Although Intel has chosen not to provide official drivers for Windows 7 for the NUC6CAYH there is anecdotal information on the internet that all hardware can be made to work with no issues left in Device Manager. Apparently all drivers needed exist, it's just that they are not (yet) listed comprehensively and (made) available to the owners of the NUC6CAYH.
I can only guess to the reasons for this choice. Maybe Intel felt getting all the drivers WHQL certified by Microsoft was too... ??? for an OS getting near to the end of its official support?
Personally I'd be willing to settle for (some) non-WHQL drivers. As long as they work. Surely Intel has enough people in its employ with more than enough experience in making or assembling drivers that offering a complete set of drivers isn't too much of a problem. Maybe mention that prolonged support for these drivers cannot be guaranteed for the above reason. I think I wouldn't be the only one who'd appreciate being allowed our own free choice of which (still under active or "extended" support by Microsoft) Windows version we'd prefer to use. I'm also not the only one with reservations regarding Windows 10.
Please don't engage in a "Why not Windows 10?" discussion. It's off-topic here. This is about getting Windows 7 to work. There are enough other places for such a discussion.
I've gotten quite far finding working drivers. It's just that the reports about getting it all to work are a tad incomplete and I'm still left with a few issues in Device Manager.
The initial problem to get Windows 7 installed on this device with only USB 3 ports can be solved with the "ASRock Win 7 USB Patcher" with which an installation medium can be made that supports USB 3, which Windows 7 doesn't support natively. After the installation the most important drivers can be used from sources like ASUS, Dell, Lenovo and Intel itself too. If there is enough interest I can give some additional information and links.
At the moment I have a few issues left and I hope people can point me to the right drivers to get these last few parts of the hardware to function properly.
In the "Other Devices" section are 4 items left with (!).
PCI Data Acquisition and Signal Processing Controller
I understand this to be the: "Intel(R) Serial IO I2C Host Controller".
PCI Simple Communications Controller
I understand this to be the: "Intel(R) Trusted Execution Engine Interface"
Apparently both belonging to the "Advanced Configuration and Power Interface".
In the "System Devices" section are 3 items left with (!).
Intel(R) Atom(TM)/Celeron(R)/Pentium(R) Processor SPI Controller (x3)
And in the "Network Adapters" section there is one (!).
Intel(R) Dual Band Wireless-AC 3168
This device cannot find enough free resources that it can use. (Code 12)
If you want to use this device, you will need to disable one of the other devices on this system.
Somewhat unexpected. I installed drivers for this. I thought things like interrupt sharing were resolved quite some time ago. I hope and kind of expect this will be resolved by the installation of one of the still missing drivers that will take care of this.
This forum seems the most logical place to address this issue. Hopefully this community can offer some solutions. A complete list of working drivers would be nice. Getting Intel to make these available on their site would be even nicer.
Thanks in advance all.
Hello, Bunsen Honeydew. Thank you very much for sharing your issue with the Intel Communities Team. I will be more than glad to assist you.
In this case, as you know, the only OS supported on the NUC6CAYH is Windows 10* (https://www.intel.com/content/www/us/en/support/articles/000005628/mini-pcs.html Supported Operating Systems for Intel® NUC Products). Neither Intel® nor Microsoft* is providing support for Windows 10*. Please keep in mind that the results of installing this OS on the NUC may vary, this is why you are getting those errors with the drivers.
Intel is telling us in the product brief:
"While the NUC6CAYH is designed for Windows 10 you have the choice to install the OS, memory, and hard drive you want..."
Source: https://www.intel.co.uk/content/dam/www/public/us/en/documents/product-briefs/nuc-kit-nuc6cayh-brief... https://www.intel.co.uk/content/dam/www/public/us/en/documents/product-briefs/nuc-kit-nuc6cayh-brief...
Which should allow for more in terms of Windows versions than just Windows 10. The errors are not the result of the already installed drivers but they are the result of missing drivers. I would expect somebody at Intel would have the knowledge to point me to suitable drivers which will work with this particular hardware. That would help.
What Intel means by this is that you can choose to install Linux on the NUC if you so desire. It does not, however, imply that earlier versions of Windows will install and run cleanly. They won't. Intel firmly states this. Intel is not, and will not be, developing drivers to support earlier versions of Windows. Nor, where paths cross, will Intel be validating that existing drivers will work on these earlier versions of Windows.
Have you read Microsoft's rationale paper behind why they won't support the older versions of Windows on newer hardware? It is the problems with supporting SoC packages -- like what is used in this particular NUC -- that drove this decision. It is a valid point; it is getting too difficult to provide the emulation that is necessary to allow the older O/Ss to run on these packages. In the case of Windows 7, you are talking about an O/S that is almost 10 years old. PC Architectures have changed *significantly* since then. I too am a hater of many of the things that Microsoft has done in Windows 10 but I also have to be realistic. Expecting the HALs, which were designed to support PCs as they existed in 2009, to continue to cleanly function on modern architectures is bordering on the ridiculous.
Intel has a number of hard and fast rules. One is that nothing will be released that is outside of the plan of record (POR). Another is that nothing will be released that has not gone through formal validation (which won't happen if it's outside the POR). If it is not tested, it is not supported and it is not released. One of the things that I learned (the hard way) while at Intel was that, if you throw something over the wall firmly stating that it is totally unsupported and that you will use it at your own risk, customers will *still* have an expectation that, if problems occur, Intel will fix them. Well, because of this, Intel will simply not allow anything to be thrown over the wall. Windows 7 is not supported, period. If you don't like Windows 10, run Linux. If you absolutely want to run Windows 7, purchase an older architecture (there are still some NUC5CPYH and NUC5PPYH units around).
People, for some unfathomable reason, seem to think that, because Intel is this big company, they have resources that they can throw willy nilly at any of these issues and come up with a solution tomorrow. This simply isn't possible and it is ludicrous to think so. No organization has the ability to pivot on a dime like that. There's never enough free resources, with the right training (the rub of the issue), that you can simply throw them at a problem.
Define "over the wall".
As I said, all needed working drivers already exist since it's apparently possible to get all exclamation marks removed from Device Manager and get all the hardware working. Most of it I already have working myself. Intel is in the best position to know which currently existing drivers can work, or work best. No drivers need to be developed.
Maybe testing and supporting such drivers is the main motivation. Maybe Intel feels it wants to support Microsoft in getting everyone to switch to Windows 10 and because of that tries to "discourage" using Windows 7. If Intel still values and respects its customer's wishes it would do them credit if they would make some effort to serve their customers in this to some extent.
Surely sufficient ways exist to provide people with a complete list of links with working drivers (not necessarily hosted on an Intel site) through some unofficial channels like the forum of some large and well known tech-sites. Someone from Intel (or close enough to have the relevant knowledge) could make a private account there and post some relevant information purely as a private person without any obligations on the part of Intel. Just in a: "Hey people. Look what I have found out." way. Undoubtedly such information will find its way to the interested customers without too much trouble. All that is needed is a little bit of willingness to provide the customers with the information they want and seek, and a little creativity to think just a little bit outside of the box. Nothing really difficult about that.
Hmm... maybe a new site... "driverleaks.com". It doesn't exist yet. I wonder how long it will take before someone registers it after reading this.
I started to respond to this a dozen different ways, but decided to discard them all. I am done trying to rationalize this for you. The decisions regarding Windows 7 support have been made and will not change. As for why, think what you want. believe what you want, I really don't care. Intel does value and respect its customer's wishes and the wishes of the *vast* majority of their customers *are* being met. End of story. End of conversation.
I've been trying to install Windows 7 on NUC6CAYH for days now. And yes, I understand that it's not supported. With the help of Win 7 USB patcher from ASRock I was able to get keyboard and mouse working in Windows 7 installation program for the first time.
The problem is that when Windows boots for the first time from hard drive, it's still lacking USB 3.0 support and keyboard and mouse won't work. This is when installation is still not finished and it's asking for my name. I'm very interested if someone has found a solution.