Generational Tests on the i7-6700K: Linux Performance

Built around several freely available benchmarks for Linux, Linux-Bench is a project spearheaded by Patrick at ServeTheHome to streamline about a dozen of these tests in a single neat package run via a set of three commands using an Ubuntu 11.04 LiveCD. These tests include fluid dynamics used by NASA, ray-tracing, OpenSSL, molecular modeling, and a scalable data structure server for web deployments. We run Linux-Bench and have chosen to report a select few of the tests that rely on CPU and DRAM speed.

C-Ray: link

C-Ray is a simple ray-tracing program that focuses almost exclusively on processor performance rather than DRAM access. The test in Linux-Bench renders a heavy complex scene offering a large scalable scenario.

Linux-Bench c-ray 1.1 (Hard)

NAMD, Scalable Molecular Dynamics: link

Developed by the Theoretical and Computational Biophysics Group at the University of Illinois at Urbana-Champaign, NAMD is a set of parallel molecular dynamics codes for extreme parallelization up to and beyond 200,000 cores. The reference paper detailing NAMD has over 4000 citations, and our testing runs a small simulation where the calculation steps per unit time is the output vector.

Linux-Bench NAMD Molecular Dynamics

NPB, Fluid Dynamics: link

Aside from LINPACK, there are many other ways to benchmark supercomputers in terms of how effective they are for various types of mathematical processes. The NAS Parallel Benchmarks (NPB) are a set of small programs originally designed for NASA to test their supercomputers in terms of fluid dynamics simulations, useful for airflow reactions and design.

Linux-Bench NAMD Molecular Dynamics

Redis: link

Many of the online applications rely on key-value caches and data structure servers to operate. Redis is an open-source, scalable web technology with a b developer base, but also relies heavily on memory bandwidth as well as CPU performance.

Linux-Bench Redis Memory-Key Store, 1x

Linux-Bench Redis Memory-Key Store, 10x

Linux-Bench Redis Memory-Key Store, 100x

Generational Tests on the i7-6700K: Windows Professional Performance Generational Tests on the i7-6700K: IGP and $70 GPU Benchmarks
Comments Locked

477 Comments

View All Comments

  • SuperVeloce - Saturday, August 8, 2015 - link

    Wait, what? Skylake and 2011-3 in the same sentence? Who, for the love of god, would say such a thing? Power delivery is (again) new and very different from Haswell/Broadwell, so there is no chance to reuse 1150 and 2011-3
  • Oxford Guy - Saturday, August 8, 2015 - link

    The belief put forward was that Broadwell would be compatible with Haswell desktop motherboards and Skylake would be compatible with Haswell-E motherboards.
  • KAlmquist - Saturday, August 8, 2015 - link

    The analysis by Puget Sound Systems offers a plausible explanation of why Skylake has a higher TDP than Haswell or Ivy Bridge: the integrated GPU that comes with Skylake is faster and draws more power. It appears that if you don't use the integrated GPU, Skylake draws slightly less power than Haswell.
  • SuperVeloce - Saturday, August 8, 2015 - link

    That's definitely plausible. The other thing here is the TDP 4790K uses. 88W is too conservative for the clocks and voltages from that chip. They needed to up that I am sure.
  • bobbozzo - Saturday, August 8, 2015 - link

    Error in graph on final page:
    "Gains over Sandy Bridge.png" - the key for green says IVY bridge.
  • tuklap - Saturday, August 8, 2015 - link

    I don't know... Intel seems to keep pushing forwards every year with profit in mind. The thing that they are really making breakthrough is the non volatile, high bandwidth memory or Xpoint...

    If Xpoint will be available maybe this will give a new speed bump... But Sandy-Skylake is really good...
  • wizyy - Saturday, August 8, 2015 - link

    There is a review which shows 6600k to be quite a nice improvement over popular I5 processors in 10 recent games, over at eurogamer.net. Check it if you're a gamer thinking to upgrade your older I5.
  • SilverManSachs - Saturday, August 8, 2015 - link

    There is a good jump in IPC for the Core i5, less so for the Core i7. This makes sense as its harder to push the top end performance higher at smaller nodes but they did improve the i5 performance which is great as i5's are the most sold parts. Also, good overclocking room on the i7.

    Would be very interested to see 'Skylake vs Excavator' CPU only benchmarks on the mobile 17W parts. Please so that test for us AT!
  • soldier45 - Sunday, August 9, 2015 - link

    Spending $500+ on Skylake over my 2600k to get 3-5 fps in my games isn't really worth it. Having said that at the end of the day,I'm about to spend $700 on a 980Ti over a 780 classified so yeh I will end up going with Skylake.
  • asmian - Sunday, August 9, 2015 - link

    The interesting fact for me faced with building a new rig is how the i7-6700K compares with the 28-lane Haswell-E i7-5820K. For my usage (design/programming, no interest in SLI/Crossfire, regular Handbrake use), with very comparable mid-range boards (ASRock Z170 Extreme6+ versus ASRock X99 Extreme4 with the USB 3.1 A/C card) the price of mobo + board is almost identical at £490 or so in the UK right now - in fact, the Haswell-E combo would be £15 cheaper. All other added components (DDR4 memory, new OS, M2 SSD etc.) would be identical.

    So do the extra 2 cores at a somewhat lower eventual overclock for that Handbrake usage make up for extremely marginal extra IPC on 4 cores at a higher price (and trading a few extra features for many less SATA ports)? Somehow I doubt it... The only question remaining would be whether waiting another year or more for Skylake-E would be worth it for even more chipset features over X99, but that looks rather marginal as well.

Log in

Don't have an account? Sign up now