RT-AC68U in Media Bridge mode won't go faster than 288.9 Mbps. Main router = AX-88U. RSSI @ AC-68U is a strong -35dBm.

[theirongiant]

I have an RT-AX88U in the main office and an RT-AC68U / RT-AC1900P serving as a media bridge about 15 feet away, through one layer of drywall in a wood frame house. It's in a hallway broom closet where I added an electrical outlet to convert it into a server closet. I added active ventilation with an AC Infinity Airframe T7 installed near the top of the door, and there is a large gap at the bottom of the door for intake. Daytime temperature in the closet is about 80-82ºF with all equipment running. The RT-AC1900P serves as a media bridge to my smart thing hubs and a NAS; everything in the closet is wired up to a Netgear fanless switch. No PoE.

Both the AX-88U and AC-1900P are mounted on shelves about 6 feet off the ground.

The current signal strength and connection speed, as measured by the 1900P, is displayed in the router's dashboard:

Parent AP status :
Connected
Band: 5 GHz
Link rate: 288.9 Mbps
RSSI: -35 dBm

I'm confused as to why the RT-AC1900P won't transmit any faster than 300Mbps when it is capable of 1300 Mbps. I would have expected at least 500-600 Mbps at this distance. Anything "louder" than -50 dBm is basically excellent, so -35 dBm should be able to deliver maximum speed.

Is there a setting I'm missing on either device? Both are running the latest Merlin firmware (386.3_2).

 

[Adooni]

did you disabled Enable NAT in WAN - Internet Connection

what is CPU load at your AC1900P when you transmit data.
for test purpose can you connect those routers with cable? and recheck speed. It is looks it is not wi-fi limitation but router performance more.

do not look at theoretical max speed this is marketing talk. Asus and rest is doing it from 2009 at least when they introduce Asus RT-N16 - it has LAN 1Gbit but it is hard to get 300Mbit from it w/o anything enabled in router.

 

[L&LD]

Have you tested different Control Channels? 'Spec great' doesn't mean optimal real-world results in many examples/situations.

Have you tested different router orientations? Antennae placement? Inches may make big differences.

Test also with different heights for the main and Media Bridge routers, as much as possible.

 

[wouterv]

It shall be capable up to 1300 Mbps.
I have a RT-AC68U as main Router and a RT-AC1900U as Media Bridge, about 7 meters apart with one brick wall in between:

It goes a bit up and down over time, doesn't go below say 800 Mbps, sometimes up to the max of 1300 Mbps.
The Media Bridge setup is pretty straight forward, so is the Router setup.
The only "special" in the Media Bridge is the manual LAN setup, as follows:

The thing for you is that you have AX and AC mixed.

 

[theirongiant]

Update: I swapped to a different make/model...

I had a Netgear R7000 lying around that I hadn't used in a while.

The Asus RT-AC68U and Netgear R7000 are very similar under the hood:

• Dual Band 2.4GHz & 5GHz
• 802.11n @ 600Mbps, 802.11ac @ 1300Mbps
• Dual Core ARM Cortex A9 CPU @ 1GHz
• 128MB flash memory
• 256MB RAM

The R7000 had an old build of AdvancedTomato installed. The fork was an outgrowth of TomatoUSB by Shibby, but AT is effectively dead having had no updates since 2017. Thankfully, someone else forked Shibby to a new project in 2018: FreshTomato.org.

I flashed the R7000 back to stock, and then upgraded to the FreshTomato build.

I installed the R7000 in the same physical location as the Asus. While the reported RSSI is a bit lower -- down to -45 dBm from -33dBm -- it doesn't seem to matter much.

I just did a couple of quick tests from speedtest.net.

From a computer attached to the AC68U in media bridge mode, a typical result was about 60Mbps down, 65Mbps up.

Using that *same* computer attached to the R7000 in wireless ethernet bridge mode, a typical result is now about 500Mbps down, 540Mbps up.

That's a performance increase of over 800%.

What could account for that massive difference?

MAC Address	xx:xx:xx:xx:xx:xx
Wireless Mode	Wireless Ethernet Bridge
Wireless Network Mode	Auto
Interface Status	Up (LAN0)
Radio	Enabled
SSID	[redacted]
Security	WPA2 Personal (PSK) + AES
Channel	36 - 5.180 GHz
Channel Width	80 MHz
Interference Level	Acceptable
Rate	1300 Mbps
RSSI	-45 dBm
Noise	-92 dBm
Signal Quality	100

 

[ColinTaylor]

Link rate: 288.9 Mbps

The only scenario that would create this link rate is 3 streams operating at 20MHz bandwidth. The same three streams operating at 80MHz bandwidth yields 1300Mbps. So the question is why was your Asus only operating at 20MHz.

 

[wouterv]

Check the Wireless log for the bandwidth:

80 MHz in my case.

Lower data rates are usually caused by one or more of the following:

• Neighbor disturbance (other networks using the same channel).
• Steel or concrete near by the Router or Media Bridge (steel shelves?) or in between the devices.
• Electrical cabling and wiring in close proximity of the antenna's.
• Distance.

The Router and Media Bridge shall ideally have a free line of sight to each other.
Try temporary moving them out of the closet and experiment with antenna orientations.

 

[theirongiant]

Check the Wireless log for the bandwidth:
View attachment 35972
80 MHz in my case.

Lower data rates are usually caused by one or more of the following:

• Neighbor disturbance (other networks using the same channel).
• Steel or concrete near by the Router or Media Bridge (steel shelves?) or in between the devices.
• Electrical cabling and wiring in close proximity of the antenna's.
• Distance.

The Router and Media Bridge shall ideally have a free line of sight to each other.
Try temporary moving them out of the closet and experiment with antenna orientations.

That wouldn’t explain an 8x improvement in performance. The routers are in the same position. The Netgear has a weaker RSSI than the Asus but 8x the speed. The antennae are similar in length on both devices and similarly oriented.

As the previous commenter indicated, it seems the Asus was limited to 20MHz. I don’t know why.

 

[wouterv]

That wouldn’t explain an 8x improvement in performance. The routers are in the same position. The Netgear has a weaker RSSI than the Asus but 8x the speed. The antennae are similar in length on both devices and similarly oriented.

As the previous commenter indicated, it seems the Asus was limited to 20MHz. I don’t know why.

Was the Media Bridge box brought to factory default, then set to Media Bridge?

 

[theirongiant]

Was the Media Bridge box brought to factory default, then set to Media Bridge?

Yes.

 

[wouterv]

Did you check the Wireless Log of the Router, did it indicate 20 MHz for 5 GHz?
You can try to force the Router to 80 Mhz, in the Wireless - 5 GHz settings.
It can be other wireless clients as well to scale down the Router to 20 MHz, or neighbor disturbance.

 

[theirongiant]

So I did a full factory reset of the Asus RT-AC68U and then configured it for Media Bridge from the setup wizard.

Now it shows 5GHz, RSSI -35 dBm, and 1300MHz. However, the results from www.speedtest.net did not improve much. 80Mbps down, 59Mbps up.

On the plus side, my smart hubs work again. This is not a great trade-off. I'd rather not have to keep two routers in the closet to support everything.

For now, I'm connecting the smart hubs at the main router and leaving everything else connected to the R7000. The increased speed is worth the minor inconvenience / workaround (until I find a better solution).

Long term, I need to run Ethernet throughout the house, but it's difficult moving around in our attic because there's no "floor" space. It takes a lot of crawling and lumbering over roof joists (no pun intended).

 

[RangerZ]

15 feet away, through one layer of drywall in a wood frame house

In a scenario like this you can take a long drill bit and just drilled right through the wall to the next room and fish a cable. If you want pretty just add a pair of Quickport face plates and a short run of cable. Add a switch if you need more ports. Easy, cheap and FAST.

 

[theirongiant]

Thanks for the tip!

 

[theirongiant]

So... does anyone else have any thoughts about why the wireless bridge throughput would be so slow (50-70Mbps) compared to the R7000 (600Mbps+) despite having similar signal strength, similar placement, and similar specs between the two devices?

Both routers have a 1 GHz dual core A7 Cortex ARM CPU, 128MB Flash, 256MB RAM, and a dual-band 3 antennae configuration (802.11ac @ 1300 + 802.11n @ 600). The Asus AC68U has a Broadcom BCM4708A0 @ 800 MHz, and the Netgear has a Broadcom BCM4709A0 @ 1 GHz.

Those two chipset parts are literally one number away in the parts catalog. I can't believe that a 200MHz difference would translate to a 10-fold difference in speed. There has to be a bug in the software.

edit: to clarify, mine is actually the AC1900P, which should technically be FASTER than the R7000, yet the throughput is much slower. I'm using the AC68U firmware per Merlin's instructions for the AC1900P.

This post lays out the difference. https://www.snbforums.com/threads/asus-ac1900-router-vs-rt68u.35851/

 

[wouterv]

A few thoughts:

• The performance of all models from the RT-AC68 family should be about the same (I suppose as consumer it will be hard to find a real difference).
• Why talk about an RT-AC68U while it is a RT-AC1900P which was a Best Buy special?
• Why not be sure and load the current latest RT-AC1900P stock firmware?
• What Tx rate and Rx rate does the Router show for the RT-AC1900P versus the R7000 as Media Bridge (System Log > Wireless Log)?
• What does the Media Bridge (RT-AC1900P versus the R7000) indicate for LAN Port speed with the test computer (Network Map > Status)?

 

[theirongiant]

I installed FreshTomato on the AC68U AC1900P and saw a massive throughput increase to about 600Mbps in both directions. Same as the Netgear R7000 running FreshTomato.

So now I'm going to revert to the stock firmware to see if the speeds are consistent. If they are fast on stock firmware, then there is a bug (or additional optimization required) in asuswrt-merlin. If it's slow with stock firmware, then it's Asus' problem.

- I was talking about the device in the context of which firmware was recommended by Merlin (it says to use the AC68U if you have an AC1900P.
- I am about to wipe / revert to the stock firmware to see if this is an issue with Merlin's port.
- The Tx and Rx rates were similar from both sides: 1300 Mbps.
- LAN port speed was 1Gbps in both cases.

 

[RangerZ]

I just setup a Media Bridge setup (a pair of ASUS RT-AC1900P on Merlin 386.3_2). I set the channel to 80 Mhz. I am doing marginally better, but at 3 feet. Connected at 1300 Mbps, -26db. Getting 130-160 Write\190-240 Read depending on what I use to test.

I also tested with the devices in an AIMesh configuration and get virtually identical performance.

FWIW, I use the Totusoft lan speed test to measure speed internally. I prefer to eliminate the internet when testing LAN speed and I only have a 200\50 connection. I use the older 1.3 lite (right hand column) just because it's portable.

 

[theirongiant]

Update: finally reverted the AC1900P to STOCK Asus firmware.

The media bridge performance is no better. FreshTomato has at least 5x faster throughput.

@RMerlin, to bring you up to speed: a third-party firmware project forked from TomatoUSB has somehow managed to get significantly better 5GHz throughput between access points. I've tested two different Broadcom-based APs with Cortex A7 dual core CPUs at similar speeds. Both are configured in "media bridge" or "wireless ethernet bridge" mode. The stated data rate is 1300Mbps. RSSI is between -35 and -45 dBm. The access points are placed in adjacent rooms, ~15 feet total distance, including one layer of drywall in a wood frame house.

A Netgear R7000 running FreshTomato firmware achieves 600-700Mbps bidirectional throughput with an AX88U as the base AP.

An AC1900P running FreshTomato achieves the same level of performance. The same AC1900P running Asus stock or Asuswrt-merlin sees much lower throughput, ranging from 60-150Mbps.

Any thoughts about filing a bug or performance enhancement request with Asus? Any other tests you'd like me to perform? Is this a known issue?

Thanks!

 

[L&LD]

To file a 'bug report' to Asus, you need to install the stock firmware and submit your findings via the GUI.