We have found that if a local monitor is not plugged in to the NUC 5i5 the remote KVM screen remains black( in Bios and windows).
However Keyboard and mouse clicks remain active, we can see this on windows when running remote admin screen.
If we plug in a local monitor or emulate a local monitor, the remote KVM screen works well as expected.
This is rather frustrating we assumed that the built in KVM would work headless, mainly because the NUC unit itself has been designed to run headless as stated in the manual.
The Nuc in general fantastic compact unit. Our current solution is to use a modified mini display port to VGA adapter, which is around 1/4 the size of the NUC pcb which is a real shame.
Does anyone out there have a better solution, or is there already some sort of a firmware solution out there already.
Thank you for contacting the Intel® Communities.
I would like to know which is the software used to remotely access the unit.
This would be to replicate the issue from our end.
Also, I would like to know what is the exact model of NUC 5i5 that you are currently using.
Thanks for your reply
We have 12 NUC5i5MYHE units, they all do exactly the same thing. We are using up to date bios and Drivers. Although the older drivers and Bios did the same thing.
We have tried tightVNC, Ultra VNC and real VNC and get the same result.
We are using in conjunction with the intel AMT tool, but get the same result when using the VPro manager.
We also use NUC DC53427hye and have no problems, when windows headless on those units.
When running Linux headless on NUC DC53427hye we had to use nomodeset command otherwise we got a Black screen on KVM
We are fairly confident that our usage of the software is correct.
all the best
Thank you for the information provided.
I will be performing some test on my end, I will be providing you with my outcome.
Same setup here, but even more frustrating.
We've purchased some FIT Headless 4k Display Emulators and only got a partial solution!.
Booting up with a Dell Monitor attached and powered on allows VNC to work fine. Power down the monitor and the remote screen goes black. That's a puzzle because I thought that the EIDD mechanism was passive and did not need the monitor powered up.
Power back up the monitor and the graphics return fine. Now if I replace the monitor with the FIT Headless 4k (on a HDMI to mDP adapter cable) the graphics reappears. So far so good.
Now if I ask Windows to soft reboot with the FIT Emulator still in place, the graphics also restarts just fine. Excellent I thought. We must be on track.
But sadly if I shutdown the NUC and perform a cold boot using the FIT alone, the VNC session (via AMT BOIS of course) sees the BIOS flash screens and menu choices, briefly displays a Windows startup message - and then goes black. Nothing I've tried over the past two days seems to help whatsoever. Nor does trying to set a dummy VGA screen. The end result is that I still need to plug in a powered up monitor at every cold start; which entirely defeats the point of running headless.
All the very latest firmware and drivers I can find for Windows 7 Pro OS. It seems Intel have really struggled with this issue over the past year or so and given that everyone claims that the FIT Emulator magically solves the problem, I'm pretty gutted that even with one of these I'm still not able to get to a result.
Finally this morning I solved the problem for myself.
Because the NUC5i5MYHE uses mDP ports, it was necessary to use a short adaptor cable between the NUC and the FIT Headless Emulator which only comes with an HDMI plug. The cable I first purchased for the job is labelled as "Mini DisplayPort to HDMI 4K2K Adaptor" from Comsol.
On reflection I decided to try another adaptor cable which was labelled "Mini DisplayPort to HDMI Adaptor" , ie without the 4k2K feature.
And much to my delight it works! It seems there must be some active electronics in the other cable that interferes with powerup action.
I would like to thank you for providing us with the workaround and feedback for this situation.
For sure it is going to be useful for other users with this scenario.
Are you able to provide an update,as to when this issue may be resolved.
Its been a while now, and i can see other are having the same problem.
They have also had to resort to a using hardware work around.
all the best
looks like this is not only an issue with the Intel NUC, but with all Skylake AMT KVM Boards? At least Fujitsu D3417 also suffers with from that.
Will Intel update their Code in the nearest future? Why are always things that worked for years getting broken..
Are you running latest BIOS, BIOS 28 https://downloadcenter.intel.com/download/25868/BIOS-Update-MYBDWi5v-86A- Download BIOS Update [MYBDWi5v.86A]?
It seems that BIOS 27 has a fix for the issue with headless AMT KVM.
Updated Intel® ME firmware to version 10.0.47.1006.
Fixed issue where Windows installation fails with PXE.
Fixed issue with headless AMT KVM.
If you are running BIOS 28, could you download and try BIOS 27 https://downloadcenter.intel.com/download/25250/BIOS-Update-MYBDWi5v-86A-?product=84861 Download BIOS Update [MYBDWi5v.86A]?
Thought I would just check in and find out if you were able to resolve this problem with the latest recommendations given about updating the BIOS.
I look forward to hearing from you.
Bios update does not work for us.
If we delete the intel graphics driver and use the basic windows driver we can get a very low resolution screen.
Which is better than nothing, but not very usable.
all the best
Thank you for your answer on this stewie, let me investigate into this in order to provide you with further assistance.
I will keep you posted.
Did you try BIOS 27 https://downloadcenter.intel.com/download/25250/BIOS-Update-MYBDWi5v-86A-?product=84861 Download BIOS Update [MYBDWi5v.86A]?
Let me know if you can try it and what are the results.
Thought I would just check in and find out if you were able to resolve this problem with the BIOS 27 advised by rguevara.
I look forward to hearing from you.