Power Consumption and Thermal Performance

The power consumption at the wall was measured with a 4K display being driven through the HDMI port. In the graphs below, we compare the idle and load power of the Intel NUC10i7FNH (Frost Canyon) with other low power PCs evaluated before. For load power consumption, we ran the AIDA64 System Stability Test with various stress components, and noted the maximum sustained power consumption at the wall.

Idle Power Consumption

The usage of a PCIe 3.0 x2 NVMe SSD along with a low TDP processor (25W, compared to the 28W+ processors in the other systems, other than the June Canyon NUC) manufactured in a highly-optimized and power-efficient 14nm process result in the Frost Canyon NUC having a sub-5W idle power consumption even while driving a 4K display. The system is able to sustain 90W+ at the wall for a very brief duration only, with the power falling slightly south of 60W for sustained workloads.

Our thermal stress routine starts with the system at idle, followed by four stages of different system loading profiles using the AIDA64 System Stability Test (each of 30 minutes duration). In the first stage, we stress the CPU, caches and RAM. In the second stage, we add the GPU to the above list. In the third stage, we stress the GPU standalone. In the final stage, we stress all the system components (including the disks). Beyond this, we leave the unit idle in order to determine how quickly the various temperatures in the system can come back to normal idling range. The various clocks, temperatures and power consumption numbers for the system during the above routine are presented in the graphs below.

Intel NUC10i7FNH (Frost Canyon) System Loading with the AIDA64 System Stability Test

The frequencies adjust to ensure that the 30W PL1 is maintained. With just the CPU loaded, the cores can maintain around 3 GHz sustained. Adding the GPU brings the CPU cores down to 2 GHz. With only the GPU stressed, the CPU cores can stay close to their boost clocks without consuming much power. The thermal design is quite good, with the package temperature never above 90C for extended durations.

Intel NUC10i7FNH (Frost Canyon) System Loading with Prime95 and Furmark

Our custom stress test with Prime95 and Furmark shows similar characteristics. Frequencies get adjusted to maintain a 30W package power, and the package temperature does not cross 85C. Furmark loads the GPU more than the AIDA stress component, and hence, we see the CPU frequency drop down to around 1.5 GHz in this case.

HTPC Credentials - Local Media Playback and Video Processing Concluding Remarks
Comments Locked

85 Comments

View All Comments

  • PeachNCream - Monday, March 2, 2020 - link

    Ganesh explained why the NUC was opened up in the article. Besides that, NUC systems are built to be user-accessible and are pretty simple to pull apart. It's like 4 screws to get the case open and pulling one ribbon cable off the motherboard to disconnect the mechanical drive. The barebones models are the same hardware and you have to crack the case to add storage or RAM or you do not have a complete computer.
  • nico_mach - Tuesday, March 3, 2020 - link

    I don't actually disagree with his reasoning, but I definitely have mixed feelings as a review.

    I forgot about the Atom NUCs completely. There's a name I haven't heard in a long time.
  • watzupken - Monday, March 2, 2020 - link

    With the new AMD APU arriving this year, I wonder if this is dead in the water considering the higher cost of the NUC. 14nm+++ is not going to save Intel when facing off 7nm from AMD as shown in the current processor stack from both teams.
  • timecop1818 - Monday, March 2, 2020 - link

    "7nm".

    Intel's "14nm+++++" (keep adding pluses, retards) is closer to 10nm than AMD's crap.
  • Fulljack - Monday, March 2, 2020 - link

    while fabrication nomenclature are now nowhere near it's actual marketed size, Intel's 14nm are still nowhere close with their own 10nm, and couldn't compete with TSMC's 7nm
  • Qasar - Monday, March 2, 2020 - link

    keep posting useless, anti amd, crap. keep showing the rest of us, and demonstrate to the world that you are a nitwit, then we will let you hang yourself, king oft trolls
  • Lord of the Bored - Thursday, March 5, 2020 - link

    Isn't this copy/paste'd from one of your other shill threads? Come on, man. Intel isn't paying you to repost the same old bullshit, your fans demand new content!
  • Korguz - Thursday, March 5, 2020 - link

    timecop1818 cant post new content, cause it doesnt have any.
  • xenol - Monday, March 2, 2020 - link

    The NUC product line is sort of dead in the water any way, I'd argue.
  • YB1064 - Monday, March 2, 2020 - link

    It would be helpful to include a small table that of benchmarks vs CPU performance scaling. Perhaps this is hard to do, but as a naive simple example:
    Benchmark#1 - scales with threads/cores
    Benchmark#2 - scales with clocks/IPC

    This is most likely highly complicated, but if anybody can do it, it is you guys. Ian, care to take a stab?

Log in

Don't have an account? Sign up now