I'm using the NUC NUC7PJYH with Ubuntu 18.04 and so far it works almost flawless.
The suspend mode is also working without any errors, which means I can suspend (blue LED flashing) and resume and I don't see any errors in the log files.
But when I measure the power consumption it's a bit to high for my taste and I assume there is something not suspending properly, which leads to a higher power consumption.
My measurements are:
0.2-0.5W powered off
2.6W in suspend / sleep
Why 2.6W in suspend? Shouldn't it be around or less than 1W?
Are there some settings I need to tweak?
I've already tried the newest Linux kernels (up to 4.18.3) and updated to the newest BIOS versions, but nothing changed.
2.6W seems to be the correct power consumption in sleep state for your NUC. The Technical Product Specification: https://www.intel.com/content/dam/support/us/en/documents/mini-pcs/nuc-kits/NUC7xJY_TechProdSpec.pdf https://www.intel.com/content/dam/support/us/en/documents/mini-pcs/nuc-kits/NUC7xJY_TechProdSpec.pdf , in Table 8 on page 30, defines <5W power consumption for this state and it is dependent on the standby power consumption of wake-up devices used in the system.
Well, that would be disappointing for a low power device. Any laptop and even my gaming laptop is capable of using less than 1W in suspend.
And it also does not fit to some reviews, which report it to be around 1W. Some reported less than 1W.
The spec is just very safely unprecise. I mean that NUC can use <5W in idle already. <span __jive_emoticon_name="wink" __jive_macro_name="emoticon" class="jive_emote jive_macro" data-renderedposition="71_541_16_16" src="/18.104.22.1689a91/images/emoticons/wink.png">
I also tested without any wake-up devices or infact any devices attached. Anyway, I stillt hope there is some tweak possible.
kev300: Based on the information showing in the TPS that the power consumption is dependent on the wake-up devices, there are some things you can try to reduce the power consumption, on the Secondary Power Settings in BIOS:
Wake on LAN from S5
Wake from S3 via CIR
Wake from S4 and S5 via CIR
Or turn on Deep S4/S5.
thanks for your help.
I tried all the options and disabled cir completely but it did not help in S3.
The power consuption in off (S5) state is reduced by 0.1W.
kev300: You are welcome.
It is great to hear that the power consumption in off (S5) state is reduced by 0.1W.
I encourage all the peers viewing this thread that if they have further suggestions for this scenario to post all the details on this thread.
Intel Customer Support Technician
Under Contract to Intel Corporation
HI @KEgal it's interesting to read that you made the NUC7PJYH work with Ubuntu 18.04, including wakeup from S3. I wonder if you also got Ubuntu 18.10 to work properly. I am interested because of my issue discussed here:
Hi, meanwhile I've switched to Manjaro 18, because Ubuntu developed a severe stuttering issue after some update. Unfortunately Manjaro also has stuttering, but not as bad as Ubuntu. I'm trying to fix that since weeks and the standby topic is now less important to me. ;) But I can try if it still works with Manjaro. The only thing I had to do for Ubuntu 18.04 is to disable usb suspend for the wireless keyboard, so I can wake it up with it.
I suspect the BIOS cpu microcode update (spectre and meltdown) as the root cause for the stuttering, but can't downgrade. If you know something about that stuttering topic, any help is appreciated. :)
The sound is fine. But the image, no matter if it's a video or some 3D animation, if I use hardware video acceleration or not, there is a constant juddery micro stutter going on as if the frame rate drops every 2-3 seconds. But the frame rate is fine.
I did not have that with ubuntu 18.04 in it's very first version and old factory BIOS. The TV and all settings are the same.
@KEgal sorry for the delay. I just came home and made some experiments with 1080p HD video on normal 1920x1080 screens. The NUC7PJYH was on Lubuntu 18.10. There was some stutter, yes. But not much. Very little, actually. Then I wasn't sure if I would have noticed that without you mentioning it first. So I went to my main desktop machine, a NUC5i7RYH :-) and played the same source. I think you would not notice the difference. I tested with https://www.youtube.com/watch?v=QAUzWtLMnU0 . Dou you want me to test with video from filesystem too?