Official GIGABYTE Forum

USB 3.0 speed anomaly

USB 3.0 speed anomaly
« on: January 16, 2015, 12:27:36 pm »
I'm having a problem with a particular piece of hardware on the ga-z87-hd3 board: the Netgear A6210 AC USB 3.0 WiFi adapter; when connected to USB 3.0 ports, Windows 7 tells me that it can perform better when connected to a USB 3.0 port. USBTreeView shows it connected via one of the two Intel EHCI USB 2 Hubs.

Other USB 3.0 devices (flash drive, drive dock) operate on the Intel USB 3.0 hub, according to USBTreeView, and do not (naturally) generate the W7 speed warning.

I've tried connecting the A6210 to all back and both front USB 3.0 ports with the same result and warning. I've tried the other two devices in the same ports, and these operate as they should.

I've tried various xHCI hand-off bios settings: smart auto/auto/enabled/manual to no effect.

I've connected the A6210 to my partner's PC, also W7, and it operates as it should. That system is on an Asus P8P67-Pro board using Intel EHCI contollers and a Renesas XHCI controller. My assumption is that the Netgear h/w and/or drivers are incompatible with the Intel XHCI controller. The Netgear adapter has the latest drivers intalled.

Device Manager tells me, when trying to update, that the Intel XHCI drivers are the latest available, though I have seen later versions available online. When attempting to update via the Gigabyte & Intel Update Utilities, and the Gigabyte & Intel Download Centres, installation exits before completion and turns the USB system to fruit salad.

Can anyone tell me if there is an order in which updates must be performed? For instance, are certain controller updates dependent upon the chipset driver version installed? And are both dependent upon the current BIOS version?

GA-Z87-HD3
F6 bios.
Chipset: 9.4.0.1017 / 6.1.7601.17514
XHCI/Hub : 2.5.0.19
CM SS2 A | Corsair AX 760i | z87-hd3 Rev 1.0/F8 | i5 4670K
Vengeance 16GB DDR3 1600 | GTX 760
Samsung 840 Pro 256GB | Seagate 500GB
W7 Pro x64

shadowsports

  • 2258
  • 67
  • Xbox One, Drives STI, Use QVL RAM For Best Results
    • Gigabyte US
Re: USB 3.0 speed anomaly
« Reply #1 on: January 17, 2015, 05:56:33 pm »
In the majority of cases, this has to do with the driver version of the USB Controller Hub.  Second is often BIOS revision and last the OS being used.

This might be related to speed negotiation, where the board is using 2.0 speeds since 3.0 cannot be achieved.     

I'm relatively new to Gigabyte, but remember there being issues with USB 3.0 implementation on some of the z87 based boards.  Others here can probably comment further, but I believe later board revisions in at least some of the cases resolved the issue.  If you are using an external hub powered or unpowered it should be removed.   
Z390 AORUS PRO (F10) \850w, 9900K, 32GB GSkill TriZ RGB - 16-18-18-38, RTX 3080Ti FTW3 Ultra, 960 Pro_m.2, W11
Z370-HD3P (F5) \750w, 8350K, 8GB LPX 3200 - 16-18-18-38, GTX 970 FTW SC, Intel SSD, 2TB RAID1, W11
Z97X-UD5H \850w, 4790K, 32GB Vengeance, RTX 2080 FTW

Re: USB 3.0 speed anomaly
« Reply #2 on: January 18, 2015, 02:44:11 pm »
Yeah--z87 boards were regarded as a breakthrough in cyber-sport development; forum registration tripled and Xtreme trouble-shooting became a recognised clinical condition with RMAs being one of the few non-prescription remedies...

I digress [placeholder4apologies]--no external hub here, powered or not; the F6 bios update addressed a few initial issues, one being USB port recognition. However, the Netgear scenario appears to be controller specific.

The Asus board manual states that only data drives will be recognised & operate as USB 3.0 devices, yet it does appear on the USB 3.0 hub and is capable of performing  at (hypothetical) SuperSpeed, which suggests to me that the Netgear software is designed to present a profile at hand-off which results in the A6210 being handled by the USB 3.0 controller and operating on the 3.0 hub. Assuming that's a plausible hypothesis (a big assumption), then there must be a radical difference in the way the Renesas & Intel drivers allocate devices and resources.

LSS, I've attempted to update the Intel xHCI-related drivers, but that consistently fails, with the installation exiting before completion--an example extracted from the Intel logs below:
!   No devices found for driver. Skipping uninstall of previous driver.
[INF - 550]
  Installing F:\Downloads\system\Intel\Intel Components\win64_15367\Intel(R)_USB_3.0_eXtensible_Host_Controller_Driver_3.0.2.54\Drivers\Win7\x64\iusb3xhc.inf
  INF = F:\Downloads\system\Intel\Intel Components\win64_15367\Intel(R)_USB_3.0_eXtensible_Host_Controller_Driver_3.0.2.54\Drivers\Win7\x64\iusb3xhc.inf
  Execute command: F:\Downloads\system\Intel\Intel Components\win64_15367\Intel(R)_USB_3.0_eXtensible_Host_Controller_Driver_3.0.2.54\x64\Drv64.exe -driverinf "F:\Downloads\system\Intel\Intel Components\win64_15367\Intel(R)_USB_3.0_eXtensible_Host_Controller_Driver_3.0.2.54\Drivers\Win7\x64\iusb3xhc.inf" -flags 20 -keypath "Software\Intel\Difx64"
!   DriverPackageInstall error 0x3FD
!   Error opening Software\Intel\USB3 - error :  0x0
IIF will NOT initiate reboot
Exit code = 0x3FD
ResultCode = 1021
<<< 1/08/2015 23:51:35:482


I'm now wondering if a BIOS (& possibly chipset driver) update is necessary before the USB 3.0 controller drivers can be installed(?).
Other resolutions are probably somewhere beyond my cerebral event horizon.
CM SS2 A | Corsair AX 760i | z87-hd3 Rev 1.0/F8 | i5 4670K
Vengeance 16GB DDR3 1600 | GTX 760
Samsung 840 Pro 256GB | Seagate 500GB
W7 Pro x64