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. As expected, the i5-4210Y performs really well except when compared to units sporting CPUs with much higher TDPs / clock rates.

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. CPUs supporting the AES-NI instruction for accelerating the encryption and decryption processes have, till now, been the higher end SKUs. However, with Bay Trail, even the lowly Atom series has gained support for AES-NI. The Core i5-4210Y in the ZBOX CI540 nano does have AES-NI support. TrueCrypt, a popular open-source disk encryption program can take advantage of the AES-NI capabilities. The TrueCrypt internal benchmark provides some interesting cryptography-related numbers to ponder. In the graph below, we can get an idea of how fast a TrueCrypt volume would behave in the Zotac ZBOX CI540 nano and how it would compare with other select PCs. This is a purely CPU feature / clock speed 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, with the unit making an appearance in the middle of the graphs. As expected, the performance is better than that of the BXBT-1900 and the ECA LIVA.

Dolphin Emulator Benchmark

Performance Metrics - I Networking & Storage Performance
Comments Locked

48 Comments

View All Comments

  • wintermute000 - Sunday, November 2, 2014 - link

    For some people that's worth the 50% cost saving (plus the laptop can be used as... a laptop? in a pinch lol). Not everyone is OCD about their TV setup.
  • BrokenCrayons - Tuesday, November 4, 2014 - link

    As long as performance doesn't matter much, $200 USD of inexpensive Bay Trail laptop would really be a far less costly solution. Quite a few modern low-end notebooks with 11 inch screens with those sorts of specifications don't have cooling fans and are kitted out with solid state memory making dust ingestion a non-issue. Yes, there's a lot less performance potential, but being able to grab the HTPC and use it someplace else when you need to leave the living room is sort of nice too since it's still a laptop.
  • p@nc@k3s - Friday, October 31, 2014 - link

    Glad to see an Intel WiFi NIC. What is the gigabit Ethernet NIC? I hope it's not Realtek.
  • saiga6360 - Friday, October 31, 2014 - link

    This one has Bluetooth so you can probably have better luck with those remotes. I would avoid IR if I can help it.
  • abufrejoval - Saturday, November 1, 2014 - link

    I think that's a very interesting little box! Thanks for the evaluation!

    Idle power in the French article linked behind Fanless Tech seems a little higher: They quote a 9-31Watt range.

    Could the plastic foil they found between the thermal pad and the chassis actually also have increased power consumption? Don't hotter chips consume more power?

    They also say that checking for the presence of the "isolated human error" would void the warranty, which is, well, bizarre in this case...

    Can you measure DRAM bandwidth on the unit?
    I'm a little worried that the single DRAM socket would be limiting the bandwidth unnecessarily.

    I guess I'd still perfer a somewhat more massive design with the i5-4200U (same CPU price), which I've seen sustain 2.1GHz peaks pretty long as well as the 2.6GHz spikes for things like spreadsheet recalcs.

    Or even with the vastly more expensive i7-4500, all the very same silicon inside.

    Unfortunately Intel gauges and charges for perceived performance pretty well: These little Haswells do deliver quite acceptable performance for those little sprints in typical desktop work, which you can't really see under these synthetic benchmark loads, where the first thermal limits kick in pretty pretty fast.

    Thus the i7-4500U really does feel no slower than a true 3GHz i7 quad core with 8 threads, when you browse a complex web page, reformat a complex document or recalc a huge spreadsheet.

    Because it a) does actually go to 3GHz for a comple of seconds when cold and b) none of these tasks exploit multiple cores or run long enough to heat up the CPU.

    Chances are you could even put the 4200 or 4500 into the very same chassis and just risk that they arrive at speeds very similar to the 4210Y under continued load.

    But they'd still "feel" a lot faster on office work stuff.

    And I might be tempted to spend an extra 50 bucks or so on a case which foregoes convection through the case and uses large external fins instead to avoid issues with dust and cleaning staff "whet cleaning" these cases to the point where create solid composite crusts of dust and dried cleaning agents.

    Too bad Broadwell will soon render these a lot less attractive far too soon!
  • Laststop311 - Sunday, November 2, 2014 - link

    This is very true. With broadwell supposed to be even more energy efficient the Y series (or is it called core M now instead of Y series or is the core m a power level below Y and there is still a Y and a u?) the fanless mini pc will end up able to convert the broadwell energy savings into even more performance at the same temperature and the U series with the new iGPU should make it actually able to game many popular titles at a respectable 30 fps and 1366x768 resolution and maybe even some at 1920x1080. I could see myself chilling in the living room loading up a game of hearthstone which would definitely run as the ipad air 1 is known to run it well and that is just an apple a7x proc not even the new A8x.
  • Laststop311 - Sunday, November 2, 2014 - link

    and before u bash the 768p gaming quite a few xbox one titles will end up at this resolution and fps. It's not as terrible as the low numbers make it sound. And games like hearthstone and league of legends should run at 1920x1080 on the intel graphics 5000 (the igpu in the intel mini pc) successor without a hitch. Broadwell might just take this category to a new level.
  • Laststop311 - Sunday, November 2, 2014 - link

    I only see this being a good buy for the noise nazis out there. My personal favorite was the intel box that was only an extra 153 as configured and it had a u series cpu with intel 5000 graphics instead of y series with intel 4200 and an intel 530 240GB SSD instead of a samsung 840 evo 120GB SSD and dual channel 1866 ram instead of single channel 1600mhz and intel 7260 2x2 wireless instead of intel 3160 1x1 wireless.

    The intel box beats it on every single specification for only a 153 dollar premium. So it would seem that unless being fanless is the number 1 thing you care about by a large margin you would have to be senile to choose the zotac nano over the intel.
  • wintermute000 - Sunday, November 2, 2014 - link

    well TBH unless size/noise is your only concern you're better off served via a mITX or better still a mATX build. They can still look pretty good aesthetically with the right case, and with right part selection don't make any appreciable noise.
  • kaczor47 - Sunday, November 2, 2014 - link

    I have been using this box for a couple of weeks now. Main use: XBMC for movies and audio. I am surprised that the silent aspect of the device is not appreciated more here - when listening to audio on hi-end speakers, last thing I want is to hear the fan noise in the background (during the quite parts of a classical piece). Not to mention that this thing is in my living room, where occasionally I want to relax in a quiet environment - again, the hum of a fan would sometimes be irritating.

    Of course, there you can get higher performance for an extra $X,but it is all about balance.

Log in

Don't have an account? Sign up now