Office Performance: Extreme Editions

The dynamics of CPU Turbo modes, both Intel and AMD, can cause concern during environments with a variable threaded workload. There is also an added issue of the motherboard remaining consistent, depending on how the motherboard manufacturer wants to add in their own boosting technologies over the ones that Intel would prefer they used. In order to remain consistent, we implement an OS-level unique high-performance mode on all the CPUs we test which should override any motherboard manufacturer performance mode.

All of our benchmark results can also be found in our benchmark engine, Bench.

Dolphin Benchmark: link

Many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance. This benchmark runs a Wii program that raytraces a complex 3D scene inside the Dolphin Wii emulator. Performance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU. Results are given in minutes, where the Wii itself scores 17.53 minutes.

Dolphin Emulation Benchmark

WinRAR 5.0.1: link

Our WinRAR test from 2013 is updated to the latest version of WinRAR at the start of 2014. We compress a set of 2867 files across 320 folders totaling 1.52 GB in size – 95% of these files are small typical website files, and the rest (90% of the size) are small 30 second 720p videos.

WinRAR 5.01, 2867 files, 1.52 GB

3D Particle Movement

3DPM is a self-penned benchmark, taking basic 3D movement algorithms used in Brownian Motion simulations and testing them for speed. High floating point performance, MHz and IPC wins in the single thread version, whereas the multithread version has to handle the threads and loves more cores.

3D Particle Movement: Single Threaded

3D Particle Movement: MultiThreaded

Agisoft Photoscan – 2D to 3D Image Manipulation: link

Agisoft Photoscan creates 3D models from 2D images, a process which is very computationally expensive. The algorithm is split into four distinct phases, and different phases of the model reconstruction require either fast memory, fast IPC, more cores, or even OpenCL compute devices to hand. Agisoft supplied us with a special version of the software to script the process, where we take 50 images of a stately home and convert it into a medium quality model. This benchmark typically takes around 15-20 minutes on a high end PC on the CPU alone, with GPUs reducing the time.

Agisoft PhotoScan Benchmark - Total Time

HandBrake v0.9.9: link

For HandBrake, we take two videos (a 2h20 640x266 DVD rip and a 10min double UHD 3840x4320 animation short) and convert them to x264 format in an MP4 container.  Results are given in terms of the frames per second processed, and HandBrake uses as many threads as possible.

HandBrake v0.9.9 LQ Film

HandBrake v0.9.9 2x4K

The Market, X99 Refresh and our Test Setup Office and Web Performance
Comments Locked

205 Comments

View All Comments

  • sor - Sunday, June 5, 2016 - link

    Actually, without those benchmarks I wouldn't have known for sure that we were GPU limited or that CPU makes almost NO difference (I might have expected a few more FPS). I also found it interesting that one game stood out as being CPU limited.
  • IUU - Sunday, June 5, 2016 - link

    All good and awesome but Intel's persistence on maintaining many-core cpus as niche expensive
    products is a dangerous strategy, that may ultimately lead to its downfall.

    If the IT world would like to see a renewed growth, high performance computing should enter the lives of ordinary Joes. Plenty of apps could make their way into our lives via many-core cpus.
    Games with really improved AI, local voice and image recognition, reading comprehension, advanced text editing, advanced 3d printing, etc etc.

    Looking down on the needs of ordinary people, is nonsensical, for it is "simple joes", the needs of which led to many core cpus and advanced gpus,that the hpc world so much uses. No further development can come if the same old strategy is not applied.
  • Motion2082 - Monday, June 6, 2016 - link

    What other XEONs would you recommend for 8 core?
  • craveable - Tuesday, June 7, 2016 - link

    Seriously, guys. My personal opinion is to skip Broadwell-E, unless you absolutely cannot wait till Q1 2017. It's a very realistic timeframe when Skylake-E will be released. Some journalist spotted a Gigabyte motherboard on Computex 2016, suited for Skylake-E, sporting a new LGA 3647 socket and 6-channel memory controller. Memory bandwidth was always a crucial selling point of HEDT CPUs, so a transition from 4 (current) to 6 (Skylake-E) is an important incremental update.
  • craveable - Tuesday, June 7, 2016 - link

    However, further studying of leaks reveled that the next HEDT will be released in Q2 2017 and be probably named Skylake-X. In the leaked roadmap the Skylake-X availability timeframe is strangely the same as for Kaby Lake-X.
  • craveable - Wednesday, June 8, 2016 - link

    However #2, it's likely, that only server part of Skylake (EP) will get LGA 3647 and 6-channel controller. And Skylake-X will get 4 channels and the same LGA 2011-3.
  • galta - Wednesday, June 8, 2016 - link

    There is no point in saying that someone should skip upgrading. But for budget constraints, if your processor is not good enough, you have to upgrade; if it still performs as you need, you do not need to upgrade, even if new generation has the same price and 50% more performance.
    It is said that Intel is trying to milk us with a +USD1,700 cpu and that not all softwares (games) are fully optimized for more than 2-4 cores, but if you still run a i7 9xx series, it is probably time to upgrade, even if Broadwell-E increase in performance is not huge.
  • mapesdhs - Thursday, June 9, 2016 - link

    That makes no sense. You state that BW-E performance increase isn't that much, while saying someone with an X58 6c should still upgrade anyway. What for? Why would someone on a constrained budget spend so much for so little speed gain? Given the platform differences, X58 users would probably benefit more from newer storage tech such as M2 and newer USB3x, in which case a 5930K on a decent board would be more sensible (or a 5820K if they don't need the PCIe lanes). What you've suggested just sounds like upgrading for its own sake. These days there are far more nuanced options available, especially from the used market.
  • fifa17 - Sunday, June 12, 2016 - link

    I'm putting a system together and I'm stuck between Core i7-6850K and Core i7-6800K on a ROG STRIX X99 GAMING motherboard. The only confusing point is the number of PCI Express Lanes which is 28 for i7-6800K and 40 for the other. I've pre ordered the ASUS ROG Strix GeForce® GTX 1080 and don't plan on adding a second one in the future. I'm also not getting an optical drive of any sort and no HDDs. Just a 1TB Samsung 850 PRO SSD. Which CPU should I choose? Any ideas about these components?
  • legolasyiu - Wednesday, June 29, 2016 - link

    It will be best to get Core i7 6850 with 40 lanes for GTX 1080 16x with 850 Pro.

Log in

Don't have an account? Sign up now