Joined: 06 Jun 2006 Posts: 7492 Location: Dresden, Germany
Posted: Fri Oct 18, 2013 22:00 Post subject:
what i know from the code is that asus does force the xhci driver to force usb 2.0 mode for specific HDD vendors. but i dont know any reason for it. its not documented what they found out and whats wrong.
what i can say is that i have 2 board revisions of the AC68U. the newer one has more shields and also bigger heatsinks. . the higher rev is 1.68 written on the pcb . the older is 1.2x or something like that (cannot remember). but what i also have seen is that router does not detect even all devices well. some older usb serial connectors are simply not detected. my 4 TB test HDD itself works fine _________________ "So you tried to use the computer and it started smoking? Sounds like a Mac to me.." - Louis Rossmann https://www.youtube.com/watch?v=eL_5YDRWqGE&t=60s
hmmm its good to know that AC68U has a new revision. From my AC68R package box, it says HW ver. A1 , so I am not sure if I have the new revision board.
Well swe had a lot of trouble with USB 3.0 at work. The problem is, that some manufacturers ship low quality cable and have a bad controller inside the loss in the chain is too high and you often have trouble to recognize a drive or correctly transfer. I have a pile of cables next to my desk, we now ordered high quality usb 3 cables and all the drives which had trouble before work now.
Thus, if shielding is bad + bad cable + bad controller etc. then the whole chain sucks and it does not work. That's why they now specify loss for any part in usb 3.1 standard. _________________ KONG PB's: http://www.desipro.de/ddwrt/
KONG Info: http://tips.desipro.de/
Joined: 06 Jun 2006 Posts: 7492 Location: Dresden, Germany
Posted: Sat Oct 19, 2013 1:13 Post subject:
i dont think that the AC68 has a shielding problem. its a very well made pcb design. _________________ "So you tried to use the computer and it started smoking? Sounds like a Mac to me.." - Louis Rossmann https://www.youtube.com/watch?v=eL_5YDRWqGE&t=60s
what i know from the code is that asus does force the xhci driver to force usb 2.0 mode for specific HDD vendors. but i dont know any reason for it. its not documented what they found out and whats wrong.
There's an option on the webui to force the driver to work in USB2 mode. That might be complementary to any manufacturer-specific downgrade they might be doing. I don't have any additional information either as to why they are blacklisting some specific manufacturers.
BrainSlayer wrote:
what i can say is that i have 2 board revisions of the AC68U. the newer one has more shields and also bigger heatsinks. . the higher rev is 1.68 written on the pcb . the older is 1.2x or something like that (cannot remember).
There also a 1.70 revision where they seem to have some USB-related differences. See the set_pwr_usb() function inside boardapi.c:
I haven't opened my RT-AC68U yet, but since that nvram value does not exist, I assume I must also have the 1.6x revision.
On the shielding issue, someone mentioned that the shielding might be inside the pcboard, assuming its multilayers - the signal trace might be sandwiched between ground layers. No idea if that's the case. It would be weird for Asus to shield the AC56U but not the newer AC68U.
Joined: 06 Jun 2006 Posts: 7492 Location: Dresden, Germany
Posted: Sat Oct 19, 2013 14:06 Post subject:
led on off switch hasnt been implemented yet. (like for all other devices which does feature such a thing) _________________ "So you tried to use the computer and it started smoking? Sounds like a Mac to me.." - Louis Rossmann https://www.youtube.com/watch?v=eL_5YDRWqGE&t=60s
what i know from the code is that asus does force the xhci driver to force usb 2.0 mode for specific HDD vendors. but i dont know any reason for it.
The reason is that they use the not officially maintained kernel 2.6.36 which doesn't get bug fixes backported..
The HDD connection delay bug which affected drives from HP and WD was fixed month ago so it is not likely that the AC68U dd-wrt build is affected by that bug. _________________ Kernel panic: Aiee, killing interrupt handler!
what i know from the code is that asus does force the xhci driver to force usb 2.0 mode for specific HDD vendors. but i dont know any reason for it.
The reason is that they use the not officially maintained kernel 2.6.36 which doesn't get bug fixes backported..
The HDD connection delay bug which affected drives from HP and WD was fixed month ago so it is not likely that the AC68U dd-wrt build is affected by that bug.
Got any Kernel.org commit off-hand I could look at?
Broadcom did seem to have backported some things (as I found some upstream patches already applied, both on Asus and Netgear's GPL kernel).
there is already a fw available for the ac56. but yes, this firmware is compatible with the ac56 as well, you just cannot install it from the asus webgui. but if dd-wrt is already installed it will be accepted and it will also work without issues. the firmware detects the device automaticly. so in fact this image will run on alot existing BCM4708 based devices
I thought this one has a BCM4709? or do you mean that this firmware will work with BCM4708 and BCM4709?