[Solution] Win7 drivers for USB 3.0/3.1 Controllers of new Intel chipset systems

Operating Systems Windows 7/Vista/Server 2008
261

@PatrickK
My hardware is : strix z390-f gaming plus mobo, cpu intel i7 8700k, 250GB GPT ssd with Win10, 1TB MBR ssd where I want to install Win7, GeForce GTX 1070 ti mini.

262

@Illuminate Ok, I suggest that you just try it. Keep in mind that Windows puts its boot menu entries into the first drive, no matter where you install Windows. That means you need the other SSD in order to keep Windows on the other SSD bootable. Or Windows 7 puts its boot menu entry in the same drive because you have one with GPT and one in MBR, for whatever reason.

263

@PatrickK checked again, both drives are MBR, not sure why I thought otherwise. Just to make sure I’m doing this right, I stuck my iso named x64 in the slipstream folder, ran the batch file as admin, chose windows pro, let it do its thing and when it finished used that same ISO to do the install. Is there a specific way I need to install it? such as winpe or winntsetup or click certain settings when starting the install? I’m still unable to use mouse or keyboard after installation.

264


Not necessarily, you can also install win7 on pure uefi (no csm) with gpt. The only thing you have to do is disable standard vga driver and slipstream the integrated graphics driver for windows to load during the installation. Of course for 8th gen intel or above and similar, you might have to use a modified driver which you may need to sign and slipstream the certificate into the installation. You can do all this using win10pe and install with dism or modify win7 iso as previous posted with win10 install wim and uefi files.

I had to do this myself on my laptop because it lacks cwm (uefi class 3) and I can confirm it works.

Sorry for being a bit offtopic.

265

@bloodhand Your citation is not correct, in the next few words I mention that this is the case if not modified.
No man, not offtopic, your reply is great, I think that this no-CSM-topic will be an important thing for Windows 7 in the future with UEFI’s.
Do you have some instrucitons to view online or did you everything on yourself? I want to know how this works.

266

Thanks a lot Canonkong ! I installed your driver on msi z390M and it works like a charm…

Do you think it would be possible to update IRST drivers in the same way ?

I compared iaAHCI.inf & iaStorAC.inf of the 15.9.8.1050 and 16.8.3.1003 versions download from intel website and I noticed that additional VEN_8086&DEV_ID are declared for 300 series in the 16.8.3.1003 version.

As the 15.9.8.1050 version is recognized by Z370 chipset but not by Z390 (under win 7), I was wondering if the same “trick” would be apply, and if yes, which DEV_ID would be selected for Z390 chipset (= 9DD3, A352 or A353 ?).
Thanks for your answer

267


I decided to post my guide on this forum too, maybe people don’t browse notebookreview forums anymore: [GUIDE] How to install Windows 7 x64 on Acer Spin 5 laptop (guide can be adapted to more UEFI class 3 machines)

268

@Illuminate That’s the right way to do it.

269

@bloodhand Great, thank you, bookmark saved for later.

270

Integrated the signed driver into the Windows 7 image using the NTLite utility. Windows installs successfully, but each time you reboot, you have to choose the "no digital signature verification" mode. I can sign the driver on this computer with DSEO, but I want to create a working installation image. How can I do that?

271

@roman-sln If you want, you dan use my mobile slipstreaming script, it uses the option /forceunsigned in its lines, that might help you.

272

@canonkong Please upload the drivers soon, they would be really useful to me. (the generic USB controller drivers for xp, vista, 7, 8, 8.1)

273


I asked that too, but he couldn’t make them work. I also have an ACPI\USBC000 USB-C controller that will never work under xp/vista/7/8/8.1 because there is no UCSI USB Connection device class for that, only on win10.

By the way, did you succeed installing win7 on your uefi machine? https://www.sevenforums.com/installation…-my-bios-4.html If yes, how did you do it in the end?

274

Z490 chipset is very similar to Z390 chipset in shape. Device ID(Z490’s USB 3.1 xHCI Host Controller) is probably 06ED.

Z490
https://www.xfastest.com/thread-240614-1-1.html
Z390
https://www.xfastest.com/thread-223309-1-1.html
Z370
https://www.xfastest.com/thread-201075-1-2.html

275

Hello everyone, I have a project simplix.info/win7usb3 for installing USB drivers under Win7 (please use a translator if necessary). It can integrate drivers into the system distribution, or install from WinPE into a working system, or simply install drivers on a working system. Here, the installation is not using Dism, because some versions of Intel drivers conflict with each other, instead, installation takes place in a few seconds during the second stage of system installation. Prior to Win7USB3 1.5 inclusive, I used only original drivers, but now I decided to check the modified ones from this forum. The important thing is that I have a purchased certificate, so the installation of a modified driver occurs without warning.

Win7USB3 works in two stages - first, it checks what hardware is physically present, and then installs only those drivers that are needed. If the HWID is present in several versions of the drivers, then a newer one is selected. The problem is that in the modified drivers there are HWID’s, which are also in the official drivers. I have a few questions for @canonkong before I put Win7USB3 2.0 in public access:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
 
AMD:
PCI\VEN_10DE&CC_0C0330 ;Doesn't match AMD manufacturer
PCI\VEN_1022&DEV_13ED
PCI\VEN_1022&DEV_13EE
PCI\VEN_1022&DEV_145C ;Already present in v2.0.0.60
PCI\VEN_1022&DEV_145F ;Already present in v2.0.0.60
PCI\VEN_1022&DEV_148C
PCI\VEN_1022&DEV_149C
PCI\VEN_1022&DEV_15D4
PCI\VEN_1022&DEV_15D5
PCI\VEN_1022&DEV_15E0 ;Already present in v1.0.0.12
PCI\VEN_1022&DEV_15E1 ;Already present in v1.0.0.12
PCI\VEN_1022&DEV_15E5 ;Already present in v1.0.0.12
PCI\VEN_1022&DEV_1639
 
Intel:
PCI\VEN_8086&DEV_A36D
PCI\VEN_8086&DEV_A36D&REV_10 ;Superfluous?
PCI\VEN_8086&DEV_A36D&CC_0C03 ;Superfluous?
PCI\VEN_8086&DEV_9DED
PCI\VEN_8086&DEV_31A8
PCI\VEN_8086&DEV_A1AF ;Already present in v5.0.4.43
 
Lines with comments are now excluded from the search, indicating the reasons, namely:
1) "Already present" means that instead of the modified one, the original driver of the specified version will be installed.
2) "Does not match the manufacturer" - I did not understand the criterion for adding VEN_10DE to the driver with VEN_1022, stability is questionable.
3) "Superfluous?" - I did not understand the point of adding REV_10 and CC_0C03 when they contain DEV_A36D, which is already represented as a separate line.

One member of my forum has already successfully installed a test version with hardware VEN_8086&DEV_A36D. If you want to test it, write the result of the devcon command on your motherboard:
1
 
devcon find "PCI\*"
 
276

Hi simplix and welcome. A pleasure seeing you around here.

I will try to answer your questions:

1. If you have access to a nice SPC to resign, you’re in control. Removing/commentting out officially supported devices from INF won’t impact the mod. However note that the Intel device list in your post was used in the old “unsigned” (actually not timestamped) version of the mod offered in OP. Last version offered there was the SHA2-signed (and properly timestamped) which contained only the following devices:

1
2
 
PCI\VEN_8086&CC_0C03
PCI\VEN_8086&CC_0C0330
 


as suggested in OP to avoid using any specific DEV ID and be more "future-proof". But as I said, if you can resign, just offer only actual, requested devices on INF, so driver fit your installation approach.

2. canonkong found that this AMD driver can handle USB-C controller present in recent Nvidia cards, hence allowing Win7 support for it. I've not personally confirmed this because I lack the required hardware. Note the device entry follows the same generic approach used for Intel PCI\VEN_8086&CC_0C0330.

3. Redundant entries, yes. Can be removed.

Regarding devcon, just note it's not a built-in Windows tool (at least on Windows 7). It comes from WDK/EWDK/SDK or other developer-aimed kits.
277

where to download ,thanks~

278

Hi! I lost the files and now I need them again. But i can’t find a DL link.


Can someone share by PM? thanks!

My system is Z390 + coffee lake. Just restored from Clonezilla working old backup but now my mouse is kaputt :frowning:

279

Thanks, Reducer. I decided not to add the extra lines that I commented on above. I believe that it is not safe to add an unofficial driver for any device, so now I have limited myself to the following lines:

1
2
3
4
5
6
7
8
9
10
11
12
13
 
AMD:
PCI\VEN_1022&DEV_13ED
PCI\VEN_1022&DEV_13EE
PCI\VEN_1022&DEV_148C
PCI\VEN_1022&DEV_149C
PCI\VEN_1022&DEV_15D4
PCI\VEN_1022&DEV_15D5
PCI\VEN_1022&DEV_1639
 
Intel:
PCI\VEN_8086&DEV_A36D
PCI\VEN_8086&DEV_9DED
PCI\VEN_8086&DEV_31A8
 
No problem adding new HWID's after verification. The new version of Win7USB3 2.0 is already available.
280

Hi PatrickK,

I installed your deWin7x64_Pro_SP2.10.iso and it works very good.
Now I tried your slipstreaming script en mobile slipstreaming x64 v1.1.zip
on en_windows_7_professional_with_sp1_x64_dvd_u_676939.iso.
Your batch file looks very well written! But it produces errors during this statement:
dism\dism /image:mount /add-driver /driver:DriverSP2.9\x64 /recurse /forceunsigned
Error message:


Extract from C:\Windows\Logs\DISM\dism.log:


Could you please have a look? If you need more info please tell me.