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 Core m3-6Y30 in the Compute Stick form factor is given a bit of a challenge by the Core i5-4210Y in the Zotac ZBOX CI540 nano, a NUC-form factor machine.

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.

In these multi-threaded benchmarks, the quad-core processors (4C/4T) are able to score better than the 2C/4T configuration of the Core m3-6Y30.

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. The Core m3-6Y30 supports the AES-NI instruction for accelerating the encryption and decryption processes. 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 Intel Core m3-6Y30 Compute Stick 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 pure single-threaed performance advantage of the Core m3-6Y30 is evident here.

Dolphin Emulator Benchmark

Performance Metrics - I Networking and Storage Performance
Comments Locked

105 Comments

View All Comments

  • Vaz - Tuesday, June 28, 2016 - link

    Product photography seems to have hit rock bottom in anandtech reviews.
  • tamalero - Tuesday, June 28, 2016 - link

    The cheer size of the power block kinda makes the stick size useless, doesnt it?
  • xodusgenesis - Tuesday, June 28, 2016 - link

    Can this handle software decode of H.265? Or does it stress the CPU too much?
  • ganeshts - Tuesday, June 28, 2016 - link

    There is hybrid decoding for HEVC 8b - I expect it will work ok for 24p content, even 4K.

    However, 4Kp60 or HEVC 10b will be a no-go.
  • Kinematics - Wednesday, June 29, 2016 - link

    What about 10-bit h264?

    It's what's most disappointing in all this small form-factor reviews: At what point is this product *not* going to be useful to you? And all it ever does is cover stuff that I know already is going to work fine. Since I know that 10-bit won't be hardware accelerated (whereas everything that's tested is), it's impossible to even guess at how it compares in areas that you actually need a reviewer to look at.
  • DonMiguel85 - Tuesday, June 28, 2016 - link

    I'm still waiting for that GTX 960 review
  • nfriedly - Tuesday, June 28, 2016 - link

    I have one of these and one of the m5 versions and I pretty much agree with Anandtech's conclusions. They're the first compute stick that feel "fast enough" for basic office/browsing/htpc tasks. But they're definitely a bit on the expensive side.

    From what I've seen elsewhere and my personal experience the m5 version is a little faster for bursty workloads, but about on par with the m3 for anything sustained. The graphics might be a bit more powerful, though. Probably not worth the price difference.

    Also, on a different topic, Anandtech seriously needs to put some improvements into the commenting system. I had to get to page 4 just to find a single comment about the article... A way to collapse comment threads would probably be good enough, although voting or some kind of moderation might be even better.
  • lhorror - Wednesday, June 29, 2016 - link

    core m3 version on video - https://www.youtube.com/watch?v=AWcULkwryH4
  • piasabird - Wednesday, June 29, 2016 - link

    I would like to see comparison of this to something like an STX system where you can use normal DDR4 Ram and a standard CPU, with an M.2 SSD drive, and also an ITX system. All this to determine if it is worth it and also what are the advantages to having a larger drive and more RAM.
  • grand_puba - Friday, July 1, 2016 - link

    This website is going from bad to worse. There are few reviews and at least half of them are about cases and shit like that. I am going to remove you from my bookmarks.

Log in

Don't have an account? Sign up now