Found a thread on the "Asus Republic of Gamers" forum discussing same USB3 issue on Asus P67 boards.
On page one of the thread, a user describes the problem in reference to the front-panel ports of his CoolerMaster HAF-X case:
"My haf-x only works if I plug the 2 USB3 ports in the back usb using loopthrough.
If I use the motherboard USB3 header, the Haf-x usb3 front ports work with usb2 devices but are extremly unstable with USB3."
Does that "extremely unstable" part sound familiar?
And then on page 3, there is a response from someone whose forum name indicates they work at ASUS:
"Found out from RD this morning: It's a known issue with the NEC USB 3.0 that if your recipient device uses JMicron generally have poor compatibility, but it also largely depends on the design of the recipient device too (i.e. not everyone is made equally)."
Now that comment doesn't really address front vs rear ports acting differently, but... The one device I have that doesn't work on the motherboard header ports uses a JMicron controller. The other USB3 devices I have (Corsair and ASMedia) work fine. (And of course the JMicron device works on the rear ports.)
NOTE: Bear in mind yesterday i had re-installed the original USB 3.0 Renesa 18.104.22.168 drivers direct from Intel`s download center site and my problems still existed.
An hour ago i un-installed or deleted the light greyed out areas once again in both Storage Volumes & USB devices that indicate previously detected devices as we discussed yesterday.
D\L the latest Renesa USB 3.0 version 22.214.171.124 drivers and installed them. Shut the pc down and rebooted.
Tested the previous known working USB 3.0 front port housed with the AFT USB 3.0 Card Reader that is plugged into the rear DZ68BC USB 3.0 I/O ports and it works just fine also with the new Renesa 126.96.36.199 drivers. After this i once again deleted the Storage Volume & USB devices greyed out detections so the next port test would require a driver d\l again from a new test port site.
Then tested the Storm Enforcer top left USB 3.0 port connected to the dreaded DZ68BC internal USB 3.0 BB 19pin header connection and with utter amazement it was detected, installed the drivers and works just great transferring a large 1.5gb file at over 100mbs using the below Adata S102 16gb Flash drive that previously would not work with the originally installed Renesa USB 3.0 188.8.131.52 drivers from Intel`s Download center site.
Then tested the front top right Storm Enforcer USB 3.0 port and it reacted sort of quirky at first but finally responded and it appears so far to be working just fine also.
I have NO idea how long they will work since some folks have claimed they have just quit working or become intermittent, but it looks like the guy i mentioned on Newegg last night that could not get his Rosewill USB 3.0 Pci card to work until he installed the latest .1 drivers might just be right.
I still have absolutely no clue as to why the rear DZ68BC USB 3.0 ports work but the problem in my case and others is always associated with the internal 19 pin Motherboard header.
So as i suspected NO Intel Bios update cures this problem, nothing wrong with the Coolermaster Storm Enforcer front case ports, cabling or 19 pin female connector, and swapping out the motherboard surely would not cure it either.
Of course like alot of these phenoms only time will tell but im glad this fiasco is cured and hopefully not just temporary.
Might not cure all the USB 3.0 issues out there with various 3.0 plug-in devices but i would highly suggest folks install these latest Renesa 184.108.40.206 drivers released on 1/19/12 that Intel has still not added to windows updates or to there download center site.
I checked the "220.127.116.11" driver package, and according to the the release notes it still uses the same exact driver files as the previous package (nusb3xhc.sys 18.104.22.168, and nusb3co.dll 22.214.171.124). The only thing changed was the installer, which was updated to eliminate the need to "uninstall first" when previous version was 126.96.36.199. Even says that the "Windows Logo Information" is the same as 188.8.131.52, which means the driver didn't actually change.
Could you check the actual driver file version for nusb3xhc.sys in device manager and see if it is 184.108.40.206 or still the older 220.127.116.11?
If they are still 18.104.22.168, then I strongly suspect the problems aren't gone, but just hiding temporarily due to the cleanup/redetect.
Your right in Device Manager clicking on properties/ driver of each USB 3.0 controller or hub it still shows 22.214.171.124 but you explain to me why they are ALL working now and yesterday after re-installing the original USB 3.0 126.96.36.199 d/l from Intel they were still screwed up and would not work. I simply installed the newer .1 version over the top without uninstalling the previous.
Plus explain the .1 version curing the guy on Newegg`s problem ?
I certainly dont disagree as i noticed after checking in Device Manager they did not reflect .1 after i initially installed them, but for whatever reason they are working now unless like i mentioned time will tell and they once again will fail or become intermittent.
Well, it's exactly the same driver files -- so the "explanation" would be that something else is at play. Perhaps the very act of uninstalling / redecting the devices again, or that it is just an intermittent problem in the first place -- so sometimes it works, sometimes it doesn't. Really easy to misidentify something as a potential solution under those circumstances.
I'll give the new driver package a try, and see if it has any effect on the JMicron device that I replicated the problem with earlier.
Did you uninstall the old driver first, or just run the new installer package?
Uninstalled old device detections from device manager, loaded the 188.8.131.52 driver package, rebooted.
Connected the Startech USB3 dock (Jmicron based) to the front (motherboard) USB3 and got the same bad behavior as before. It connects, shows "installing", then there is a rapid succession of connects/disconnects and then not recognized. Also tons of errors in event log. Tried it on the rear USB3 ports and it works fine.
None of this is surprising, as the driver files in the 184.108.40.206 package are still the same ones delivered with 220.127.116.11 !!! I captured a copy of the files before running the installer (nusb3hub.sys, nusb3xhc.sys, and nusb3co2.dll). Did a comparison to the files installed by the "18.104.22.168" package, and they are 100% binary identical to the old files.
Just as the 22.214.171.124 release notes say, the drivers have not changed -- they are still 126.96.36.199 -- and the only change is that the installer doesn't require "uninstalling" very old versions before upgrading.
No i did not un-install the previous drivers. I just installed the new .1`s over the top of the old ones. I sure dislike it when we run into problems that cannot be rectified with a specific known solution. So many times mere speculation on the causes & possible remedies we read & hear about not only on the various forums but also the grabbing for a solution mentality consumers receive from Product technical support can surely get frustrating for us all. The final straw is when folks are told by Support to RMA a product for replacement and end up with the exact same issue.
As you previously mentioned hopefully a new Renesa firmware along with actual new drivers will be released to possibly once & for all resolve the issue for so many folks we read about all over the internet with various brands of motherboards & USB 3.0 plug-in devices.
Thanks again for all your help & advice. It is surely appreciated.
Decided to take one more run at this, since the "new" driver package (with old drivers) didn't help me.
I was led to this second attempt by a message on another forum, where a user had reported successfully using a device on a FRONT port only after it had been "initially detected" on a REAR port. I also noted this was the sequence you detailed in your successful attempt, in that you tried the device on the REAR port first and then subsequently plugged it into the front port. Well, it worked for me too.
Keep in mind, though, that my Corsair 16gb USB3 stick and my Rosewill USB3 drive work just fine being plugged into the FRONT port -- without ever first being on the REAR port.
So I think there is a LOT going on here... The sequence of removal, rebooting or not, sequence of detection, and specific device all seem to have a role and could result in different outcomes. Also, I think some of these devices communicate differently than others: Some communicate directly as a USB device and others use the USB as a "bridge" to get to PCI/SCSI driver support. That may be why some of my devices (the Corsair and Rosewill) always work fine on the front port, while the StarTech/Jmicron would not work on the front port until I used the specific sequence noted above under "second attempt".
I still think the ultimate solution lies with a new Renesas firmware (and perhaps drivers). But given that we have now both had success on the FRONT ports with devices that previously had problems there... It appears there may be some workaround related to the sequence of letting a device be "first detected" on the rear ports. Or perhaps it is something else we both did, but failed to recognize as significant... I would have to do several more test sequences to proclaim this as the definitive reason for our success, and I'm running out of interest since -- for the moment -- everything is now working. :-)
I will want to revisit this if/when a new Renesas firmware is available. Perhaps we can revive this thread at that time.
I am still of the opinion that solving the problems described in this thread will require a firmware update, not drivers alone. And Renesas hasn't provided any new firmware for months. :-(
Seems like Intel (and a lot of other OEMs) made a mistake sourcing USB3 from Nec/Renesas.
In case anyone is still monitoring this thread, I wanted to provide an update.
Renesas has released new firmware, 188.8.131.52, for the USB3 controllers on the DZ68BC. It can currently be found on the "station drivers" web site. The latest drivers, 2.1.36, are also available there. Intel has not yet posted either of these in the Intel download center.
I updated to the latest Renesas drivers (2.1.36), and flashed the firmware of both Renesas USB3 controllers to the new 184.108.40.206.
Repeating the tests outlined in my earlier posting (see above, dated 3-29-2012), I got exactly the same (bad) results. The combination of the 220.127.116.11 firmware and 2.1.36 drivers do NOT fix the problem. As before, ALL my usb3 devices work fine on the "rear" USB3 ports, but the Jmicron-based devices fail on the "front" USB3 ports. Also, the event logs still contain the "Driver PCI returned invalid ID for a child device (FFFFFFFFFFFFFFFF00)" message at each boot.
I guess it is time to get a USB3 hub, plug it into one of the rear ports, and put some electrical tape over the front ports.
FYI... If anyone else wants to try the 18.104.22.168 firmware, beware that it is not yet tested by Intel and is a bit tricky to load on the dz68bc. The files available at "station drivers" will fail when more than a single Renesas controller is present (and the DZ68BC has 2). To load this firmware, you must download the 22.214.171.124 update package AND the 126.96.36.199 package, then use the utility supplied with 188.8.131.52 (w200fw35.exe) along with the firmware file supplied with 184.108.40.206 (F402003.MEM) to accomplish the update. Also, you must first run the w200fw35 with the "/srom ?" to get the bus address of your Renesas controllers, then perform the update twice (once for each controller) specifying the "/address 05-00-00" and "/adress 06-00-00" (assuming your controllers come up with bus addresses of 0x05 and 0x06).