Performance Metrics - II

In this section, we mainly look at benchmark modes in programs used on a day-to-day basis, i.e, application performance and not synthetic workloads.

x264 Benchmark

First off, we have some video encoding benchmarks courtesy of x264 HD Benchmark v5.0. This is simply a test of CPU performance. The cTDP-Up of the Skylake BRIX enables it to to get slightly higher numbers in the second pass when compared to the Kaby Lake Cubi2-005BUS. Actual x264 encoding performance with the Kaby Lake processors and using the latest x264 releases (with AVX512 AVX2 support) is bound to be even better compared to the numbers below.

Video Encoding - x264 5.0 - Pass 1

Video Encoding - x264 5.0 - Pass 2

7-Zip

7-Zip is a very effective and efficient compression program, often beating out OpenCL accelerated commercial programs in benchmarks even while using just the CPU power. 7-Zip has a benchmarking program that provides tons of details regarding the underlying CPU's efficiency. In this subsection, we are interested in the compression and decompression MIPS ratings when utilizing all the available threads.

7-Zip LZMA Compression Benchmark

7-Zip LZMA Decompression Benchmark

TrueCrypt

As businesses (and even home consumers) become more security conscious, the importance of encryption can't be overstated. Almost all modern CPUs support AES-NI for acceleration of the cryptographic processes. The Core i7-7500U in the Cubi2-005BUS does have AES-NI support. TrueCrypt, a popular open-source disk encryption program can take advantage of the AES-NI capabilities. In the graph below, we can get an idea of how fast a TrueCrypt volume would behave in the MSI Cubi2-005BUS and how it would compare with other select PCs. This is a purely CPU feature / clock speed / TDP-based test.

TrueCrypt Benchmark

Agisoft Photoscan

Agisoft PhotoScan is a commercial program that converts 2D images into 3D point maps, meshes and textures. The program designers sent us a command line version in order to evaluate the efficiency of various systems that go under our review scanner. The command line version has two benchmark modes, one using the CPU and the other using both the CPU and GPU (via OpenCL). The benchmark takes around 50 photographs and does four stages of computation:

  • Stage 1: Align Photographs
  • Stage 2: Build Point Cloud (capable of OpenCL acceleration)
  • Stage 3: Build Mesh
  • Stage 4: Build Textures

We record the time taken for each stage. Since various elements of the software are single threaded, others multithreaded, and some use GPUs, it is interesting to record the effects of CPU generations, speeds, number of cores, DRAM parameters and the GPU using this software.

Agisoft PhotoScan Benchmark - Stage 1

Agisoft PhotoScan Benchmark - Stage 2

Agisoft PhotoScan Benchmark - Stage 3

Agisoft PhotoScan Benchmark - Stage 4

Dolphin Emulator

Wrapping up our application benchmark numbers is the Dolphin Emulator benchmark mode results. This is again a test of the CPU capabilities, and the cTDP-Up of the Skylake BRIX makes it come out on top.

Dolphin Emulator Benchmark

Performance Metrics - I Networking and Storage Performance
POST A COMMENT

32 Comments

View All Comments

  • MrSpadge - Friday, December 30, 2016 - link

    First paragraph:
    "The Kaby Lake-U (KBL-U)series with 15W TDP CPUs was introduced along with the 4.5W Kaby Lake-Y ones in Q3 2014."
    You mean Broadwell here instead of Kaby Lake, don't you?
    Reply
  • ganeshts - Friday, December 30, 2016 - link

    bad typo, with the link in it correctly linking to the Q3 2016 article. It has been fixed. Reply
  • Voldenuit - Friday, December 30, 2016 - link

    No tear down? Cooling system? System layout? Noise measurements? Reply
  • Great_Scott - Friday, December 30, 2016 - link

    It doesn't really matter anyways.

    This kind of system is a waste of time: laptops have greater functionality for less price, and the same components. The U-series doesn't distinguish between the i5 and i7 beyond clock speeds.

    You can go to any site online and get a U-series laptop for ~$400 US that also includes RAM and Storage.
    Reply
  • barleyguy - Saturday, December 31, 2016 - link

    For an HTPC, a NUC or UCFF is a lot more convenient than a laptop. They generally have less fan noise, fit in less space, and boot up with the TV (or projector) as the primary monitor.

    (I'm typing this on a Zotac Z-Box in my living room.)

    That said, I'm really skeptical of the U-series processors in general. My Z-Box has an i5-4200u, and I've had some issues with throttling under load. Hopefully they've improved from the 4200u to the 7500u.
    Reply
  • kmmatney - Monday, January 2, 2017 - link

    and the laptop will have an OS as well... Reply
  • milkod2001 - Tuesday, January 3, 2017 - link

    and it also has screen, keyboard and tracking device to start with and usually cost much less.

    I can only see good use of NUCs if the cheapest & crappiest NUCs are considered to buy to replace Intel P4 10 years old machines with existing monitors, keyboards and mice. For anything else: laptop is much better solution unless you need to Vesa mount your computer on telly.
    Reply
  • niva - Tuesday, January 3, 2017 - link

    I don't know about all that, please show me the laptop with a Kaby Lake "i7" chip and comparable specs that really stacks up to this NUC and obsoletes it.

    I think these boxes are ideal for installing linux on it, paying for a windows license when buying a laptop really irks me anyways so that's money down the drain too. I just bought an Acer Aspire E 15 from Amazon for $350 but that thing had a terrible drive in it, an i3-7100U, and only a single stick of 4GB RAM. Of course I had an old SSD from my previously dead laptop I could put in and I dumped linux on it which is fine with 4GB of RAM at this point, but you guys are seriously ignoring some of the components thrown into this NUC.
    Reply
  • BrokenCrayons - Tuesday, January 3, 2017 - link

    *shoulder taps* Psst, from one Linux user to another, Windows licenses are not a waste because close to or over 99% of laptops that ship with Windows end up as ....*drum roll*... Windows latops.

    As far as your Acer's sad configuration is concerned, welcome to the wonderful world of budget computers. You got what you paid for. If you want better new hardware, you'll have to dig a little deeper into the wallet.
    Reply
  • ganeshts - Friday, December 30, 2016 - link

    You guys are hard to please. In any case, this is a bog standard NUC, nothing special about it.

    Cooling system - all that matters is effectiveness, and I hope readers agree when I say that we have the most comprehensive evaluation of the effectiveness of the cooling solution of mini-PCs in the review circuit. Look at the graphs in the 'Power Consumption and Thermal Performance' section.

    Noise measurement - we only do subjective eval in these reviews. Providing noise numbers for these types of PCs (i.e, non gaming mini-PCs) is pretty much useless because the noise floor is too high and these types of PCs are too quiet in our evaluation setup. Creating a noise measurement lab is not worth the investment for the number of PC reviews that I do per year. You can find dedicated guys like SPCR ( Silent PC Review - http://www.silentpcreview.com/ ) who fill that market niche with excellent reviews and articles. (I would imagine even they would not find the Cubi 2 and other similar PCs interesting enough to do noise measurements)
    Reply

Log in

Don't have an account? Sign up now