There were at least two driver updates since I started having the problem - the most recent update was 8151 - but I have to roll all the way back to version 7529 (dated 11/13/2019) to get rid of the problem.
I have a support case open. The technician said the case would be elevated, then he ignored further emails from me and I haven't heard anything for ages. I have just updated the support case which has had no Intel input since 3 July 2020. Intel's support is really poor.
Sorry to hear about the repeated problems, but I have to report... I have not seen ANY problems with sleep/wake since I went to build 8587 (and then to 8629). It might be worth looking into external factors that may be contributing - in my example there was an issue with the CyberPower Power Panel driver that was preventing sleep.
I can also add that very often simply fixing the driver itself isn't sufficient or possible, so there may be dependencies between the newest driver and the newest BIOS (just for example)... so it is worth making sure that everything is completely updated before concluding the new driver "doesn't work".
I'm not an Intel fanboy nor an employee - but I do work for another large software company, and these issues are often more complicated than they appear. From my perspective, I posted to a community forum, not expecting Intel to respond, but they did, and opened a support case for me, and after a little prodding they got me a build to test and followed up to check on my progress. That's better than I'd get from Microsoft or Apple (typically anyway) so I can't really get on board with all the criticism.
Not to mention it isn't helpful.
Yes, and that was very irritating the way they took it to private message with you and left the rest of us in the dark. I cannot accept such excuses when the hardware is Intel's as is the driver. Sending a signal to wake the monitor is pretty basic. I'm sure if they were bothered they could find a solution. I'm running nothing particularly unusual or esoteric. Just the basic apps.
The case I opened in mid-August was handled very expeditiously. I was sitting in my daughter's dorm room, realized the monitor was not responding as it should after I'd set up the system, and then connected to Intel support over chat. The agent took over the NUC & ran through a number of checks rather quickly. He then indicated that Intel would replace the NUC if necessary. It was just a week later I decided to take that step.
As far as testing the NUC firsthand next week, her college is presently barring dormitory visitors so I guess that's out.
So I did get a case open, they sent me the generic 'wake problems' troubleshooting page. I actually did do the BIOS recovery by using the jumper inside, the instructions say this clears HDCP keys and some other things, so it seemed worth a try. I noticed that in my BIOS settings I had HDMI CEC enabled, I have since tried disabling that. So far, no change in anything. I was having a lot of problems and noticed I was on the 8681 driver, not sure if I had installed it or if Windows bumped me up to it from 7529. There's also an 8691 build now but it's dated older than 8681 so have not yet tried it.
I also experimented with the HDMI firmware 1.73 and noticed no difference, so I am now back on 1.77.01 which was the recommended version for HDMI to HDMI connections.
A few days ago I added another message to my support case 04678907 with Intel and still no reply. Intel has not added anything since 3 July. Really poor for a support case.
I decided to lock in the working 7529 graphics driver and prevent Windows from updating it. This driver works perfectly and later drivers do not, so by following the instructions here the driver cannot be updated. Please note this means even a manual install will fail, so keep a note of the instructions so you can easily reverse them should Intel get itself into gear and show that it cares.
@SSuss What are the details of your setup? Maybe your original problem is not the same root cause as the rest of us. Are you running a 4k monitor with HDMI to HDMI? I noticed my problems when I upgraded to Win 10 20H1 (2004), but I am not convinced that in itself was the cause.
I'm running Windows 20H2 now. HDMI to HDMI, Dell U2718Q. Two HDDs installed, NVMe WD Black SN750 (boot) and SATA SSD Samsung 850 Pro (data).
Mmmmm other things attached to it... Microsoft Classic Intellimouse (USB)... a 1987 IBM Model M keyboard via USB/PS2 Soarer's adapter. Nothing exotic.
I can tell within minutes if a driver is going to cause problems. If I replace my HDMI->HDMI cable with a USB C->Displayport cable the monitor will go through the motions of sleeping after whatever time is specified in Windows settings but will immediately wake up, then try and sleep and wake up in a continuous cycle. I know if I experience this issue then even though substituting the HDMI cable allows it to sleep normally the monitor will not wake up in the morning. 7529 behaves perfectly with either cable.
I had the problem before I updated to 2004.
My monitor is an LG UHD 24UD58.
So my driver updated to 126.96.36.19981 again, must be this is what Windows Update is pushing out now. I noticed in the SSU output I generated for another issue that in there it says 188.8.131.5291 in one place and 8681 in another place, kind of strange. I'm not quite to the point of denying automatic updates for the display driver, I'd like to see if this gets resolved.
So I mentioned that the 7529 driver works in my support case, so they updated the case saying ok keep using 7529 and closed the case! Such awesome customer service. They did say that they had referred this to some other group, so maybe it is the other group that closed the ticket. Really disappointing how Intel handles this.
BTW Windows Update does provide 8681 now, my luck with that one has been hit or miss. I had a good streak of about 3 days where it seemed to work well and then it started failing again. I also took the preview update to Win10 20H2, hoping that would help but of course it didn't. Just for fun, I removed the display driver entirely back to the Microsoft basic display driver and let Windows install 8681 instead of using the Intel installer. Have not yet tried the latest beta drivers, maybe there is still hope.
Despite this conversation taking a different path, I want to come back to the USB problem.
My last adjustment was to disable (in device manager - USB devices - Allow the computer to turn off this device to save power) power management.
Also in Powersettings i've disabled the USB Selective Suspend Settings.
For now, 2 weeks later, neither of the 2 NUC's i've adjusted reported the issue.
One problem about the USB selective supend setting: Not possible via GPO
I've been on a good streak for a few days now on the 8681 driver version installed by Windows Update, interesting thing, I tried to install the newly released 8783 (non-beta) driver and it told me I had the same version installed already. Looking in device manager and running the Intel SSU it shows 8681 though with a date of 9/4. I guess I am going to leave well enough alone for now but unless the 8681 version pushed by Windows is different than the one others have used I don't have a lot of faith that this is fixed.
I re-opened my case and they re-closed it saying there is not yet a fix available, here's the exact response:
"We don't have a driver that can fix the issue at this point but can continue checking the Downloadcenter.intel.com on the regular basis to see when we come up with a new driver for this NUC."
EDIT: I just searched the NUC8i7BEH drivers and they are now listing 8681 as available for the NUC and it is certified against Win10 20H1......maybe the NUC version is customized in some way vs the generic driver? For the longest time, 8141 was listed as the supported version for this particular NUC.
I have a NUC7CJYH, connected to a LG 24UD58-B monitor, same as the posters PDeac1 & MikeBurrito have. The NUC is 2 years old and has always had occasional instances in which the monitor has failed to detect a signal after the NUC wakes up. Now it is a constant problem. I live in NJ and my daughter uses the NUC at college in Ohio.
I had Case# 04769622 opened with Intel on August 13 when I was in my daughter's dorm room. The tech took over the NUC and ran through a number of checks. The NUC's Windows 10 system is fully updated. The tech had no idea this is an active thread and I only found it this afternoon. Obviously, the tech will not be getting back to me with a fix. We had also discussed the possibility of a warranty replacement. But, as it is apparent this affects several NUCs, that seems like a waste of time. Besides, my daughter would not be able to swap out the guts into a replacement NUC, or simply the circuit board.
I assume my daughter's NUC might have the v184.108.40.20641 driver but only if it was installed by Windows. Her NUC was not used since March until a week ago and I only recall the Intel update app providing a new BIOS. I did not pay attention to what drivers Windows was installing but the tech informed me all drivers were up to date.
I would not know where to find the older drivers mentioned by some posters. But, Intel provides this warning: "After a driver has been updated to a Windows DCH Driver, it's possible to roll back to a legacy driver; however, rolling back isn't recommended as it involves a complex process that could result in system instability. This system instability is especially pertinent to graphics drivers."
A method that works 3 times out of 4 to have the monitor respond after the NUC awakens is to reseat the HDMI cable. That tells me this is a hardware defect. Reconnecting the HDMI cable likely causes a mild surge in the signalling voltage which tells the NUC a monitor is connected. So, my theory is that the NUC is failing to detect a monitor when it awakens.