 6(b) 9 12(b) 18 24(b) 36 48 54 ]
HT Capable:
	Chanspec: 5GHz channel 159 40MHz (0xd89f)
	Primary channel: 157

 

[speedingcheetah]

What does it show on the Wireless Log page?

I'm not 100% sure that the icon report is accurate, because I don't fully trust InSSIDer either. I need to find a third method to validate which channels are really used - Asus's own utility for the USB-AC53 might help me when I find some time to look again at this.

I use WiFi Analyzer app for my phone. There is also WirelessMon I use on Windows......although...so far..I have noticed that they report network as N and not AC...guess they haven't updated yet. even inssider says its N.

The Intel wifi application shows AC networks...but I hate Intel's ProSet software

 

[speedingcheetah]

Stp

Is it better for overall performance to have Spanning tree protocol enabled or disabled in the Switch control tab?

 

[PrivateJoker]

My RT-N66U (.32MW) gives conflicting info (I think??) for channel usage under wireless log.

Under 5ghz SSID
1st line: channel 161u (this means 157 & 161, aka 161u, aka 157l, aka 5785mhz & 5805mhz)
5th line: channel 159 40mhz (0x1e9f)
6th line: control channel 161

The 5th line I think being erroneous info.

OSX identifies it as "161, -1" (aka 161 & 157) when I go into detailed NIC properties

Also, different countries have different DFS/TPC regulations on different channels. There are many theoretical ways you could pick a channel manually and have it not actually be able to pair up with the channel it "should" or perhaps it does, but at reduced power. So always compare your RSSIs and actual throughput (not just link speed) on the channel you may choose manually & an automatic selection as well. Theoretically set on "auto" could even allow your AP to be the first one to move to a better frequency should the one it's on get interference, or DFS activity that wasn't there before, or DECT phones, etc.

 

[PrivateJoker]

Thank you for great work, Merlin. I would really appreciate some advice.

I am trying to setup a wireless network bridge between two adjacent buildings. I have two RT-AC66U routers with 374.32 firmware, one configured as AP and the other as a Media Bridge. The AP is connected to a gigabit network and the media bridge is connected to a laptop via a gigabit nic. I am measuring connection speed across the bridge by copying an ISO from a network share. When the two routers are 10 feet apart, I am only using 12.5% of gigabit connection. When the laptop is directly connected to the network the copying task consumes 35% of the gigabit connection. I have tried changing the control channel and 20/40/80 bandwidth frequencies with no improvement. The AP point's power output is set at 200mW. Would switching to 374.33 make a difference? What can I do to improve performance? How much bandwidth should I realistically expect at ~125 foot distance?

I'd start a new thread, this isn't really a firmware question, this is a AP question. These APs are not designed to be used for linking buildings wirelessly. Their antennas are omni-directional and arrayed, and they also aren't designed to run day in & day out with Tx power of 200mw. Look at Engenius & Ubiquiti for starters. It would be almost negligent to speculate on the bandwidth you should "expect" over a given distance because there are so many factors in play. At a bare minimum, anticipate on mounting some sort of equipment outside, ideally in an elevated location with direct line of sight to the other location.

 

[RMerlin]

Is it better for overall performance to have Spanning tree protocol enabled or disabled in the Switch control tab?

STP should always be kept enabled, unless you have a specific reason for needing to disable it.

 

[savagehenry]

STP should always be kept enabled, unless you have a specific reason for needing to disable it.

Personally, The only specific reason I need is not having a layer 2 loop in my network but that's just me and my OCD.

 

[Jurassic]

Usb

I'm very disappointed with USB speed in this build. Usually on old firmwares the speed was near 14-15 MB/s and now it's never more than 8 MB/s and when downloading smth over Ethernet it falls down to 2 MB/s

I read about priority changes by Asus but priority of 0 doesn't differ from too low anyway. How can it be fixed without restoring previous firmware?

Thanks.

 

[DrTeeth]

Their antennas are omni-directional and arrayed, and they also aren't designed to run day in & day out with Tx power of 200mw.

They are as that is the legal output of some jurisdictions.

 

[diyguy]

Hi,

I just upgraded and I having a major problem. My connection to the router times out periodically, enough to a point that it will drop an SSH session. Has anyone experienced this before ? Is there a fix ?

 

[RMerlin]

I'm very disappointed with USB speed in this build. Usually on old firmwares the speed was near 14-15 MB/s and now it's never more than 8 MB/s and when downloading smth over Ethernet it falls down to 2 MB/s

I read about priority changes by Asus but priority of 0 doesn't differ from too low anyway. How can it be fixed without restoring previous firmware?

Thanks.

I was getting very bad performance with the lower priority. Setting it back to 0 gave me the same performance I was having with older firmwares where it was set too high - around 15 MB/s.

At 0, it's a balanced approach. If your router isn't too busy doing other things, then you will get full USB performance. If however you are running a torrent client, the two will have to share performance.

Since the device is primarily a router, I don't want to see routing performance affected by USB, hence I don't want to set it back to the highest priority.

If for some reason you think this is a problem, use the "renice" command to give smbd a higher priority (renice -19)

 

[RMerlin]

Hi,

I just upgraded and I having a major problem. My connection to the router times out periodically, enough to a point that it will drop an SSH session. Has anyone experienced this before ? Is there a fix ?

We'll need more information to be able to even start helping you.

- What router model?
- Are you connected wifi or wired?
- Are you talking about an SSH session to the router, a LAN server, or an external server?

 

[diyguy]

We'll need more information to be able to even start helping you.

- What router model?
- Are you connected wifi or wired?
- Are you talking about an SSH session to the router, a LAN server, or an external server?

n66u

wired and wifi- pings to the router fail periodically and it is consistent on different machines

ssh and gui session to router time out

 

[PrivateJoker]

They are as that is the legal output of some jurisdictions.

And 1 watt is the legal output on some of the higher 5ghz channels in the US (and 4 watts in the UK I believe). If you think you're getting the best throughput with client devices, and aren't being affected by TPC or DFS regulations (I don't know if you'd be made aware of it anyway), and your router operates in a temperature zone that you're comfortable with - go sick.

 

[speedingcheetah]

Tx power adjustment

I was wondering if the Tx power adjustment actually does anything...for 5ghz band.

On my AC56U...i have tested it...and there no difference in signal strength according to wifi analyzer app on S3...or insidder on my laptop...no matter what setting is used.....I am testing about 20-30feet away....no difference when Tx power adjustment is set to 200....or to 1. I am using control channel 36.

 

[PrivateJoker]

I was wondering if the Tx power adjustment actually does anything...for 5ghz band.

On my AC56U...i have tested it...and there no difference in signal strength according to wifi analyzer app on S3...or insidder on my laptop...no matter what setting is used.....I am testing about 20-30feet away....no difference when Tx power adjustment is set to 200....or to 1. I am using control channel 36.

I honestly doubt it, both based on RSSI from a client and actual real power being emitted.

I'm not sure if you manually picked channel 36 or it did. But, Tx power, channel selection, and channel width have a weird and often fluid relationship between them though. It might see DFS activity and limit your power or change your channel, or both, but also the things it's looking out for are often frequency hopping by design and intermittent in nature. So things it's asking your router to do tonight might not be present tomorrow.

If you're going to manually pick a channel, it just might not accept the Tx power and/or the channel width you select because of other RF stuff it sees and you don't. Different things happen any way you slice it, but I think if you really wanted to insure you were, using the widest possible channels, for example, you'd want to have channel selection on auto.

5ghz is relatively new to consumer WiFi, and it's a nice chunk of spectrum (5180-5825mhz! in the US vs 2412-2484 on 2.4ghz) but a big part of why Wi-Fi got it was automatic, unavoidable, (yet dynamic and relatively seamless) interoperability with nearby other transmitters (mostly weather radar and aviation use radar [which may actually be weather radar also, I can't tell if it's used for planes or clouds]). Different channels have different regulations in different parts of the world, and even if a channel is under DFS regulation, for example, that doesn't mean it is automatically imposed upon your router.

http://www.summitdata.com/blog/?p=752
http://en.wikipedia.org/wiki/List_of_WLAN_channels
http://www.wi-fi.org/knowledge-center/glossary/8
http://www.cisco.com/en/US/prod/col...aecd801c4a88_ps5279_Products_White_Paper.html