Do Motherboard manufacturers modify Intel Drivers?



I think you are mistaken here. What would prevent a motherboard manufacturer like AsRock from modifying a TXTSETUP.OEM file which is basically a text file which requires no source code to perform? Unless there is some legal contract with Intel that specifically states they couldn’t but I doubt Intel would go that far. But Intel would never release the source code of their SATA controller driver. I think they would keep this locked down unless they had a NDA and licensed it to some company and if so we would never know which companies.

What kind of proof were you looking for? I copied the TXTSETUP.OEM from the original source files. One came downloaded directly from Intel’s site, the other from Asrock’s site.

Did you inspect each TXTSETUP.OEM file I provided in the spoilers?

Are you saying that the changes between the two files don’t appear legitimate or significant?

Do you feel that these TXTSETUP.OEM files I posted are not authentic in any way?

I would not lie and tamper these files to prove any point. I copied them exactly from the source file.

Or are you not believing any manufacturer would ever change the TXTSETUP.OEM file to differ from what Intel has on their own site?


Furthermore I repeat: Intel’s Download Center has never “officially” released any Intel drivers for customers, who are using an ASUS, Gigabyte, ASRock or MSI mainboard. The delivery of suitable drivers for their mainboards is the task of the mainboard manufacturers themselves and not the Chipset manufacturers.

Yes I’m aware that Intel themselves wouldn’t offer say modified drivers for Asus or other motherboard manufacturer and have them listed on Intel’s site that would be the company themselves releasing the drivers on their own non-Intel site. The only time I’ve seen a non Intel driver released on Intel’s site would probably be for some non-Intel onBoard sound driver and non-Intel onBoard USB 3.0 from NEC / Renesas for their Intel desktop motherboards. I suppose at the time Intel needed a 3rd party manufacturer to integrate onBoard audio and they were still too far away from finishing their own Intel USB v3.0 while other companies had been doing this years prior. If I recall Intel didn’t add the Intel USB v3.0 until Z77 or 7 Series. I believe even 6 Series and earlier only companies like NEC and Asmedia had finalized their own USB 3.0 versions before Intel so Intel came very late into the game.

This coincides with this article dated in April 2012.
https://www.cnet.com/news/intel-delivers…-chips-finally/

As my first Z77 did include Intel USB 3.0 ports on it. I even recall on the motherboard box stating Intel USB 3.0 finally introduced and PCIe 3.0 was introduced as well which is why Ivy Bridge is so revered.

Anything prior to the release of Intel USB 3.0 was Renesas USB 3.0 based.

If April 2012 was the earliest Intel USB 3.0 was introduced then it doesn’t make sense.

https://downloadcenter.intel.com/product/65855

https://downloadcenter.intel.com/downloa…y?product=65855

USB 3.0 Driver: Intel® USB 3.0 eXtensible Host Controller Driver for Intel® 7 Series/C216 Chipset Family
Version: 1.0.10.255 (Latest) Date: 2/3/2014

Since we know Intel never released Intel USB 3.0 drivers for XP or Vista we can conclude that only Windows 7 was the starting point for driver versions.

If you look closely and we assume April 2012 was the earliest possible release date of a Intel USB 3.0 driver and in less than 2 years time Feb 2014 the driver version is still 1.0.10.255.

You can only conclude that the first Intel USB 3.0 version must have been 1.0.00.XXX. I see no way they would have started at version 15 and then dropped back to or gone from version 1 and jumped to version 15 in less than two years time.

But even for Renesas I could only find an earlier driver for it on board a X58 Motherboard dated Dec 2010 v2.0.30.0.
Then Version 2.1.28.1 Dated April 2012 in about 1.5 years time difference.
https://downloadcenter.intel.com/downloa…-Desktop-Boards

So judging by these two USB 3.0 versions for Renesas vs Intel neither seems to advanced their versions numbers much to have reached 15.



Getting back to the TXTSETUP.OEM, in this case I have found two versions of the same driver one from Intel and one from AsRock which contained two different TXTSETUP.OEM files for the Intel SATA controller driver. The one modified dated earlier was AsRock which leads me to believe Intel provided AsRock the driver first and AsRock manipulated it themselves for their motherboard. Either way it was altered whether or not Intel released it first or AsRock did as the TXTSETUP.OEM files do not match which I believe you are disputing would ever occur.

Did you do a google search for these two Floppy based SATA v10.8.0.1003 driver packages from Intel and AsRock?

If you can’t locate them I’ll go hunt them down for you if you don’t believe my TXTSETUP.OEM files I posted are authentic. I downloaded these files ages ago so I know these are authentic files since I gathered them directly from the websites of Intel and AsRock and not some 3rd party site.

@XPLives :
Since I have much more important things to do and don’t see any reason for a withdrawal or correction of my previous statements regarding your question “Do Motherboard Manufacturers modify Intel Drivers?”, I will not continue this useless discussion.
My simple answer is and stays: "No!"
If you don’t believe me, ask Intel. They would be very interested in getting the proofs for your assertion.
By the way: Microsoft would never give its WHQL stamp to an Intel driver, which has been modified by someone else.



I think before we can make a firm concrete answer we have to establish a few things.

We both have important things to do but I didn’t bring up this manufacturer will never modify Intel’s files.

But I’m asking for a direct answer from you.

Did you examine the two Spoilers in the earlier message where I posted the two differing TXTSETUP.OEM files?

One from Asrock and the other from Intel?

All I need to know is do you consider these files the same or modified files?

As for does Intel care if some motherboard manufacturer alters the TXTSETUP.OEM file? I would say no they don’t care because if they did they would have stopped AsRock a long time ago if this was an issue. And as far as I know AsRock is still selling motherboards with Intel SATA controllers so this is a clear indicator Intel does not care that they did this. I also found evidence this is not the first time this has been done.

If you want to compare the original TXTSETUP.OEM from Intel to the one I posted in the Spoiler go right ahead.
I didn’t alter it as I said before.

[Intel] TXTSETUP.OEM v10.8.0.1003
Monday, October 17, 2011, 3:34:12 PM
5.19 KB (5,322 bytes)

Direct link:
https://downloadmirror.intel.com/20768/e…_f6flpy-x86.zip

Extract the zip file and locate the TXTSETUP.OEM you will find this matches the one I pasted.

Once we are in agreement this is an unaltered “GENUINE” Intel originated driver file I will locate the AsRock version for you to compare.

@XPLives :

The content of both files is different, but none of them has been modified by ASRock. Both files have been edited and digitally signed by Intel, before they were sent to Microsoft with the demand to get a WHQL stamp. As last step these drivers were delivered to the mainboard manufacturers.
Why should an OEM like ASRock modify any driver? All on-board Controllers and the related drivers are built/compiled by the chipset manufacturers (Intel, Realtek, Marvell etc.) and not by the OEMs.
It is the task of the chipset manufacturers to deliver matching drivers to the OEMs and it is the task of the OEMs to make these original (untouched) drivers available for the consumers.

The content of both files is different, but none of them has been modified by ASRock. Both files have been edited and digitally signed by Intel, before they were sent to Microsoft with the demand to get a WHQL stamp. As last step these drivers were delivered to the mainboard manufacturers.
Why should an OEM like ASRock modify any driver? All on-board Controllers and the related drivers are built/compiled by the chipset manufacturers (Intel, Realtek, Marvell etc.) and not by the OEMs.
It is the task of the chipset manufacturers to deliver matching drivers to the OEMs and it is the task of the OEMs to make these original (untouched) drivers available for the consumers.
.




Sorry I’m not sure I know what you mean are you saying Intel modified it and gave it to AsRock and so does every other MB manufacturer before they are resent to Microsoft by each MB manufacturer? Wouldn’t that add a lot more middle man involvement for Intel now that you say AsRock sends their file to Intel and has Intel sign it and then sends it to Microsoft to add the WHQL stamp and then sends it back to AsRock.

The TXTSETUP.OEM file is just a text file. Why would AsRock or all other MB companies go through all the trouble to do this?

It make more sense that the driver.sys is not modified nor the driver.cat file. Modifying the TXTSETUP.OEM file would seem safe as long as the company tests their modifications thoroughly before releasing it on their site.

You keep asking why would the OEM like AsRock modify their TXTSETUP.OEM file. It’s because they are simplifying the install process for the user. If you’ve ever installed the SATA AHCI driver by a floppy drive you will know only the first 4 storage devices will show up in the list the rest you will have to navigate. So if you have more than 4 some people may not realize this and assume there are only 4 to choose from. The TXTSETUP.OEM install wasn’t made to be user friendly in that regard so it isn’t apparent to the user there are more than 4 to choose from which is why I assume AsRock eliminated a bunch of these storage devices from their TXTSETUP.OEM file to avoid any confusion for the user when manually selecting their “matching” SATA controller or get a BSOD 7B trying over a dozen other incorrect ones through trial and error.

So are you now saying that even though the AsRock modified the TXTSETUP.OEM you don’t consider this as being modified? I never said the driver.sys file was ever modified at all. Examine both TXTSETUP.OEM files from both files. Are they the same size?

I have located for you the exact file matching the file I got from AsRock.

Here it is in its original form and you can extract the TXTSETUP.OEM file from the AsRock package and compare this to the Intel TXTSETUP.OEM I linked.

You will find I made NO MODIFICATIONS of the TXTSETUP.OEM files as shown in the previous post and were copied exactly as is in their original state as they were downloaded from both sites originally when released many years ago.

I have moved the Intel download link here to make it easier for you to download both and compare.

AsRock TXTSETUP.OEM for Floppy Installation
http://asrock.pc.cdn.bitgravity.com/Driv…0.8.0.1003).zip

Intel TXTSETUP.OEM for Floppy Installation
https://downloadmirror.intel.com/20768/e…_f6flpy-x86.zip

After comparing both are you still saying that the TXTSETUP.OEM files are identical and unmodified? I consider this TXTSETUP.OEM from AsRock to be modified to be different than Intel’s whether it was done by Intel or AsRock first there’s no way to tell since neither of us were involved in that process. However the modification date on the AsRock appears earlier than the date released by Intel. This is why I believe Asrock got the driver from Intel first and modified it to suit their motherboard. Whether AsRock had to send the driver modifications back to Intel afterwards and Intel then sends it to Microsoft I wouldn’t know if this ever took place. Explain why there is a 5 month gap between the AsRock and Intel modified dates for the TXTSETUP.OEM for same SATA driver version?

As far as I know not all companies will get a driver WHQL certified for every version released. Even nVidia does not WHQL certify every single version released.


UPDATE:
I did a Binary comparison of both versions (AsRock and Intel) for these files:
IASTOR.SYS
IAAHCI.CAT
IASTOR.CAT

They are bit for bit identical and no modifications were detected. This tells me that altering the TXTSETUP.OEM file does not appear to break the certification.
So the certification seems to only apply to the driver.sys file.

@XPLives :
The file named TXTSETUP.OEM is not a driver, but a simple text file, which can be altered by anyone by using an editor. Nevertheless I doubt, that a mainboard manufacturer would do that. Most likely Intel offers various variants of that file to the OEMs.

No, the digital signature applies to the *.INF file as well and to all driver files, which are listed within the INF file (e.g. coinstaller *.dll files).

No, the digital signature applies to the *.INF file as well and to all driver files, which are listed within the INF file (e.g. coinstaller *.dll files).


Did you do the file comparison? There are no DLL files in this package.

The .INF files are not altered either as I said before only the TXTSETUP.OEM file. Check both files I linked. Every file seems identical when compared aside from the TXTSETUP.OEM so this does not seem to affect the certificate. Obviously the .INF files can’t be tampered seems correct since I have done a lot of this which breaks it.

So are you now considering these two TXTSETUP.OEM files identical and untampered?

@XPLives :
Please stop this absolutely irrelevant and useless discussion!
As you certainly know, the file named TXTSETUP.OEM is only required within the first (TEXTMODE) phase of a fresh installation of old Windows Operating Systems up to Windows XP and Server 2003. Since it is neither a driver nor a signed member of a certain driverpack, it can be altered by everyone.
So I ask you:
1. How relevant is the question, who had altered the text of a certain TXTSETUP.OEM file?
2. What lets you think, that it is/was the mainboard manufacturer, who did it and overtakes the responsability of an installation failure caused by a wrong entry?

I have already answered this question.

I have already answered this question.




I was only showing you proof that this file TXTSETUP.OEM appears altered since you wouldn’t state what kind of proof you needed so you must require the original files with links to the source manufacturer site to believe it and compare it yourself. I cannot find this same identical TXTSETUP.OEM on Intel’s side so this leads me to believe it was only done by AsRock and not Intel.

Yes this is used primarily during the Unattended or Manual Text Mode portion for driver installation. But it does not lose its significance in driver installations. I have found that TXTSETUP.OEM to be quite vital in an Unattended Coffee Lake installation.

I already answered to you why I believe this it was Asrock’s doing. The TXTSETUP.OEM file was purposely parsed down to 4 essential items(2 RAID, 2 AHCI) from the original larger selection found in the Intel version has 15 items.

These 4 items correlate to the Z68 Motherboard’s own SATA Hardware ID and removed the others. That is why I believe it was altered by AsRock and not Intel. Because if not why didn’t Asrock just use the same one on Intel’s site with 15 items instead of provide one with only 4 items?

Anyhow I did not start this discussion. I merely responded to your query and your disbelief that any files could be modified by a MB manufacturer other than Intel. This is the proof I believe shows this is not always true. I cannot see a reason why Intel will specifically waste their time to modify a TXTSETUP.OEM file for every MB manufacturer so you can only conclude since this file can be modified safely without breaking the certification that AsRock did it and not Intel.

I am guessing by your response that you do not feel that AsRock or any MB manufacturer tampers with the Intel driver package. Since you are calling this useless discussion I will leave this alone despite the evidence to the contrary which you asked for proof and we can both agree to disagree on whether this is modified from the MB manufacturer and move on.

Where is the proof, that a mainboard manufacturer has modified an Intel driver? I haven’t yet seen it.

I never claimed Windows Vista/7/2K8 required the TXTSETUP.OEM for USB 3.0 drivers. The TXTSETUP.OEM was proof of modification in an Intel driver package by another manufacturer.

Actually I didn’t open up a discussion I was relabeling the title. The discussion started with your comment and I was addressing it.

[Request] Intel USB 3.0/3.1 support for the newest Intel Chipsets running Win7 (2)
That is nonsense. No mainboard manufacturer has ever touched and will ever touch any driver or INF file delivered by any chipset manufacturer.


Although I have some Intel SATA drivers that do contain the TXTSETUP.OEM file for Windows 7 and 8 OS. So TXTSETUP.OEM files did not die with Windows 2003 Server it seems.

Floppy_RSTe(v4.1.0.1047).zip


# Copyright (c) 2009-2011 Intel Corporation
# *****************************************************************************************************
# *****************************************************************************************************
# ** Filename: txtsetup.oem
# ** Revision: DRIVER_INFO
# ** Abstract: Windows* OEM File for Intel(R)Rapid Storage Technology Enterprise RAIDPort SATA Drivers
# *****************************************************************************************************
#
[Disks]
#
# This section lists all disks in the disk set.
#
# is a descriptive name for a disk, used when
# prompting for the disk
# is a file whose presence allows setup to recognize
# that the disk is inserted.
# is where the files are located on the disk.
#
#
disk1 = “Intel(R) Rapid Storage Technology Enterprise Driver”, iaStorF.sys, \
#
#
[Defaults]
#
# This section lists the default selection for each ‘required’
# hardware component. If a line is not present for a component,
# the default defaults to the first item in the [<component_name>]
# section (see below).
#
# <component_name> is one of computer, display, keyboard, mouse, scsi
# is a unique string to be associated
# with an option.
#
SCSI = iaStorA_C600
#
[scsi]
#
# This section lists the options available for a particular component.
#
# is the unique string for the option
# is a text string, presented to the user in a menu
# <key_name> gives the name of the key to be created for the component in
# HKEY_LOCAL_MACHINE\ControlSet001\Services
#
; iaStorA.inf
iaStorA_C600 = “Intel(R) C600 Series Chipset SATA RAID Controller”
#
; iaAHCI.inf
iaStorA_AC600 = “Intel(R) C600 Series Chipset SATA AHCI Controller”
#
; iaStorS.inf
iaStorS_C600 = “Intel(R) C600 Series Chipset SAS RAID Controller”
iaStorS_SATA0 = “Intel(R) C600 Series Chipset SAS RAID (SATA mode) Controller”
#
[Files.scsi.iaStorA_C600]
driver = disk1, iaStorA.sys, iaStorA
driver = disk1, iaStorF.sys, iaStorF
inf = disk1, iaStorA.inf
catalog = disk1, iaStorA.cat
#
[Files.scsi.iaStorA_AC600]
driver = disk1, iaStorA.sys, iaStorA
driver = disk1, iaStorF.sys, iaStorF
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat
#
[Files.scsi.iaStorS_C600]
driver = disk1, iaStorS.sys, iaStorS
driver = disk1, iaStorF.sys, iaStorF
inf = disk1, iaStorS.inf
catalog = disk1, iaStorS.cat
#
[Files.scsi.iaStorS_SATA0]
driver = disk1, iaStorS.sys, iaStorS
driver = disk1, iaStorF.sys, iaStorF
inf = disk1, iaStorS.inf
catalog = disk1, iaStorS.cat
#
# [HardwareIds.scsi.id]
# corresponds to an id entry in the HwComponent section.
# specifies the device id for a mass storage device.
# specifies the service to be installed for the device.
# The service is specified by the file name of its executable image without a .sys extension.
#
[HardwareIds.scsi.iaStorA_C600]
id = “PCI\VEN_8086&DEV_2922”, “iaStorA”
id = “PCI\VEN_8086&DEV_2826”, “iaStorA”
#
[HardwareIds.scsi.iaStorA_AC600]
id = “PCI\VEN_8086&DEV_1D02”, “iaStorA”
#
[HardwareIds.scsi.iaStorS_C600]
id = “PCI\VEN_8086&DEV_1D60”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D61”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D64”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D65”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D68”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D69”, “iaStorS”
#
[HardwareIds.scsi.iaStorS_SATA0]
id = “PCI\VEN_8086&DEV_1D6A”, “iaStorS”
id = “PCI\VEN_8086&DEV_1D6B”, “iaStorS”
#
# [Config.subkey_name]
# Specifies the name of a key under the Services\DriverKey tree where Windows places the specified value.
# <value_name> specifies the name of the value to be set.
# <value_type> specifies the type of the registry entry.
# The value_type can be one of the following: REG_DWORD, REG_BINARY, REG_MULTI_SZ
# specifies the value; its format depends on value_type.
#
[Config.iaStorA]
value = Parameters\PnpInterface,5,REG_DWORD,1
value = Parameters,BusType,REG_DWORD,8
value = Parameters\Device,EnableQueryAccessAlignment,REG_DWORD,1
#
[Config.iaStorS]
value = Parameters\PnpInterface,5,REG_DWORD,1
value = Parameters,BusType,REG_DWORD,A
value = Parameters\Device,EnableQueryAccessAlignment,REG_DWORD,1
#
[Config.iaAHCI]
value = Parameters\PnpInterface,5,REG_DWORD,1
value = Parameters,BusType,REG_DWORD,B
value = Parameters\Device,EnableQueryAccessAlignment,REG_DWORD,1

To which extent has the TXTSETUP.OEM been a proof for a driver modification by another manufacturer???

The fact, that Win7 and Win8 cannot and will not use this file, is the proof, that the related SATA drivers have not been designed for these Operating Systems. Maybe the chipset manufacturer just forgot to remove the related file.

To which extent has the TXTSETUP.OEM been a proof for a driver modification by another manufacturer???

The fact, that Win7 and Win8 cannot and will not use this file, is the proof, that the related SATA drivers have not been designed for these Operating Systems. Maybe the chipset manufacturer just forgot to remove the related file.




I have no answer to this as I thought you would have more of a clue since you deal with Storage drivers more with newer OS Vista+.

Since MSAHCI is compatible with even Coffee Lake I haven’t had a need to test any Storage drivers. I just found the TXTSETUP.OEM files inside the zip file. I found it mysterious as well and I have not yet attempted an unattended installation of Vista or Windows 7 to see if this works with those files? However a TXTSETUP.OEM might be useful for USB driver integration as I performed an unattended install test with them on XP so it appears it works.

As for the proof of a driver modification by another manufacturer. I’ll explain more in detail in the next message.

Where is the proof, that a mainboard manufacturer has modified an Intel driver? I haven’t yet seen it.




You keep claiming where is the proof? I showed AsRock altered the driver package. The TXTSETUP.OEM files do not match.

Did you even download the two files I linked? Did you examine both TXTSETUP.OEM files contents?

Did they match? Are they 100% identical? Are the file byte sizes identical?

You stated that no motherboard manufacturer ever modified an Intel driver package.

Are the TXTSETUP.OEM files the same from the Asrock and the Intel driver links for Floppy Driver Installation of the same version?

I never stated any driver.sys file was altered. You stated nothing would ever get modified from what Intel has on their site and I showed you that the TXTSETUP.OEM files do not match.

So are you now claiming Intel did those modifications specifically for AsRock for their Z68 MB? That Intel filtered down 15 items to 4 items in the TXTSETUP.OEM themselves?

Are these TXTSETUP.OEM files identical between AsRock and Intel? If not there is no further discussion because if they were identical then that would prove nothing was modified by AsRock and we wouldn’t be discussing this. Prove to me AsRock didn’t modify the TXTSETUP.OEM file and how would Intel know which IDs to filter out to match the AsRock specific motherboard? Are you going to state that Intel modified each and every TXTSETUP.OEM for every AsRock line of motherboards? Why are the modified dates different by 5 months aside from the TXTSETUP.OEM and file layouts being internally altered? Wouldn’t it make more sense for AsRock to use Intel’s identical unmodified version?

Again if your sense is mismatching TXTSETUP.OEM files doesn’t prove it’s been modified then we will have to agree to disagree on our ideas of modification.

@XPLives :
ASRock is offering Intel’s drivers and not ASRock’s drivers.
Why do you insist claiming, that ASRock has modified Intel’s drivers? From whom do you know, that the modification hasn’t been done by Intel?



AsRock is offering Intel’s drivers but they have modified the TXTSETUP.OEM file? You haven’t responded did you inspect each file I linked to each original website Asrock and Intel?

If so did you examine that the TXTSETUP.OEM files do not match?

Again what makes you think Intel will modify every single TXTSETUP.OEM file to fit every motherboard of AsRock, or other manufacturer when they don’t have to?

What makes you think AsRock didn’t modify the TXTSETUP.OEM to suit their motherboard after Intel provided them the base driver package. Modifying the TXTSETUP.OEM file as you also noted shouldn’t invalidate the certificate so wouldn’t it make more sense that the manufacturer would alter this file than Intel doing it themselves for every MB manufacturer who wanted to cater to their motherboard HW IDs?

If you’re not even examining the TXTSETUP.OEM files I gave you links to and comparing them then you’re not even analyzing the facts. So it would be pointless to try and prove to you they were modified by AsRock since you refused to look at the two files and compare the two file packages for the differences.

An educated guess would be the driver on Intel’s site was originally given to AsRock 5 months prior. AsRock modified the TXTSETUP.OEM file and suited their MB’s HW IDs. If AsRock had all 15 HW IDs I would think they might have just added a ; to comment the others out or just left all 15 HW IDs present to support future motherboards. So maybe this is what really happened. They didn’t get all 15 HW IDs but a blank slate. AsRock just filled in the ones for their MB and called it a day.

I can turn this around and ask how do you know AsRock would never touch the TXTSETUP.OEM file with 100% certainty? Unless you have some inside information or work for Intel / AsRock how would you know who modified that TXTSETUP.OEM file? Why would the TXTSETUP.OEM date from AsRock predate Intel’s official version by 5 months?

Why didn’t AsRock wait 5 months and just use the one from Intel’s site instead? Or if you’re saying Intel did all the modification and if this were true why did Intel even bother wasting their time to modify it for AsRock 5 months before releasing it on Intel’s own site? Wouldn’t it have saved everyone plenty of time if Intel released the only driver unmodified and let every manufacturer host the same file on their own server? So all files hosted match from all manufacturers to Intel’s with no changes? As long as it contained every Hardware ID out there then it shouldn’t matter where you got the driver so Intel could have not modified any TXTSETUP.OEM file assuming they did it according to you.

Maybe we will never know the truth. But the modified date time stamp shows AsRock had it 5 months before Intel released the driver on their site.

If Intel does the modification why didn’t they release it sooner than 5 months if they already had all the Hardware IDs ready?

A driver really only consists of binary components, the .sys file, an .inf guide, and possibly helper items like dynamic link libraries (with the main driver in the .sys file.) This “OEM” file sounds like some extra metadata file included in ASRocks’s driver bundle, and as the .sys file is identical to that provided by Intel, ASRock did not modify the driver regardless of whatever they did to the OEM file.

This would be like saying they modified the driver by changing a word in the provided readme.txt.

I’ve never heard of anything aside from the .sys file having a provided certificate .cat file either, the OEM file would not need to be resigned as it wasn’t hashed/verified to begin with as it is not part of the driver binary.

You only use the drivers for parts of the motherboard, things like the chipset, USB controller, and so forth - you could feasibly have multiple “motherboard IDs” which use the same set of drivers, and the IDs provided in this OEM file are probably intended as a guide to see if the driver set for the drivers themselves to see if they need to even be installed.

@Omicron :

When using the Device Manager manual driver installation method the TXTSETUP.OEM will not be required because you are already inside the OS. All you will need is the .INF and the .SYS and whatever files the .INF lists required for the device. Modifying the .INF will break the certification but in some ways this is the only way to compact the drivers list needed to be installed for the device to work. Often times they add files that are not required. I was able to reduce a couple hundred MBs of files of space used by an nVidia driver down to less than 20MBs compressed for driver installation. So there was tremendous bloat in nVidia installer packages.

I don’t think you have a firm enough grasp of 2000/XP/2003 F6 installation methods regarding the TXTSETUP.OEM as it used to load drivers during the initial OS installation stage using a Floppy Disk Drive. Fernando may not even be familiar with using this method either as I doubt he uses floppy disk drive equipped systems which is why nLite is the easier solution for beginners.

The TXTSETUP.OEM is not a simple readme.txt file. If it were it would have no effect on the F6 driver installation process and you could remove it. If you attempt this it will error if not present.

Nlite performs a similar automated method of this but it does some unnecessary modifications of certain OS files and is why it’s usually used by beginners. More advanced skills are required to use the TXTSETUP.OEM method. I had to perform close to a hundred different test installs to get a better sense of how it works during Unattended install experiments.

The TXTSETUP.OEM file is needed during the F6 stage to install a driver through a Floppy disk drive or a USB Floppy disk drive.

The TXTSETUP.OEM offers a selection menu of devices which can be customized. But without it you could not install the driver using the F6 method or could also be used to perform an Unattended Installation with the drivers detected during the OS install process. If you only had the .INF and the .SYS files it would not be enough as it will continue to request the TXTSETUP.OEM file.

I never said AsRock modified the actual device driver as AsRock would never modify the actual driver or it would break the certification and add more work for Intel or MS to recertify it and highly doubtful Intel would “EVER” release the source code of their Intel SATA driver to other companies unless they had a NDA. I said AsRock modified the driver package that differed from the Intel one which you clearly are not understanding. The packages are different regarding the TXTSETUP.OEM files and the layout of the files inside it as well. My point was AsRock modified this TXTSETUP.OEM and not Intel which Fernando either disagrees with or doesn’t understand and thinks Intel did it. So I’ll have to clearly spell it out further as the proof is already provided and maybe unaware.

If you examine the two TXTSETUP.OEM files:

The AsRock one will show these items to choose from during the F6 setup process.

[Actual ASROCK List First Seen]
Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP <— DEFAULT SELECTION (aAHCI32_DesktopWorkstationServer)
Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP64
Intel(R) Desktop Express Chipset SATA RAID Controller - Windows XP
Intel(R) Desktop Express Chipset SATA RAID Controller - Windows XP64


[Defaults]
scsi = iaAHCI32_DesktopWorkstationServer <— This line in the AsRock TXTSETUP.OEM determines what default device is if first selected in the F6 Setup Menu once the Floppy Disk has read the driver info.

The AsRock default refers to: Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP
PCI\VEN_8086&DEV_1C02&CC_0106

The Intel default refers to: Intel(R) Mobile Express Chipset SATA RAID Controller
PCI\VEN_8086&DEV_282A&CC_0104

The Intel match to Asrocks calls it Intel(R) Desktop/Workstation/Server Express Chipset SATA AHCI Controller

—Lining them up---------------------------------------------------------------------------------------------------------------
You can now compare the two:
Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP <— AsRock description
Intel(R) Desktop/Workstation/Server Express Chipset SATA AHCI Controller <— Intel description
-----------------------------------------------------------------------------------------------------------------------------------

Intel would never add the words “- Windows XP” or “- Windows XP64” to their device description for the items in the TXTSETUP.OEM found on Intel’s website.
And you can see in the AsRock version the words “/Workstation/Server” have been removed.
This clearly shows you that the Asrock version is different from the Intel version. The Hardware IDs are also completely different for the default selection settings.


The reason why you see this in my opinion it was modified specifically by AsRock for their internal floppy drive F6 installation. They purposely filtered down 15 items found in the Intel’s original version and parsed it down to just these 4 or if they had access to the driver earlier on 5 months prior then they only added their motherboards 4 listed options and Intel added the rest to their own TXTSETUP.OEM later to support more:
1 for XP 32-Bit AHCI
1 for XP 64-Bit AHCI

1 for XP 32-Bit RAID
1 for XP 64-Bit RAID

During the F6 stage driver installation you will only see these four items listed to choose from listed above and the rest are not selectable or shown.

On the Intel’s original Floppy setup it will be different and will have 15 options total.

[Defaults]
scsi = iaStor_Mobile <— This line in the Intel TXTSETUP.OEM determines what default device is if first selected in the F6 Setup Menu once the Floppy Disk has read the driver info.

[THIS IS THE MENU LIST BELOW]
Intel(R) ICH7R/DH SATA AHCI Controller
Intel(R) ICH7M/MDH SATA AHCI Controller
Intel(R) ICH9R/DO/DH SATA AHCI Controller
Intel(R) ICH9M-E/M SATA AHCI Controller
Intel(R) ICH10D/DO SATA AHCI Controller
Intel(R) ICH10R SATA AHCI Controller
Intel(R) 5 Series 4 Port SATA AHCI Controller
Intel(R) 5 Series 6 Port SATA AHCI Controller
Intel(R) 5 Series/3400 Series SATA AHCI Controller
Intel(R) Desktop/Workstation/Server Express Chipset SATA AHCI Controller
Intel(R) Mobile Express Chipset SATA AHCI Controller
Intel(R) ICH7R/DH SATA RAID Controller
Intel(R) ICH7MDH SATA RAID Controller
Intel(R) Desktop/Workstation/Server Express Chipset SATA RAID Controller"
iaStor_Mobile = “Intel(R) Mobile Express Chipset SATA RAID Controller” <— DEFAULT SELECTION
Since the DEFAULT item is located at the very bottom of the file you should see these four items listed initially and would have to scroll up to see the rest.


[Actual INTEL List First Seen]
Intel(R) ICH7R/DH SATA RAID Controller
Intel(R) ICH7MDH SATA RAID Controller
Intel(R) Desktop/Workstation/Server Express Chipset SATA RAID Controller

"Intel(R) Mobile Express Chipset SATA RAID Controller" <— DEFAULT SELECTION (iaStor_Mobile)


Since the Default for the Intel version is iaStor_Mobile is for “Intel(R) Mobile Express Chipset SATA RAID Controller”.
If the user was using this on the AsRock Z68 Desktop Hardware ID and the user were to hit ENTER and select this item the installation would fail with the BSOD 7B early on.

This is why the AsRock version appears modified by AsRock for the intent of selecting the correct matching Hardware ID by default for a Windows XP 32-Bit SATA AHCI driver installation for their Z68 series motherboard. This filtering down of other Hardware devices from being shown was by “intent” from AsRock. Intel would never customize the TXTSETUP.OEM in this manner for a Floppy disk installation. If you observe all other Intel Floppy Based SATA driver packages they would never add “- Windows XP” or “- Windows XP64” to any of their items and the default highlighted initial item would most likely will be iaStor_Mobile which would highlight “Intel(R) Mobile Express Chipset SATA RAID Controller”.

Since I’ve done numerous F6 Floppy disk installations over the years and even modified the TXTSETUP.OEM there is no other reasonable conclusion as to why the Asrock version vs the Intel Floppy driver packages are different.

If you are still not convinced you could try installing the Intel SATA driver package yourself and use a USB floppy on a Sandy Bridge Z68 system during the F6 stage and use the default Intel(R) Mobile Express Chipset SATA RAID Controller and it will give you a BSOD 7B.

Now if you had a matching Z68 Hardware Id motherboard which hopefully it should be the same for another Z68 motherboard and repeated the process with the AsRock Floppy SATA driver it should work without a BSOD 7B.

If you can list any Intel Floppy based SATA storage driver off Intel’s own site that shows Intel doing this same exact kind of TXTSETUP.OEM modification adding - Windows XP and - Windows XP64 to the device description I’d like to see it. Intel would never add the OS to the name of the device in this manner and I found another case of this happening in another AsRock MB chipset Floppy SATA driver package. There might be other Floppy packages from Asus and Asrock with modified TXTSETUP.OEM files like this since they are related companies but not on Intel’s own site.

Also one other difference I noticed was the Intel 64-Bit Driver package is isolated just like the Intel 32-Bit Driver package with their own individual TXTSETUP.OEM file and not shared. Only the AsRock package combines both the XP 32-Bit and XP 64-Bit driver selectable options into one convenient TXTSETUP.OEM file. Only AsRock could have modified it this way. I could not find any Intel XP 32-Bit and XP 64-Bit TXTSETUP.OEM integrated together like this on Intel’s site.

By using this TXTSETUP.OEM method of Unattended installation it also allowed me to get Intel USB 3.0 controllers working in XP for the first time never done by anyone else. Sometimes the hardest way turns out is the best way…



[BREAKDOWN OF THE ASROCK VS INTEL TXTSETUP.OEM FILES]
http://asrock.pc.cdn.bitgravity.com/Driv…0.8.0.1003).zip
[Asrock] TXTSETUP.OEM v10.8.0.1003
Monday, May 09, 2011, 4:18:10 PM
2.43 KB (2,493 bytes)



; Copyright (c) 2003-11 Intel Corporation
;#############################################################################
;#
;# Filename: TXTSETUP.OEM
;#
;#############################################################################
[Disks]
bit32 = “Intel(R) Rapid Storage Technology Driver”, \i386\iaStor.sys, \i386
bit64 = “Intel(R) Rapid Storage Technology Driver”, \AMD64\iaStor.sys, \AMD64


[Defaults]

scsi = iaAHCI32_DesktopWorkstationServer

;#############################################################################
[scsi]

; iaAHCI.inf
iaAHCI32_DesktopWorkstationServer = "Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP"
iaAHCI64_DesktopWorkstationServer = “Intel(R) Desktop Express Chipset SATA AHCI Controller - Windows XP64

; iaStor.inf
iaStor32_DesktopWorkstationServer = "Intel(R) Desktop Express Chipset SATA RAID Controller - Windows XP"
iaStor64_DesktopWorkstationServer = “Intel(R) Desktop Express Chipset SATA RAID Controller - Windows XP64


;#############################################################################

; iaAHCI.inf
[Files.scsi.iaAHCI32_DesktopWorkstationServer]
driver = bit32, iaStor.sys, iaStor
inf = bit32, iaAHCI.inf
catalog = bit32, iaAHCI.cat

[Files.scsi.iaAHCI64_DesktopWorkstationServer]
driver = bit64, iaStor.sys, iaStor
inf = bit64, iaAHCI.inf
catalog = bit64, iaAHCI.cat

; iaStor.inf
[Files.scsi.iaStor32_DesktopWorkstationServer]
driver = bit32, iaStor.sys, iaStor
inf = bit32, iaStor.inf
catalog = bit32, iaStor.cat

[Files.scsi.iaStor64_DesktopWorkstationServer]
driver = bit64, iaStor.sys, iaStor
inf = bit64, iaStor.inf
catalog = bit64, iaStor.cat



;#############################################################################
[Config.iaStor]
value = “”, tag, REG_DWORD, 1b
value = “”, ErrorControl, REG_DWORD, 1
value = “”, Group, REG_SZ, "SCSI Miniport"
value = “”, Start, REG_DWORD, 0
value = “”, Type, REG_DWORD, 1

;#############################################################################

; iaAHCI.inf
[HardwareIds.scsi.iaAHCI32_DesktopWorkstationServer] <— Default Hardware ID for AsRock TXTSETUP.OEM
id = “PCI\VEN_8086&DEV_1C02&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI64_DesktopWorkstationServer]
id = “PCI\VEN_8086&DEV_1C02&CC_0106”,“iaStor”

; iaStor.inf
[HardwareIds.scsi.iaStor32_DesktopWorkstationServer]
id = “PCI\VEN_8086&DEV_2822&CC_0104”,“iaStor”

[HardwareIds.scsi.iaStor64_DesktopWorkstationServer]
id = “PCI\VEN_8086&DEV_2822&CC_0104”,"iaStor"




https://downloadmirror.intel.com/20768/e…_f6flpy-x86.zip
[Intel] TXTSETUP.OEM v10.8.0.1003
Monday, October 17, 2011, 3:34:12 PM
5.19 KB (5,322 bytes)


; Copyright (c) 2003-11 Intel Corporation
;#############################################################################
;#
;# Filename: TXTSETUP.OEM
;#
;#############################################################################
[Disks]
disk1 = “Intel(R) Rapid Storage Technology Driver”, iaStor.sys, <br />
[Defaults]
scsi = iaStor_Mobile

;#############################################################################
[scsi]

; iaAHCI.inf
aAHCI_7RDH = "Intel(R) ICH7R/DH SATA AHCI Controller"
iaAHCI_7MMDH = "Intel(R) ICH7M/MDH SATA AHCI Controller"
iaAHCI_9RDODH = "Intel(R) ICH9R/DO/DH SATA AHCI Controller"
iaAHCI_9MEM = "Intel(R) ICH9M-E/M SATA AHCI Controller"
iaAHCI_10DDO = "Intel(R) ICH10D/DO SATA AHCI Controller"
iaAHCI_10R = "Intel(R) ICH10R SATA AHCI Controller"
iaAHCI_5 = "Intel(R) 5 Series 4 Port SATA AHCI Controller"
iaAHCI_5_1 = "Intel(R) 5 Series 6 Port SATA AHCI Controller"
iaAHCI_5_1_1 = "Intel(R) 5 Series/3400 Series SATA AHCI Controller"


iaAHCI_DesktopWorkstationServer = “Intel(R) Desktop/Workstation/Server Express Chipset SATA AHCI Controller” <— This one matches closest to the one used by AsRock’s


iaAHCI_Mobile = “Intel(R) Mobile Express Chipset SATA AHCI Controller

; iaStor.inf
iaStor_7RDH = "Intel(R) ICH7R/DH SATA RAID Controller"
iaStor_7MDH = "Intel(R) ICH7MDH SATA RAID Controller"
iaStor_DesktopWorkstationServer = "Intel(R) Desktop/Workstation/Server Express Chipset SATA RAID Controller"
iaStor_Mobile = “Intel(R) Mobile Express Chipset SATA RAID Controller”

;#############################################################################

; iaAHCI.inf
[Files.scsi.iaAHCI_7RDH]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_7MMDH]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_9RDODH]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_9MEM]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_10DDO]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_10R]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_5]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_5_1]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_5_1_1]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_DesktopWorkstationServer]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat

[Files.scsi.iaAHCI_Mobile]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaAHCI.inf
catalog = disk1, iaAHCI.cat


; iaStor.inf
[Files.scsi.iaStor_7RDH]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaStor.inf
catalog = disk1, iaStor.cat

[Files.scsi.iaStor_7MDH]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaStor.inf
catalog = disk1, iaStor.cat

[Files.scsi.iaStor_DesktopWorkstationServer]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaStor.inf
catalog = disk1, iaStor.cat

[Files.scsi.iaStor_Mobile]
driver = disk1, iaStor.sys, iaStor
inf = disk1, iaStor.inf
catalog = disk1, iaStor.cat


;#############################################################################
[Config.iaStor]
value = “”, tag, REG_DWORD, 1b
value = “”, ErrorControl, REG_DWORD, 1
value = “”, Group, REG_SZ, "SCSI Miniport"
value = “”, Start, REG_DWORD, 0
value = “”, Type, REG_DWORD, 1

;#############################################################################

; iaAHCI.inf
[HardwareIds.scsi.iaAHCI_7RDH]
id = “PCI\VEN_8086&DEV_27C1&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_7MMDH]
id = “PCI\VEN_8086&DEV_27C5&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_9RDODH]
id = “PCI\VEN_8086&DEV_2922&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_9MEM]
id = “PCI\VEN_8086&DEV_2929&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_10DDO]
id = “PCI\VEN_8086&DEV_3A02&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_10R]
id = “PCI\VEN_8086&DEV_3A22&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_5]
id = “PCI\VEN_8086&DEV_3B29&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_5_1]
id = “PCI\VEN_8086&DEV_3B2F&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_5_1_1]
id = “PCI\VEN_8086&DEV_3B22&CC_0106”,“iaStor”

[HardwareIds.scsi.iaAHCI_DesktopWorkstationServer]
id = “PCI\VEN_8086&DEV_1C02&CC_0106”,“iaStor” <— This one matches closest to the one used by AsRock’s

[HardwareIds.scsi.iaAHCI_Mobile]
id = “PCI\VEN_8086&DEV_1C03&CC_0106”,“iaStor”


; iaStor.inf
[HardwareIds.scsi.iaStor_7RDH]
id = “PCI\VEN_8086&DEV_27C3&CC_0104”,“iaStor”

[HardwareIds.scsi.iaStor_7MDH]
id = “PCI\VEN_8086&DEV_27C6&CC_0104”,“iaStor”

[HardwareIds.scsi.iaStor_DesktopWorkstationServer]
id = “PCI\VEN_8086&DEV_2822&CC_0104”,“iaStor”

[HardwareIds.scsi.iaStor_Mobile]
id = “PCI\VEN_8086&DEV_282A&CC_0104”,“iaStor” <— Default Hardware ID for Intel TXTSETUP.OEM


EDIT by Fernando: To minimize the length of the post and to improve its readability, I have
a) replaced the unneeded fully quoted post by a directly addressing and
b) put the content of the 2 text files into “spoilers”.

No???
As Omicron pointed out very clearly, the TXTSETUP.OEM is neither a driver nor part of the digitally signed driver package. So how could ASRock modify an Intel driver without breaking its signature?

All drivers, which are or have been offered on Intel’s “Download Center”, were/are designed for mainboards, which were manufactured by Intel.
From whom do you know, what sort of drivers Intel has released to the other mainboard manufacturers like ASRock?

JFYI: Simple text files like the TXTSETUP.OEM can be used with 32bit and 64bit Operating Systems. The text can be absolutely identical.

Ah, I think we just have a case of people being on different pages here.

No, I’m not highly familiar with the old driver install process of the “PE” text mode environment of Windows XP/2003 (5.1/5.2) other than using it once to load an AHCI driver. I pretty much ditched 2003 when Windows 7 came out as there was finally a reason to upgrade - the 5.X kernel series became antiquated to me.

I was just saying to Windows 7, something like that OEM file is comparable to a text file as it isn’t even touched to my knowledge - and that my discussion comes from a higher familiarity with newer 6.X drivers.

Now, if a manufacturer managed to actually modify a .sys binary driver file, that would be “modification of the driver” per my experience.

I’m actually sort of curious, there are xHCI/USB 3.0 drivers that actually work on XP? I had no idea anyone did/tried that.