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.

Video Encoding - x264 5.0

Video Encoding - x264 5.0

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 i7-4770R, being the flagship Crystal Well part for Haswell, hasn't been left behind. It does have AES-NU support and TrueCrypt, a popular open-source disk encryption program which 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 BRIX Pro 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:

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

Dolphin Emulator Benchmark

Performance Metrics - I Gaming Performance Evaluation
POST A COMMENT

55 Comments

View All Comments

  • Dirk Broer - Sunday, June 22, 2014 - link

    RAM speeds have a BIG impact on the performance of IGPs, both for Intel and AMD. Speed is even <a href="http://www.anandtech.com/show/7364/memory-scaling-... important</a> than Latency. Reply
  • Hrel - Monday, June 16, 2014 - link

    Except that paying for that "faster" RAM costs you money and yields no tangible benefit beyond 1866. The price/performance sweet spot is 1866.

    http://www.anandtech.com/show/7364/memory-scaling-...
    Reply
  • Dirk Broer - Sunday, June 22, 2014 - link

    "you can easily get 2133 for a decent price" You've seen the photograph, with SoDIMMs? Where do you buy 2133 MHz SoDIMMS for a decent price? Reply
  • TwistedKestrel - Saturday, June 14, 2014 - link

    I don't understand how graphics performance is worse if the RAM in the (2) is faster in every way. Even if you didn't use the XMP profile, it should still be faster! Reply
  • TwistedKestrel - Saturday, June 14, 2014 - link

    I meant to say, worse in ANY scenario. The (2) should win out every time, not just some of the time. Reply
  • ganeshts - Saturday, June 14, 2014 - link

    In areas where they are close or (1) wins out, the error is within measurement margin, but in the gaming scenarios where there is a big difference, (2) wins handsomely.

    Some benchmarks are also not dependent on the memory (CPU-bound).
    Reply
  • Samus - Sunday, June 15, 2014 - link

    It might be worth investigating whether the crystal well gpu can decrease CPU throughput when it is heavily taxed. That would explain why giving crystal well more memory bandwidth slightly decreases CPU performance. Although they are separate parts of the die, they communicate across the same bus and are flip-chip limited. Reply
  • etamin - Saturday, June 14, 2014 - link

    My guess is that the difference is more related to thermals than RAM. Intel iGP's are less dependent on DRAM speeds than AMD's, especially the Crystal Well parts. Outside of the one Tomb Raider benchmark at 1080p, the performance drop isn't significant (at that FPS, it's all kind of irrelevant anyways). Since the max power draw is higher on the 2nd edition, it's probably attributed to a combination of silicon quality and throttling. Reply
  • dylan522p - Saturday, June 14, 2014 - link

    Why did you use a 840 Evo. It is inferior to the MX100 which is also cheaper. For the same price you can get a drive with better nand and double the storage. Reply
  • ganeshts - Saturday, June 14, 2014 - link

    Well, this piece has been in the works for a couple of months now.. When I built the refreshed config, the Samsung 840EVO was the best value for money.. I see that the MX100 came to market just a couple of weeks back.. Reply

Log in

Don't have an account? Sign up now