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)

C-Ray doesn't care much for the overclock, indicating that the bottleneck is elsewhere.

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

The Molecular Dynamics module of the test certainly prefers more physical cores , with the overclock giving the result a small raise but still lagging behind the Core i5 parts.

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 NPB Fluid Dynamics

Fluid Dynamics appreciates the overclock, and we sit in the middle of the Core i5 parts and well above the previous generation Core i7s.

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 strong 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

With Redis single thread speed as well as IPC is king, so with an overclocked Skylake it does rather well.

Core i3-6100TE Professional Performance: Windows Core i3-6100TE Gaming Performance: High End GTX 980/R9 290X
Comments Locked

62 Comments

View All Comments

  • nathanddrews - Thursday, March 17, 2016 - link

    Predicating the value of i3 overclocking as a whole on the 6100TE is absurd, don't you think? The 35% OC of the 6100TE is still slower than the base clock speed of the vanilla 6100 (3.7GHz), which when OCed over 4.2GHz offers a tremendous value. For the same price - why on earth would you buy the much slower 35W variant if your goal is overclocking?

    http://ark.intel.com/compare/90729,88181

    From the standpoint of motherboard/BIOS SKYOC availability alone I would agree that the case is now somewhat moot, but I think the case that has already been made months ago by DigitalFoundry and others is that the SKYOC i3 (normal i3 CPUs that you can buy) makes a very strong value. It's a shame that mobo vendors caved so quickly.
  • dualsmp - Thursday, March 17, 2016 - link

    I suspect politics is a work here. Also, the conclusion would be more clear cut with the 6100 rather than a bit nebulous with the 6100TE.

    Remember how fast Intel shutdown overclocking after the Clarkdale i3? When the Clarkdale i3 was overclocked it was nipping at the heels of Intel's more expensive parts at the time. The Clarkdale i3 offered too much value for the money when overclocked, so Intel had to shut it down.
  • ImSpartacus - Thursday, March 17, 2016 - link

    You're probably right about the politics. It's a shame, but Anandtech is probably under Intel's finger.
  • alistair.brogan - Thursday, March 17, 2016 - link

    Is there hope for an AMD Zen dual core that will overclock? That's all I'd recommend for people :)
  • ImSpartacus - Friday, March 18, 2016 - link

    If zen is only 40% better ipc than excavator, then it won't be leading Intel and amd will still need high clocks compete. So I would bet that the stock zen parts might be using up a fair bit of that overclocking headroom.
  • alistair.brogan - Thursday, March 17, 2016 - link

    I think a properly overclockable i3 would hit 4.5ghz, just like an i5 can, so imagine the results in that case. It wouldn't "maybe" be worth it, it would absolutely by awesome. Which is why Intel doesn't allow it..... don't want people buying 4.5ghz i3's and saying good bye to the expensive i5s
  • dragosmp - Thursday, March 17, 2016 - link

    Thanks for the review, it's a good reference point.

    I thought at some point the lowest end Xeon v5 (4C/8T) was quite a bit cheaper than the 6700K and much more available. If that could be OCed, and I think some motherboards like Gigabyte's Cxxx say they could, that would have been the fastest CPU while keeping some money in the pocket.

    Agreed with getting older and putting a higher value on our time, but if my kid decides to leave me alone for a while I still like to tinker with OCing
  • Samus - Friday, March 18, 2016 - link

    It's true Xeon's are usually priced between the i-series and i-series k equivalents (there are i5 and i7 based Xeon's) so it would be cheaper than a "k" part while giving you ECC and full vPro support, not to mention Xeon's appear to have soldered heatspreaders when starting with Ivy Bridge, everything but Xeon's were shipped with pretty subpar TIM/paste. That only slightly improved with Devils Canyon...they still weren't fused lid-to-die like a Xeon.

    There are a few theories why Intel uses solder on Xeon CPU's. One is obviously temperature and 24/7 consistency. The most plausible reason, though, is the weight and pressure of server cooling solutions (compared to most OEM PC coolers and water blocks. I weighed a tiny 1U Xeon cooler once. This thing was like 20mm tall, and it still weighed 870 grams. That's 2 lbs! Pure block of copper fins. Tightened down to 25Nm and exposed to the pressure, vibration and temperatures of a 1U environment for years, you can take a guess why Intel makes the Xeon lids more durable.
  • extide - Thursday, March 17, 2016 - link

    Please do that future piece, anyone who is seriously considering doing this will definitely not be sing the T or TE series chips, and as you mentioned you can't even really get them in the retail market. I would really like to see what a 4.4-4.6Ghz i3 can do.
  • LostWander - Thursday, March 17, 2016 - link

    The idea is to show how close to i5 performance you can get while paying the price of an i3. Although it would be interesting to see how well the i5 overclocks too so we can see how close to K performance we can get for a non-k price.

Log in

Don't have an account? Sign up now