[Beta] Asuswrt-Merlin 384.16 beta (and 384.13_5) are available

[RejZoR]

Is it normal that 3rd core on RT-AX58U with Beta2 FW doesn't process anything? It's always at 0% where first two cores keep jumping up and down all the time.

 

[maxbraketorque]

Don't put too much stock in the DSLReports test. It's not 100% accurate. The fact that current networks in general are getting more saturated these days with the amount of people working from home WILL skew test results. And there's been quite a few reports for the past year or so about the DSLReports test results becoming increasingly inaccurate, probably due to an increase in usage.

I ran my vacation home AC86U through the DSL Reports speed test just before (running 384.15) and just after upgrading to 384.16b2, and there was no significant difference in test results. Both tests showed substantial bufferbloat on the inbound 100 mbps connection and almost no bufferbloat on the outbound 10 mbps connection. My impression of the indicated bufferbloat is that ISPs are controlling the ramp to full speed on the inbound and/or outbound connections.

 

[GHammer]

Wow, you got to be sh*ing me. My ISP apparently has my old MAC address locked in their system and this was the reason why it just flat out refused to work. Copied my old router MAC into WAN settings and the thing miraculously started working.

Don't know who you have, but on Comcast I've always had to power cycle the cable modem then add a new router.

 

[RMerlin]

Had RT-AX58U hooked and it refuses to work. Just switched it for RT-AC87U and bam, I could post this post instantly. I just don't get it what the hell is going on. It's not like I can even adjust anything, it just straight doesn't work.

When using a cablemodem, you need to power down the modem for about 10 minutes to release your lease at the ISP level, otherwise it will refuse to provide a lease to the router's new MAC address.

This is well known with DOCSIS.

 

[RejZoR]

I'm not on cable. I'm on fiber optics. I did replace the crappy stock Innbox router ISP gave me with ASUS router. Could be they use same system for some reason. Will try to leave it off for same amount of time and see if it's the same...

 

[Gar]

When using a cablemodem, you need to power down the modem for about 10 minutes to release your lease at the ISP level, otherwise it will refuse to provide a lease to the router's new MAC address.

This is well known with DOCSIS.

This technique worked for me with cable. However, when I changed to fiber there is no way to substitute a new router without calling the ISP to release the MAC address, OR, edit the MAC field to use the same one the ISP already has. Frustrating, the ISP says that's just the way it is set up.

 

[RMerlin]

This technique worked for me with cable. However, when I changed to fiber there is no way to substitute a new router without calling the ISP to release the MAC address, OR, edit the MAC field to use the same one the ISP already has. Frustrating, the ISP says that's just the way it is set up.

Could be that with fiber, it's the only way they can authenticate the customer, while with cablemodems, they have to whitelist the modem's own MAC in their database.

In that case, MAC cloning is indeed one solution, if you can't get the ISP to make the change on their end.

 

[Makaveli]

Don't think I've seen this issue on fiber. But I guess each ISP will be different.

However on my setup I have the GPON module pulled out of the isp provided modem and going directly into a Media converter then lan cable to Asus Router.

Just pulling the power on the media converter is enough for me to get a new IP address.

 

[unsynaps]

When using a cablemodem, you need to power down the modem for about 10 minutes to release your lease at the ISP level, otherwise it will refuse to provide a lease to the router's new MAC address.

This is well known with DOCSIS.

Had to do this with FIOS as well. Or yank the power on the ONT for a bit.

 

[bluepoint]

I'm not on cable. I'm on fiber optics. I did replace the crappy stock Innbox router ISP gave me with ASUS router. Could be they use same system for some reason. Will try to leave it off for same amount of time and see if it's the same...

Usually, the router provided by your ISP has a section to release/renew the IP, at least FIOS's does, so before changing router release the IP so that the ISP's DHCP server can issue another.

Ooops I just re-read your OP, in this case release the IP from your old 87U before replacing it with your new one.

 

[Gar]

Could be that with fiber, it's the only way they can authenticate the customer, while with cablemodems, they have to whitelist the modem's own MAC in their database.

In that case, MAC cloning is indeed one solution, if you can't get the ISP to make the change on their end.

I've found that few "tech support" folks at my ISP know how to do it which results in a call-back from someone that does, so I clone instead.

 

[whitey019]

I ran my vacation home AC86U through the DSL Reports speed test just before (running 384.15) and just after upgrading to 384.16b2, and there was no significant difference in test results. Both tests showed substantial bufferbloat on the inbound 100 mbps connection and almost no bufferbloat on the outbound 10 mbps connection. My impression of the indicated bufferbloat is that ISPs are controlling the ramp to full speed on the inbound and/or outbound connections.

What browser are you using? I have seen Chrome cause significantly more buffer bloat than Firefox.

 

[digits n bits]

Is it normal that 3rd core on RT-AX58U with Beta2 FW doesn't process anything? It's always at 0% where first two cores keep jumping up and down all the time.

My RT-AX58U does the same thing. The third core of the CPU rarely is being used at all. Stock firmware or Merlin it shows the same process.

 

[Ozymandias]

My RT-AX58U does the same thing. The third core of the CPU rarely is being used at all. Stock firmware or Merlin it shows the same process.

The fourth core of my RT-Ax56U is rarely (never noticed it being used) either.

 

[Torson]

It looks like your 3rd core on the AX58U does pretty much what the 4th core on the AX56U does - nothing perceptible to the eye. While writing this I just realized I don't even know what color on the graph the 4th core is shown as. Probably a new algorithm based on the 'just in case' concept...

 

[axaro1]

It looks like your 3rd core on the AX58U does pretty much what the 4th core on the AX56U does - nothing perceptible to the eye. While writing this I just realized I don't even know what color on the graph the 4th core is shown as. Probably a new algorithm based on the 'just in case' concept...

I can confirm the same behaviour on my AX56U, run a speedtest if you are curious about Core 4's color

 

[maxbraketorque]

What browser are you using? I have seen Chrome cause significantly more buffer bloat than Firefox.

Chrome. I'll try with Firefox next weekend.

 

[zenmota]

Using a RT-AC86U router with the 384.16 Beta 2 firmware. I notice this when messing around with the QoS.
Go to Adaptive QoS
Then QoS
Under Wan Packet Overhead Select Cable (DOCSIS)
Then Click Apply
After it does it's thing of Applying the settings, Under Wan Packet Overhead will revert back to Select Preset: (The value to the right will stay at 18 though).
I just starting messing around with QoS so I don't know if this was happening with any earlier versions.

Thanks for the time....
I

 

[L&LD]

@zenmota this is normal.

 

[RejZoR]

Something is wrong with IPTV thing on RT-AX58U with Beta2. It just doesn't work. I've used same settings as on RT-AC87U, but on this one TV works, on RT-AX58U it doesn't. I'm using Port 4 and I need to use VID value 2000 on it.

In menu it says IGMP Proxy DISABLED because it can't exist with IPTV (then why is it even there then?) due to hardware limitation, but on RT-AC87U TV works even if I manually disable IGMP Proxy so what gives?!