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 Celeron J1900 with active cooling performs about twice as better as the passively cooled LIVA, but can't match the Haswell CPUs in the other mini-PCs.

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. The presence of four distinct cores helps the unit move to the middle of the pack in the decompression ratings.

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 Silvermont, even the lowly Atom series has gained support for AES-NI in some SKUs. Unfortunately, AES-NI is not a feature available in Bay Trail-D. AES operations have to be done in software. TrueCrypt, a popular open-source disk encryption program can take advantage of the AES-NI capabilities, but fall back to a software implementation in the absence of AES-NI. 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 GIGABYTE GB-BXBT-1900 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, and the trend seen in earlier CPU-focused workload graphs are evident here also.

Dolphin Emulator Benchmark

Performance Metrics - I Networking & Storage Performance
Comments Locked

35 Comments

View All Comments

  • nathanddrews - Tuesday, October 28, 2014 - link

    Size doesn't matter.
  • vailr - Saturday, October 25, 2014 - link

    This review is too scatterbrained. The reviewer says "Intel HD Graphics" but doesn't specify which version; "not being sold in the North American market", yet also says: "a US power cord".
    Wut? Anand is seeing his old web site go down the drain.
  • nickb64 - Sunday, October 26, 2014 - link

    "Intel HD Graphics" is exactly what Intel calls it. No number or anything.

    http://ark.intel.com/products/78867/Intel-Celeron-...
  • sjprg2 - Saturday, October 25, 2014 - link

    Why would any company make the SSD controller a SATA 2 instead of a SATA 3?
    also a 5 GHZ wireless would make this a great remote backup.
  • abufrejoval - Sunday, October 26, 2014 - link

    You'd need to ask Intel that: This is a system on chip or SoC and SATA comes off that chip.
    Intel might say that SATA 2 saves power and that might even be true. More likely they are just afraid they'll sell less mainline CPUs and chipsets if these SoC were too powerful.

    In practice I find the difference only important on jobs I wouldn't want to do on this system anyway.

    The wireless card isn't soldered on: In theory you could get another card that supports 5GHz for minimal bucks.
  • nothing immortal - Sunday, October 26, 2014 - link

    No kabini result? i want to see this cheap amd chip compared with this expensive intel box. Really, it should be there.
  • abufrejoval - Sunday, October 26, 2014 - link

    As much as I love the BRIX in terms of form factor, case look and sturdyness etc. I could never quite fathom their choice of notebook type active cooling systems, especially since they are full metal cases (top cover also has metal under a cover which is plastic for "looks").

    Well of course these notebook fans are cheap, being mass produced in incredible numbers, but they get far too noisy under load. I tried a 15Watt Haswell BRIX and had to return it heartbroken: I loved the little box, but I couldn't stand the constant fan speed changes, which gave me far too much feedback on just how hard the CPU was working.

    I run the GIGABYTE GA-J1900N-D3V which is completely passive and I thus know it's possible to design a passive box in the BRIX form factor.

    And if by all means they need to share a fan design with the bigger boxes, why can't they put a bigger one in, say something Noctua, which never more than whispers?

    Although one of the biggest attractions of these BayTrails is that you can design systems all passive which just run forever somewhere in a corner completely silent and sealed with no chance of choking on dust.

    BTW: I'm glad you finally measured clock speeds and power consumption! Yes, the J1900 (and the J1800 likewise) *never* run at anything but turbo clocks under load. The "official" clock rates are simply bogus numbers, perhaps designed to keep people away from these SoCs or to "manage performance expectations": With Intel everything is possible.

    Idle power clearly speaks for the BRIX and the integrated power supply, which is probably a far better match than the Pico-PSU and external 12V brick that I am using on the GIGABYTE GA-J1900N-D3V, which gives me 10/28 Watts on idle/full load.

    BTW: I've never noticed that a single DRAM channel limited the BayTrail SoCs in any way: Most likely that's because they only ever have a single channel DRAM interface anyway.

    And for Core 2 comparisons: I didn't run too many benchmarks but the J1900 at 2.41 GHz matched 80% of my QX9100 @ 2.40 GHz on all pure CPU benchmarks I tried (POVray and Cinebench R15 raytracing) at 28 vs. 65 Watts.

    If they managed to make this completely passive and silent this would be the perfect desktop. It runs pretty much every x86 OS right out the box with decent but not awe inspiring performance at 1920x1200 or below. Video at least on Windows with is great with all that hardware support from an Intel HD VPU including QuickSync and 3D is limited only by speed not by feature availability, which often enough is all you want (e.g. Compiz 3D effects on Linux).

    Can't expect gaming performance from a system costing less and using less power than an entry level grapics card.
  • chizow - Monday, October 27, 2014 - link

    @Ganesh T S: did you get a chance to test this with WMC? Specifically, high bitrate CableCard streams where you change the resolution from windowed to fullscreen or portrait/landscape. My Asus T100 wasn't able to handle this well at all so I ultimately stopped using it as a WMC device and finally sold it. It would hang for up to 60 minutes changing context/resolution and I could never figure out if it was the slow eMMC storage, the GPU, single-channel RAM or what.

    I wanted to get a Celeron NUC for an HTPC but ultimately this scared me off and I went with the i3 NUC instead, which cost more than 2x as much.
  • zlandar - Tuesday, October 28, 2014 - link

    I know Celeron is just a name but every time I've used Intel chips below i3 the performance sucks.

    Rather pay the price premium for an i3 than risk being unhappy with a stuttering NUC.

    The comparison with other NUCs seem way off as you mostly compared it against i5/i7 systems and no i3. It's like stacking a HTPC video card against a mid to high range gaming video card.
  • leonhk1 - Wednesday, October 29, 2014 - link

    Hi,
    I have stock of Brand New Samsung GALAXY Note 4 for sale at $500 only, sealed in box with 1year warranty.
    Interested buyer should E-mail me at: megas83@yahoo.co.uk

Log in

Don't have an account? Sign up now