Beta Asuswrt-Merlin 386.1 Beta (stage 2) is now available

[DanNuke86]

Have carried out a dirty flash to beta4 from the previous beta4 build Had an issue with my AX86U where I wouldn’t get full speeds over wifi. Used to get full speeds on my AX88U. Seems to have been like this since I bought it.

Reading through a lot of the comments on various threads, it seems the microtrend is an issue. I’ve done WAN speeds to check and then carried out WiFi speed tests, I’ve withdrawn consent and then checked WiFi speeds and found them back to full speed again.

Put QoS back on and found WiFi speeds not matching manual D/L & U/L set speeds (using autoBW). Turned QoS back off again (but privacy hasn’t been withdrawn this time) and speeds back to full over WiFi again?

I don’t get why this is happening. Can anyone tell me if this is an issue with the QoS or trend micro, or current firmware?

 

[PDinDetroit]

2021-01-10 Upgrade Notes

Pre-Merlin BETA State:
RT-AX88U: HW Ver - A1.1, Production Year - 2019, AiMesh Router, RELEASE ASUS, 386_41249, Location - Basement (Point of Presence)
RT-AC86U: HW Ver - R1.5, Production Year - 2018, AiMesh Node, BETA ASUS, 386_41157, Location - Living Room
RT-AC86U: HW Ver - R1.5, Production Year - 2017, AiMesh, BETA ASUS, 386_41157, Location - Office
Backhaul: Ethernet, CAT 6 cables
WIFI:
2.4 GHz, Channel 6, Bandwidth 20 MHz, Beamforming Disabled
5 GHz, Channel 36, Bandwidth 20/40/80 MHz, Universal Beamforming Disabled
2.4 GHz & 5 GHz Guest Networks for IoT

Followed @LimJK upgrade path (again): https://www.snbforums.com/threads/a...e-2-is-now-available.69116/page-2#post-649590

NOTES:
- RT-AX88U requested manual boot. Took the time to restart Cable Modem - pfSense FW - NetGear Smart Switch
- 2.4 Guest Network - Nearly all clients went to RT-AX88U as AiMesh Router. Selected each Guest Network, Sync to AiMesh Node already set to All, and clicked Apply. This resolved the issue and Clients moved to the appropriate AiMesh node quicker than before.

Merlin BETA State (same roles):
RT-AX88U: 386.1_beta4b
RT-AC86U: 386.1_beta4
RT-AC86U: 386.1_beta4
Backhaul: No changes

Monitoring...

 

[miroco]

Don't mess with the switch, it will affect latency.

Isn't it better to keep energy efficient ethernet (EEE) disabled to avoid link flap errors?

Jan 10 11:04:17 kernel: eth4 (Ext switch port: 3) (Logical Port: 11) (phyId: b) Link Up at 1000 mbps full duplex
Jan 10 11:05:02 kernel: eth4 (Ext switch port: 3) (Logical Port: 11) (phyId: b) Link DOWN.
Jan 10 11:05:05 kernel: eth4 (Ext switch port: 3) (Logical Port: 11) (phyId: b) Link Up at 1000 mbps full duplex

ASUSWRT-Merlin RT-AX86U 386.1_beta4 Fri Jan  8 22:43:23 UTC 2021
admin@RT-AX86U-AB12:/tmp/home/root# pwr show
WIFI Off            Disabled
DISK Off            Disabled
PCI ASPM            N/A
UBUS DCM            N/A
CPU Off             N/A
CPU Wait            N/A
CPU Speed           N/A
XRDP Gate           N/A
NET Down            Disabled
PHY Down            Disabled
PHY EEE             Enabled
PHY APD             Enabled
SF2 DGM             Enabled
DRAM SR             N/A
AVS                 N/A
WLAN Down           N/A
PMD Off             N/A
admin@RT-AX86U-AB12:/tmp/home/root#

 

[jeff3820]

Has anyone compared temperatures and router power states between 384.19 and 386.1b4b for an AC86U? Would be interesting to see.

With beta 4: 2.4GHz 53C, 5GHz 58C, CPU 76C
With beta 4b: 2.4GHZ 53C, 5GHZ 58C, CPU 71C

 

[XIII]

RT-AC86U:

• 386.1 Beta 2: 2.4 GHz: 53°C - 5 GHz: 58°C - CPU: 73°C
• 386.1 Beta 4b: 2.4 GHz: 52°C - 5 GHz: 58°C - CPU: 75°C

Never installed Beta 3/4 and never checked on 384.19...

 

[Tom Ames]

keep us posted on what it settles to, please. it will have been heat-soaked and will take time to cool.
my AC86 is running at 71-72C in 18-19C ambient on 384.19. I can accept a few degrees increase, but would prefer to keep things on the cooler side without going to an active solution
I accept that electronics will always emit heat from work while powered, but if coding updates are supposed to optimise how they operate, shouldn't power consumption and heat drop off as that happens?

Now four hours after b4b install and it seems settled at 79-81*C. Ambient is 21*C.

 

[bbunge]

keep us posted on what it settles to, please. it will have been heat-soaked and will take time to cool.
my AC86 is running at 71-72C in 18-19C ambient on 384.19. I can accept a few degrees increase, but would prefer to keep things on the cooler side without going to an active solution
I accept that electronics will always emit heat from work while powered, but if coding updates are supposed to optimise how they operate, shouldn't power consumption and heat drop off as that happens?

Mine settled to 82C in a 19C room on 386.1 beta 4b. Went back to 384.19 and CPU is now at 69C. When I get my cooling fan I may go back to 386.1.

 

[JSinFCVA]

RT-AC86u, after flashing beta4b CPU wait enabled. temperature 77deg. Thanks,Merlin

Ditto. Thanks!

 

[heysoundude]

Now four hours after b4b install and it seems settled at 79-81*C. Ambient is 21*C.

Mine settled to 82C in a 19C room on 386.1 beta 4b. Went back to 384.19 and CPU is now at 69C. When I get my cooling fan I may go back to 386.1.

RT-AC86U:

• 386.1 Beta 2: 2.4 GHz: 53°C - 5 GHz: 58°C - CPU: 73°C
• 386.1 Beta 4b: 2.4 GHz: 52°C - 5 GHz: 58°C - CPU: 75°C

Never installed Beta 3/4 and never checked on 384.19...

With beta 4: 2.4GHz 53C, 5GHz 58C, CPU 76C
With beta 4b: 2.4GHZ 53C, 5GHZ 58C, CPU 71C

Thank you, kind folks! I can accept a possible 7-10 degree increase, but I'm still going to wait for v386.1 Release...which I expect shortly, given historical release patterns - there's always a flurry of putting out little hotspots like this before @RMerlin pushes the commit button ;-)

 

[jsbeddow]

I believe @RMerlin just said yesterday that we may be looking at a few more weeks before 386.1 goes to a final Release, given the recent issues that cropped up.

Beta - Asuswrt-Merlin 386.1 Beta (stage 2) is now available

@RMerlin -- looking around sources for various routers to check on temp issues many are discussing. One thing I note is relatively recent alterations to kernel configs: CONFIG_DEBUG_SLAB=y CONFIG_DEBUG_SLAB_LEAK=y That makes mem allocs pretty busy. Not saying it's all of it, but for beta...

www.snbforums.com

 

[heysoundude]

I believe @RMerlin just said yesterday that we may be looking at a few more weeks before 386.1 goes to a final Release, given the recent issues that cropped up.

<shrugs> time doesn't have much meaning lately on lockdown...

 

[Smokey613]

I dug out some fans I had been using to cool a cabinet and double sided taped them to the back of my main RT-AC86U, blowing in on the heatsink. I then disabled cpu wait and eee. My normal cpu temp usually stays around 77-81. This is my current temps with the above settings after 30 minutes. I may chance a test of the 386.1 beta4b, maybe.....

 

[webdriver]

Interesting. RT-AC86u, still on 386.B4 merlin (didn't try B4b)
before CPU wait turn on 81~82C
after CPU wait turn on 76~77 (-5)

Router sitting in a closet.

 

[skeal]

I dug out some fans I had been using to cool a cabinet and double sided taped them to the back of my main RT-AC86U, blowing in on the heatsink. I then disabled cpu wait and eee. My normal cpu temp usually stays around 77-81. This is my current temps with the above settings after 30 minutes. I may chance a test of the 386.1 beta4b, maybe.....

View attachment 29332View attachment 29333

I've been using a laptop cooler I bought from Amazon for 11 dollars Canadian. I get the same results with my AX88U.

 

[shabbs]

Interesting. RT-AC86u, still on 386.B4 merlin (didn't try B4b)
before CPU wait turn on 81~82C
after CPU wait turn on 76~77 (-5)

View attachment 29335

That's an expected behaviour. Mine has done the same after enabling.

 

[webdriver]

That's an expected behaviour. Mine has done the same after enabling.

Not sure if B4b will drop another several degrees LOL

 

[shabbs]

Not sure if B4b will drop another several degrees LOL

I would not expect that. I believe it's essentially doing what you just did manually.

 

[dogger66]

Speedtest 68u 386.1b4 router based speedtest maxs out at ~270Mbps...yet attached laptop gets the the best the service (750/20) provides. Anyone know why the routers internal test can't measure what it actually able to do?

 

[dave14305]

Speedtest 68u 386.1b4 router based speedtest maxs out at ~270Mbps...yet attached laptop gets the the best the service (750/20) provides. Anyone know why the routers internal test can't measure what it actually able to do?

View attachment 29336

The router test does not benefit from the NAT acceleration that LAN clients take advantage of, so your router tests are constrained by the router’s CPU, which isn’t capable of hitting 750 Mbps on its own.

 

[elorimer]

Is it possible to change EEE on B4b and set it to enabled, just to see if CPU temp drops a little more?

For the 86U, yes and no. In a terminal I've enabled EEE but the terminal would hang. After closing the terminal and reconnecting, I can see EEE enabled, but it had no effect on temps. I've put it into a script on boot and the boot was massively irregular. So, my conclusion is DO NOT DO THIS.
Also, on B4 (is there a B4b for the 86U--I'm on whatever is up now) the CPU wait fix doesn't work.
But I've put a fan on my 86U, which was 80C on .19 and 95C on B3, and it is at 60C on B4.

EDIT: I wasn't on B4b after all. With it installed, I'm still at 60C with the fan, 87C without with CPU wait enabled.