Release RT-AC86U 11th January 2021 Firmware version 3.0.0.4.386.41634

[alex1236hk]

@ASUSWRT_2020

I also forgot to mention the other day, I checked my nodes for a temperature reading, I had two nodes travelling around 86C, with the exception of one node, which was at 101C.

What have ASUS changed other than "CPU Wait", to cause such a large variation in temperatures in this public release firmware?

With these temperature increases, which may lead/cause a total failure of a router, is/are ASUS willing to replace the router (free of charge), because of a faulty firmware, which is causing these temperature increases?

Or is a router running in excess of 100c, 24/7 ok?

Please enlighten us all, as to why ASUS may think these excessive temperatures are considered ok, by previous firmware releases?

We all await your response.

You are advised to "defer" Asus' choice in the matter, like someone said on page 2.
New Feature: 100c egg frying cpu!

 

[OzarkEdge]

You are advised to "defer" Asus' choice in the matter, like someone said on page 2.
New Feature: 100c egg frying cpu!

You are misquoting my experience that I shared in this thread to help inform others. I did not advise anyone to defer to Asus' choice in the matter. Misquoting one word to make a useless wise crack about frying eggs doesn't help anyone.

You have already been told by Merlin himself that the change to the CPU wait state was made by Asus to fix a bug. If you want to whine about it, leave me out of it.

OE

 

[follower]

Is this a real temperature or just a temperature value bug? Nobody knows yet right?

 

[bitsbytes]

Is this a real temperature or just a temperature value bug? Nobody knows yet right?

I was thinking the same thing. Is it really a physical temperature increase or just a bug in reading the values??

 

[bbunge]

I was thinking the same thing. Is it really a physical temperature increase or just a bug in reading the values??

Real temperature.

Edit: Am a bit concerned this morning. Not with the temperature but with RAM usage rise overnight. RAM used went from 72% to 93% with nothing unusual traffic wise. OK, I do have a swap that the system can use but loads of others don't and I've seen at least one case in the past day or so where it appeared a crash on an AC86U could have been attributed to out of RAM.
Also, I set the CPU Wait to off to see where the temp went.

 

[OzarkEdge]

Real temperature.

Edit: Am a bit concerned this morning. Not with the temperature but with RAM usage rise overnight. RAM used went from 72% to 93% with nothing unusual traffic wise. OK, I do have a swap that the system can use but loads of others don't and I've seen at least one case in the past day or so where it appeared a crash on an AC86U could have been attributed to out of RAM.
Also, I set the CPU Wait to off to see where the temp went.

My RAM installs around 55% and is holding around 75%, which is normal with AiProtection running.

OE

 

[RMerlin]

With this temperature increase, which may lead/cause a total failure of a router, is/are

It won't. The CPU will throttle its cores if the temperature reaches 100C, which is still well below the allowed max Tjunc.

 

[VANT]

It won't. The CPU will throttle its cores if the temperature reaches 100C, which is still well below the allowed max Tjunc.

but that is not the solution to the problem.
The problem is that Asus has disabled CPU wait

 

[alex1236hk]

since when ac86u became a buggy hardware to maintain
all kind of cpu wait, insufficient of ram, failing to upgrade
seems like asus is trying to "making" you upgrade to an AX model

 

[OzarkEdge]

It won't. The CPU will throttle its cores if the temperature reaches 100C, which is still well below the allowed max Tjunc.

Cool!

Knowing that my 90C peak is well below max Tjunc and that there is a fall back to CPU throttling is good to know.

Do you know if there is a way to detect throttling, if it occurs?

OE

 

[OzarkEdge]

but that is not the solution to the problem.
The problem is that Asus has disabled CPU wait

What is the problem... a hotter CPU?

OE

 

[bitsbytes]

since when ac86u became a buggy hardware to maintain
all kind of cpu wait, insufficient of ram, failing to upgrade
seems like asus is trying to "making" you upgrade to an AX model

Not long actually, the fw before this one that I'm using now has been running problem free for me 99% of the time. And I think I'm using every feature the router provides. Even this one seems to be running well for members who used it, it's just this temperature problem which if you are a "casual" user you wouldn't even notice it.

 

[128bit]

Not long actually, the fw before this one that I'm using now has been running problem free for me 99% of the time. And I think I'm using every feature the router provides. Even this one seems to be running well for members who used it, it's just this temperature problem which if you are a "casual" user you wouldn't even notice it.

well, there is a documented problem with the 2.4 guest ui "ports." the first port or section stopped working on that last rel. another user had a similar problem and the work around was to use port 2 or 3.

 

[RMerlin]

but that is not the solution to the problem.
The problem is that Asus has disabled CPU wait

I never said it did, I`m simply correcting some misinformation. Your router will not die because of this issue.

Do you know if there is a way to detect throttling, if it occurs?

The System Log will say something along the line of "CPU core 0 has been turned off" or something - I don't remember the exact wording. Check on the forums, a few users posted the log message a week or two ago (probably in my own Beta thread). What it does is it turns down some of the cores until the temperature drops back below 100C (I know it shuts down at 100C, I don't know the exact temperature at which it turns it back on).

What is the problem... a hotter CPU?

The biggest problem will be a drop in performance as CPU cores get disabled.


Asus are already aware of the problem BTW.

 

[[Jedi]]

@ASUSWRT_2020

I also forgot to mention the other day, I checked my nodes for a temperature reading, I had two nodes travelling around 86C, with the exception of one node, which was at 101C.

What has ASUS changed other than "CPU Wait", to cause such a large variation in the router temperature in this public release firmware?

With this temperature increase, which may lead/cause a total failure of a router, is/are, ASUS willing to replace the router (free of charge), because of a faulty firmware, as a result, which is causing this temperature increase?

Or, is a router running in excess of a temperature of 100c, 24/7, considered ok/safe?

Please enlighten us all, as to why, ASUS think this excessive temperature increase is considered ok/safe, by previous firmware releases for this router?

We all await your response.

Hi Neil62,
I totally agree.

A CPU temperature of 100°C cannot be considered normal nor safe.

I can't believe that Asus has found only this way to solve some issues (issues that I can't notice on 386.40451).

If that's true, I agree that Asus should replace the affected routers.

I never said it did, I`m simply correcting some misinformation. Your router will not die because of this issue.



The System Log will say something along the line of "CPU core 0 has been turned off" or something - I don't remember the exact wording. Check on the forums, a few users posted the log message a week or two ago (probably in my own Beta thread). What it does is it turns down some of the cores until the temperature drops back below 100C (I know it shuts down at 100C, I don't know the exact temperature at which it turns it back on).



The biggest problem will be a drop in performance as CPU cores get disabled.


Asus are already aware of the problem BTW.

Hi Merlin,
I wouldn't be so sure that the router won't take damage with the CPU constantly running around 100° C.

We're glad to know that Asus is aware of the problem but, as customers, we have also the right to ask Asus to intervene as soon as possible.

We hope to see this problem fixed soon.

Thank you.

Bye.

 

[DarkKnight75]

Well...I'm not taking chances either way as I need stable environment for work and zoom school sessions. Will wait till next release..meantime humming along latest .384 at 83 days uptime .

 

[webdriver]

Real temperature.

Edit: Am a bit concerned this morning. Not with the temperature but with RAM usage rise overnight. RAM used went from 72% to 93% with nothing unusual traffic wise. OK, I do have a swap that the system can use but loads of others don't and I've seen at least one case in the past day or so where it appeared a crash on an AC86U could have been attributed to out of RAM.
Also, I set the CPU Wait to off to see where the temp went.

Well I had the same experience of constantly memory reach high percentage, and I remember one time it does crash the router. I have raised similar question but as I am also trying out some add-ons so no clue if it's normal or not.

 

[follower]

https://www.ti.com/lit/an/sprabx4b/sprabx4b.pdf

Go to Hell ASUS

 

[Neil62]

Asus are already aware of the problem BTW.

Lets hope there is a fix from ASUS soon..

 

[Neil62]

It won't. The CPU will throttle its cores if the temperature reaches 100C, which is still well below the allowed max Tjunc.

Thank you for the information