I had to pull CMOS jumper out and then back in to get my 8i7HVK to turn on once after I unplugged it and moved to another room and once it just shut off on its own and needed the CMOS to be reset. Everything on the NUC is up to date and its been working fine for a few days.  I don't want to just start pulling at the battery as it seems to be taped onto the board in some way covered by tape.  Should I replace the battery and/or the jumper cable and if so does anyone know the exact steps?  The instructions listed for other NUC's don't seem to exactly apply.  Thanks.    

I have an intel NUC NUC7I7BNH acting as a media center and haven't had any issues until the last couple of days.

My system is shutting down and when re-booted saying it was the result of overheating.

The issue only occurs when I am streaming content from the BBC web site via MS edge.  If I download the content it is fine and streaming from other sites Netflix's and Amazon don't seem to have any issues.  That and I haven't really done that much checking other browsers as I don't want to damage the unit but getting it too hot too many times.

All of the drivers and bios are up to date.  I have tried adjusting the setting in the bios to keep the processor cooler, but I am surprised that it is only the BBC site that seems to cause the issue.  I had process monitor open while streaming and the processor only seemed to be running at <15% so it isn't very stressed.

Does anyone have an idea as to what can be causing this.

Dont know when it happen first but since a week or so, video and audio suddenly "stutter", when clicking forward into timeline at a running video. And it happens with Firefox, Opera and Chrome. No problem with VLC or other mediaplayer playing videos from local sources.

Sometimes its sorted after restart browsers, but in most cases NUC has to restart. Problem occours on YT, Vimeo and other videoplatforms.

Hope the attached video will show.

Problem is more or less reproducible after a couple of clicks into timeline of video.

System is Win 10 pro with all updates (Clean install ~4 weeks ago)

Intel drivers currend versions

Browsers currend versions and new installed without addons for testing

I've absolutely no idea what going on here, because never seen such a problem.

Thanks in advance

Hermann

Hi all,

l got a NUC7i5BNH from a couple of months which does not work as expected, I'll try to make it clear to understand and reproduce.

My previous NUC was a Skylake one, so starting with the new Kaby Lake unit I soon noticed it couldn't reach CPU package C-States lower than C2 in Windows 10 when CPU is idle and the display is powered on:

this produces higher temperatures, higher power consumption and higher noise than expected (and necessary), so I started investigating from BIOS configuration.

After some testing, and with my great surprise, I found that "Legacy boot" option disabled was the culprit for this behaviour.

Premises:

- Unit is NUC7i5BNH equipped with 2x16GB RAM Crucial CT2K16G4SFD8213 and M.2 2280 AHCI SSD Crucial CT500MX200SSD4 MX200

- Operating system used for testing is always an UEFI clean install of Windows 10 x64 1803 up to date (latest build 17134.12) with no crap/trash, just a clean system for testing purpose

- Drivers are up to date as per NUC7i5BNH Intel support page

- UEFI firmware is up to date (BNKBL357.86A Version 0065 6/6/2018)

- BIOS was always flashed via jumper recovery (F9 default settings loaded after power off and power chord uplug)

- NUC is connected to a DELL display via HDMI port or to a Samsung Smart TV via HDMI port (both display are full HD resolution 1920x1080), with same results

- Monitoring software used is HWInfo via its sensors interface

Better issue description:

BIOS default settings enable both "UEFI boot" and "Legacy boot":

with this configuration, when CPU is idle C-States work as expected and CPU can reach package C8 with the display active, using both "Balanced"* and "Power Saver" Windows 10 plans (* to get C8 with "Balanced" powerplan may be necessary  to enable AHCI Link power management (via Windows registry) and changing the value in power settings gui to "HIPM+DIPM" or "Lowest").

When display goes off due to power saving timer, CPU can reach lower C-States (C9 and C10)

This is the expected (per Intel docs) behaviour, so power comsumption, heat and temperatures get benefits when there is no load.

If  "Legacy boot" option is disabled, when CPU is idle can reach only package C2 state if the display is active.

When display goes off due to power saving timer, CPU can now reach package C8 and lower (do note that if you just unplug HDMI cable or manually power off the display via power button, CPU still remains in C2).

This is not the expected behaviour, because system cannot get advantage of its lovely built-in power saving features.

I am an IT-Pro guy and a passionate user, so I tried to find a solution by myself but no luck:

- I tried changing any related or unrelated bios setting in every crazy combination possible without any effect, you need "Legacy boot" enabled

- I tried any available graphic driver for this unit and also generic non-NUC from Iris Plus 640 page (newer tried was Version:24.20.100.6136), no difference

- I tried tuning every Windows power setting - also advanced hidden ones - in every power plan, but no difference

- I tried also (just for fun) loading Windows 10 from USB drive with no other drive connected to the unit, no difference

Do note the same thing happened on previous BIOS version I went through before last revision (0062, 0063, 0064), so downgrade to that revisions is useless.

Do also note that any recent Linux distribution I tried (Kernel from 4.15 up to 4.17) do not suffer this problem:

also without loading firmware blobs on boot, CPU can reach lower package C-States - as Intel Powertop reports - just as expected also when  "Legacy boot" option is disabled in BIOS and the display is active (may be necessary tuning some options to enable all power saving features, but it is expected and works fine).

Paradox here is Windows 10 is the only supported operating system by Intel for this NUC and - as per following Intel recommendations - should be installed only in UEFI mode to avoid issues.

Supported Operating Systems for Intel® NUC Products - "UEFI Boot is required to install Windows 10 - this is set by default in BIOS Setup. Using Legacy Boot results in multiple errors in Device Manager."

Some thoughts:

I think it is clearly a bug, it shows itself only in Windows 10 but I do not think Windows 10 is the culprit itself:

as reported, leaving "Legacy boot" option enabled everything works as expected, so I suppose this problem is bios related but obviously I can be wrong (as I never thought that such option could have such impact on power saving features).

UEFI and legacy boot initialize hardware in different ways and some of the features involved in power saving options are based on firmware blobs, anyway this is out of my control and I just don't have any other idea on how to try fixing it, so I hope someone can "pass the ball" to the right team and investigate about this issue, which could be present on other Kaby Lake NUCs as well.

To me and - I suppose - to the largest part of owners and/or potential owners - leaving "Legacy boot" option enabled is not an actual solution nor an option because:

- it prevents booting via UEFI PXE and micro sd slot (due to PCIe interface)

- it prevents (or may prevent) booting from NVME PCIe SSD, and if you disable it you'll loose idle power saving features

- if "secure boot" is needed and so enabled, "Legacy boot" is consequently disabled and you'll loose idle power saving features

- power saving features like CPU Package C-States and Speed Shift are really important arguments of choice when buying these units and should work as expected at least in the only OS supported by Intel

Steps to reproduce:

To get Idle C-States NOT working with display active:

- In VisualBIOS, load BIOS default settings with F9 and disable "Legacy boot", save with F10 and confirm.

- Boot Windows 10 and monitor CPU C-States with HWInfo (or any other similar software): C2 will be maximum package C-State reached in idle  with every power plan when the display is active.

To get Idle C-States working with display active (the expected behaviour):

- In VisualBIOS, load BIOS default settings with F9, save with F10 and confirm.

- Boot Windows 10, select "Power Saver" powerplan* and monitor CPU C-States with HWInfo (or any other similar software): C8 will be maximum package C-State reached in idle with the display active (C9 and C10 with display off via power saving timer).

* You can get C8 also with "Balanced" powerplan, but it may be necessary to enable AHCI Link power management (via Windows registry) and changing the value in power settings gui to "HIPM+DIPM" or "Lowest".

I hope it can be fixed, thanks in advance to everyone who can investigate about it.

Best regards to all the community,

M

So shortly after windows boots everything starts to hang. Does not matter what it is. I could open up chrome browser, video games, file explorer, ect. I notice it most on my video game emulators. It goes from as soon as i click it and running to almost 2 minutes before it loads. After that i can open anything for an example settings in windows and it will take 2-3 minutes before it opens. Same in the file explorer. It will say its working on it and it will take minutes to open what my $200 desktop can do in seconds. I have tried fresh windows install, I am up to date on all my drivers and bios is up to date. This is frustrating considering how much this cost and I did not touch it for months thinking it was a driver issue of some sort. Any help would be great thanks.

I was fiddeling with the performance settings of my NUC5i5RYH / BIOS:

Date : 06/04/2018

Vendor : Intel Corporation (www.intel.com)

Version : RYBDWi35.86A.0370.2018.0604.1034

-Board-

Name : NUC5i5RYB

Vendor : Intel Corporation (www.intel.com)

and run into following problems:

The 'voltage offset' settings in Visual Bios under 'performance' tab are not respected after saving.

You can set whatever you want with no effect  - except the voltage offset setting for graphics.

With intel XTU tool you will see then after boot into windows that no offset which was previously set and saved in bios is applied - except the one for graphics. Other hardware info tool do confirm this.

If you set a value less than '-0.5' e.g. voltage offset with XTU  for one of these values (core / ring (uncore) / graphics: change together to same value in XTU) the NUC will crash and recover in BIOS.

So you can set offset value for with XTU but not in BIOS (exept the offset for graphics).

'RealTimePerformanceTuning' on/off does not affect this issue.

Next issue ist the scaling in BIOS for all negative offset values:

It starts with '-0.5' and continues with  '-0.10, -0.15, -0.20...' till '-0.50' again!

The lowest possible negative value should start with '-0.05' not '-0.5'.

Using positive Values the scaling is correct.

EDIT:

- Same issues with BIOS Date : 07/09/2018

Version : RYBDWi35.86A.0371.2018.0709.1155

- only lowest negative value affected with wrong scaling of offset values

I just got a samsung evo ssd hard drive and put it in the second slot as a secondary drive. It does not appear in the bios or windows 10?

Do these drives only work in slot 1?

I own this NUC for about a month and eveything was fine until now.

Today, when I turned on my nuc, almost all the led lights didn't worked. Even the light on power button and skull.

I've tried to reset bios but no luck.

For some reason the only lights that currently working are first two on the front panel. I can turned them on and off in the led manager, but other lights not react on any settings.

I've never updated bios in my nuc and didn't not installed any new software in the last weak.

I plan to buy a VR Headset to use with  NUC8I7HVK and I have looked at the 2 best products, Oculus Rift  and HTC Vive.

HTC Vive

I have run the program SteamVR Performance Test and the Test Result is attached.

The conclusion is that the NUC Graphic does not fulfil the recommended minimum.

The recommendation is to upgrade of the Graphic board!

The minimum Graphic requirement is similar capacity as Geoforce GTX 1060 or AMD Radeon RX480, which are about 20% faster than RX Vega M RH.

The processor are ok.

Worth to mention is that the test program on the Vive Web does not report correctly for the NUC. It reports Intel Graphic instead of AMD, but Steam VR Performance Test do it right.

Oculus Rift

Here it's going better. The test program reports that everything is ok, but a little minus for Graphics.

Oculus has two requirements specification instead of one from HTC Vive (Recommended minimum).

1. Minimum. This is a little lower requirement than HTC Vives and the NUC fulfils it.

2. Recommended. The NUC dows not fulfils it.

Anyone here that has used VR Headset with the NUC?

Experience….

The only way to upgrade is to add an external graphic card through Thunderbolt bus.

/Bjorn

Hello,

We dev a HMI for a device (it's not our device we bought it) which embedded a NUC (NUC7i7DNB)
And our final customer ask us to add a function to enable/disable the power of NUC's usb front panel (which are deported on device front panel).

What we want :

Disable/enable specific usb port or the entire root hub power.

If we need to change bios setting from os and reboot to apply it, it's ok.

Additional infos :

There are no external device connected on ports when we want to do such command.

We need to completely remove power from usb port (we though they want to protect their device against "usb killer" attack, or such vandalism)

We can do it in the bios settings, but we want to allow the user to do it from the OS (Which is Ubuntu 18.04)

What we tried/saw :

- It's look like root usb on this NUC don't allow power switching (sudo lsusb -v)

     The hub 0 "Linux Foundation 2.0 root hub" (Bus 001 Dev 001) says : wHubCharacteristic 0x000a     No power switching (usb 1.0)

     The hub 1 "Linux Foundation 3.0 root hub" (Bus 002 Dev 001) says : wHubCharacteristic 0x000a     No power switching (usb 1.0)

- Despite this, we tried to send sendClearPortFeature(PORT_POWER) to all root hub's ports, after that, root hub Port Status said power on ports is away, but that false, +5V still present between pin+ & pin-

- We tried to unbind driver of hubs, without success

- It's write in USB2.0 specification  "The roothub provides the same functionality for dealing with USB topology as other hubs" (Chapter 10.2.8 Root Hub)

And  "This state (Powered-off) is supported for all hubs. A port transitions to this state in any of the following situations: From any state except Not Configured when the hub receives a ClearPortFeature(PORT_POWER) request for this port" (Chapter 11.5.1.2 Powered-off)

Did we miss something ?

Is it possible to do such operation from OS ? on this NUC ?

Thanks in advance

Alex

Source :

https://www.mjmwired.net/kernel/Documentation/usb/power-management.txt

http://sdphca.ucsd.edu/Lab_Equip_Manuals/usb_20.pdf

Code source used to test :

https://github.com/yy502/hub-ctrl

Ok so loaded up the New Forza Horizon 4 Demo and Disaster happens I get this message

So after trying to update drivers to latest all a epic fail reglardes of the messages I was getting saying I now have Adrenalin Edition 18.9.1 Optional Installed it makes no Difference and shows Version

Radeon Software Version 18.7.1 in the settings so Kind of in a bind as the Demo Keeps crashing at the exact same spot

Right after this message

Game will then Lock up

The NUC meets the Recommended specks Bios is up to date

Intel® NUC

Intel Core i7-8809G 3.1GHz (up to 4.2GHz Turbo)

2x 16GB DDR4 2400+ SODIMMs

Radeon RX Vega M GH 1063-1190MHz GPU, 800MHz Memory 4GB

512 GB M.2 SATA SSD Operating System and Apps

1TB M.2 SATA SSD Game's only Windows and Steam

4 TB External Hard drive USB-C for Photos and Back up files

Integrated Wifi Intel® Wireless-AC 8265 + Bluetooth 4.2

Ands my head is starting to hurt

Hi all,

we're using a NUC7 in a casing, so are implementing a remote button to turn it on and off. This button has a built-in LED, operating at 5V and 65mA max.

Is this too much for the header? I found the pins for the LED supply 5V in the TPS, but nowhere do they speak of the max current to draw.

Can anyone supply us with some more info? :-)

Thanks!!

We nave four 7I5BNK's that have been purchased over the last year. These are out of a batch of 20.  All of them Blue Screen at times.  Most Blue Screen shortly after the first login for the day.  If they have been running for a while they do not blue screen.  The Bugchecks are never the same.  When this first started happening we choose one of the first two to do extensive testing on.  Ran Memcheck on it for a couple of days without any errors found.  We then replaced the complete NUC and all hardware and that did not resolved the problem. We did find on that one if we disconnected the Ethernet cable and added a usb ethernet cable the issue almost went away. We suspected that particular unit was affected from our networking setup as the network connection comes through the desk phone.  We have had that in the past with one of our desktop computers.  We put in a Network Card in the desktop and stopped using the Network Jack on the Motherboard and those issues went away.  Now we have purchased more of these units and yesterday one of the new batch blue screened.  I immediately added the USB Ethernet jack and again first thing this AM it Blue Screened.  In Checking it our tech hear someone else mention theirs had done this yesterday also.  I have tried to keep them on the latest Bios, however we did not use 65 one these since there was no EB for it.  The Tech who was setting these up did not have the skills to use the other methods.  I will try to get 65 installed on them this week.  I will also use the Intel Driver Update Utility to see if any other drivers are out of date.  The installer was supposed to use only the latest Drivers that I had downloaded for them.

Hi,

I'm using my NUC7PJYH with a Crucial MX100 512GB SSD and 2x8GB DDR4-2400 Hynix HMA81GS6AFR8N-UH. Unfortunately I cannot activate dual-channel  mode and my memory speed is only 14GB/s reading and 10GB/s writing.

Is there a fix?

I am trying to install ESXi 6.7 on a NUC8I7HNK -

I've tried several different BIOS versions and in both Secure boot and Non-Secure Boot.

Each time it fails after loading all of the ESXi going into the Interactive installation with the following.

On the screen

“Shutting down firmware services…..

Using “simple offset” UEFT RTS mapping policy”

The NUC is now locked up.

Would really like to get this working.

The NUC7 works fine but liked the NUC8 with the two NICs built into the motherboard.

Any ideas on what to try?

I've seen this asked here and on Reddit but nothing in the VMWare forums - so please don't point me there but  I get an error message when booting to install ESXi 6.7

"Shutting down firmware services...

Using 'simple offset' UEFI RTS mapping policy"

UC8i7HVK version had a BIOS update that fixed but the HNK is still broken.  This is a bit ridiculous...

I've NUC8i7HVK on order. I would like a spare power supply and IR remote; supported under Windows 10. Can someone please help? Thank you.

I had to pull CMOS jumper out and then back in to get my 8i7HVK to turn on once after I unplugged it and moved to another room and once it just shut off on its own and needed the CMOS to be reset. Everything on the NUC is up to date and its been working fine for a few days.  I don't want to just start pulling at the battery as it seems to be taped onto the board in some way covered by tape.  Should I replace the battery and/or the jumper cable and if so does anyone know the exact steps?  The instructions listed for other NUC's don't seem to exactly apply.  Thanks.    

Does anyone else have a 5 sec boot delay, from the time the power button is pushed to the time the Intel NUC logo appears?

There seems to be a configuration issue with the Hades Canyon, after the 5 sec delay it will boot to a screen which has "Setup" and "BBS" on the upper right corner. It then proceeds to the normal boot screen and booting windows.

The 6th and 7th gen NUC's didn't have this issue, they booted to Windows in less than 2-3 seconds, and that's without fast boot enabled.

I am on the 47 version bios with 32GB of Corsair RAM 2400Mhz, if that matters.

I have started to get the error message in the attachment, now and then. It comes randomly when I use a browser.

The drivers are the latest.

Any idea what this is?

/Bjorn