RT-AC68U / RT-AC68P / RT-AC1900 / RT-AC1900P

[Pierino]

Ok, so we can't even identify them by fcc numbers anymore? Lol
My cfe is 1.0.2.5.

 

[SLT]

I'd like to contribute to this whole revision mess with my findings. After reading a couple of threads I think this would be a BCM4709C0 CPU since "nvram get cpurev" returns a value.

RT-AC68U bought in Sweden (MediaMarkt), Nov 24 2016

MFG. YEAR 2016
H/W Ver: E1
F/W Ver: 3.0.0.4.380_3264


System log
...
Aug 1 00:00:18 kernel: eth0: Broadcom BCM47XX 10/100/1000 Mbps Ethernet Controller 6.37.14.126 (r561982)
...
Aug 1 00:00:18 kernel: eth1: Broadcom BCM4360 802.11 Wireless Controller 6.37.14.126 (r561982)
Aug 1 00:00:18 kernel: eth2: Broadcom BCM4360 802.11 Wireless Controller 6.37.14.126 (r561982)
...


cat /proc/cpuinfo
Processor : ARMv7 Processor rev 0 (v7l)
processor : 0
BogoMIPS : 1998.84

processor : 1
BogoMIPS : 1998.84

Features : swp half thumb fastmult edsp
CPU implementer : 0x41
CPU architecture: 7
CPU variant : 0x3
CPU part : 0xc09
CPU revision : 0

Hardware : Northstar Prototype
Revision : 0000
Serial : 0000000000000000


nvram get cpufreq
1000,666


nvram get cpurev
c0

 

[wouterv]

Aug 1 00:00:18 kernel: eth1: Broadcom BCM4360

It would be nice to clarify above system log message, I mean determine if it really tells you the wireless chip type or not.
Is there some one with an RT-AC1900P who can lookup the system log for this message and see if it shows "43602" instead?

 

[wocram]

It would be nice to clarify above system log message, I mean determine if it really tells you the wireless chip type or not.
Is there some one with an RT-AC1900P who can lookup the system log for this message and see if it shows "43602" instead?

On my RT-AC1900P:

Nov 30 21:42:54 kernel: eth1: Broadcom BCM4360 802.11 Wireless Controller 6.37.14.126 (r561982)

 

[wouterv]

On my RT-AC1900P:

Nov 30 21:42:54 kernel: eth1: Broadcom BCM4360 802.11 Wireless Controller 6.37.14.126 (r561982)

In other words, that system log message is generic for the RT-AC68/RT-AC1900 series (probably fixed in the firmware, which is equal for all models) and does not tell you the real wireless chip type.

 

[escape75]

I wonder if the RT-AC68U revisions that have a c0 running 1000 mhz can be switched to 1400 mhz?
My RT-AC1900P is clkfreq 1400,800 and cpurev c0 ...

 

[Odkrys]

RT-AC68U/R/W/A don't have shield so those USB 3.0 speed are up to 45~50mbps (Read)

But AC68P & AC1900P have shield so can reach 70mbps (Read)

Then how about RT-AC1900 ?
It is renaming of AC68U C1 or AC68P C1 ?
Its antennas seem same as AC68P.

Did anyone test this?

 

[RMerlin]

But AC68P & AC1900P have shield so can reach 70mbps (Read)

Shielding has nothing to do with the speed. It's the faster CPU of these two models that make the difference.

 

[Odkrys]

Shielding has nothing to do with the speed. It's the faster CPU of these two models that make the difference.

oh.. then what this review?

http://www.smallnetbuilder.com/wire...-band-wireless-ac1900-gigabit-router-reviewed

AC68P disabled Reducing USB 3.0 interference option, but not AC68U ?
it is not relate with shield ?

oh.. o_ o...
I saw other people's reivew and he said ac68p usb 3.0 speed was 65(r)/50(w).
My ac1900p speed is same with it (65~68/47~49) and then where is the 400mhz? hmm..

 

[Roy2001]

Shielding has nothing to do with the speed. It's the faster CPU of these two models that make the difference.

I agree with this. When I copy/write big files with HDD attached to router, CPU usage is 100% for both cores, overclock would make the speed faster. So at least for AC68U/P, USB speed bottleneck is SoC speed.

 

[sfx2000]

I agree with this. When I copy/write big files with HDD attached to router, CPU usage is 100% for both cores, overclock would make the speed faster. So at least for AC68U/P, USB speed bottleneck is SoC speed.

A router is not a good place to keep files for many reasons...

Better to put stuff over somewhere else - if not a NAS, it's fairly easy to set up a RaspPi as a Samba and DLNA server there with a much more capable OS.

And this keeps all the spare cycles free for the normal Routing/Firewall stuff...

 

[Roy2001]

A router is not a good place to keep files for many reasons...

Better to put stuff over somewhere else - if not a NAS, it's fairly easy to set up a RaspPi as a Samba and DLNA server there with a much more capable OS.

And this keeps all the spare cycles free for the normal Routing/Firewall stuff...

I am only using it for occasional movie/music sharing, only 2-3 times a week. I think router is ok for that kind of light duty work.

I have a DIY NAS with a i5 3570K CPU and that's an overkill for this kind of job. I am too lazy to upgrade the HDD's which has 5 2TB's. So it is convenient for me to just attach a 8TB drive to router, and I don't even care if it fails as I have a copy in NAS which is in storage room.

 

[sfx2000]

So it is convenient for me to just attach a 8TB drive to router, and I don't even care if it fails as I have a copy in NAS which is in storage room.

Stick the 8TB on the NAS, and start a new share there...

 

[Roy2001]

Stick the 8TB on the NAS, and start a new share there...

I need to put NAS to family room, behind the audio rack, where router is placed. I don't have LAN setup, that's why I am not keep the NAS running.....

Thanksk for suggestion, I will do that, if I find anything wrong

 

[sfx2000]

It would be nice to clarify above system log message, I mean determine if it really tells you the wireless chip type or not.
Is there some one with an RT-AC1900P who can lookup the system log for this message and see if it shows "43602" instead?

Normally HW variants on ARM can provide additional info on the HW line of cpuinfo, but for whatever reason either Asus isn't reporting this, or the chip itself is not reporting it.

Example below on some other Broadcom ARM chip... some can provide a family and revision - but this is sometimes programmed in on the OTP section of a small section of NV on the chip itself... but there is no requirement to do so.

processor    : 3
model name    : ARMv7 Processor rev 4 (v7l)
BogoMIPS    : 38.40
Features    : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32 
CPU implementer    : 0x41
CPU architecture: 7
CPU variant    : 0x0
CPU part    : 0xd03
CPU revision    : 4

Hardware    : BCM2709
Revision    : a22082
Serial        : 00000000ace8b5xy

 

[sfx2000]

Example below on some other Broadcom ARM chip... some can provide a family and revision - but this is sometimes programmed in on the OTP section of a small section of NV on the chip itself... but there is no requirement to do so.

The other place one might look is in dmesg itself - the kernel might print it - see below;

$ dmesg | grep CPU
[    0.000000] Booting Linux on physical CPU 0x0
[    0.000000] CPU: ARMv7 Processor [410fd034] revision 4 (ARMv7), cr=10c5383d
[    0.000000] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache
[    0.000000] PERCPU: Embedded 13 pages/cpu @b9f62000 s22592 r8192 d22464 u53248
[    0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4, Nodes=1
[    0.003116] CPU: Testing write buffer coherency: ok
[    0.052540] CPU0: update cpu_capacity 1024
[    0.052604] CPU0: thread -1, cpu 0, socket 0, mpidr 80000000
[    0.054731] CPU1: update cpu_capacity 1024
[    0.054737] CPU1: thread -1, cpu 1, socket 0, mpidr 80000001
[    0.055305] CPU2: update cpu_capacity 1024
[    0.055311] CPU2: thread -1, cpu 2, socket 0, mpidr 80000002
[    0.055829] CPU3: update cpu_capacity 1024
[    0.055835] CPU3: thread -1, cpu 3, socket 0, mpidr 80000003
[    0.055896] Brought up 4 CPUs
[    0.056024] CPU: All CPU(s) started in HYP mode.
[    0.056051] CPU: Virtualization extensions available.
[    1.029770] ledtrig-cpu: registered to indicate activity on CPUs

 

[RMerlin]

Better to put stuff over somewhere else - if not a NAS, it's fairly easy to set up a RaspPi as a Samba and DLNA server there with a much more capable OS.

Do you have any disk benchmark of a "NAS" based on a RaspPi? I have a feeling that performance might not be much better than with a router.

(and I'm suddenly having flashbacks to the Linksys Slug I sold one to a customer back in the day, we had a USB disk plugged to it for backing up his desktop across the office...)

 

[sfx2000]

Do you have any disk benchmark of a "NAS" based on a RaspPi? I have a feeling that performance might not be much better than with a router.

(and I'm suddenly having flashbacks to the Linksys Slug I sold one to a customer back in the day, we had a USB disk plugged to it for backing up his desktop across the office...)

The Pi2/3 running as a NAS would be constrained by the USB implementation - USB2.0 and the ethernet port runs off USB and is only 100Base-T... it would be serviceable, but there are better options perhaps - even in the DIY single board computer space (Odroid C2 comes to mind with GBe)

It would be faster than a slug though - clock speed and much newer kernel and samba..

Another option would be to look at some of the Chromeboxes out there - some can be reflashed with a full debian - if I recall, the Asus Chromebox is one of them, and with a fast intel celeron (not atom based, it's haswell based and multiple USB3 and GBe, that would be a fast little server)

https://gist.github.com/freekrai/b56ff76a68392f0456bf

(FWIW - I've still got an old Linksys NSLU2 (slug) sitting in the closet - they were useful, but painfully slow - but that code base is still around, and many distro's own a bit of debt to Linksys for that... not as big of an impact as their GPL drop on the WRT54G, but close...)

 

[linuxares]

Hey!

Have annyone happened to see that the temperatures are quite high on this router?

As you can see on the provided picture it's more or less idle at the moment.

 

[sm00thpapa]

If it still works fine don't worry about. My 88U, 68P and 66U run at that temp no worries.


Sent from my iPad using Tapatalk