CPU Performance: SPEC2006

SPEC2006 has been a natural goal to aim for as a keystone analysis benchmark as it’s a respected industry standard benchmark that even silicon vendors use for architecture analysis and development. As we saw SPEC2017 released last year SPEC2006 is getting officially retired on January 9th, a funny coincidence as we now finally start using it.

As Android SoCs improve in power efficiency and performance it’s now becoming more practical to use SPEC2006 on consumer smartphones. The main concerns of the past were memory usage for subtests such as MCF, but more importantly sheer test runtimes for battery powered devices. For a couple of weeks I’ve been busy in porting over SPEC2006 to a custom Android application harness.

The results are quite remarkable as we see both the generational performance as well as efficiency improvements from the various Android SoC vendors. The Kirin 970 in particular closes in on the efficiency of the Snapdragon 835, leapfrogging the Kirin 960 and Exynos SoCs. We also see a non-improvement in absolute performance as the Kirin 970 showcases a slight performance degradation over the Kirin 960 – with all SoC vendors showing just meagre performance gains over the past generation.

Going Into The Details

Our new SPEC2006 harness is compiled using the official Android NDK. For this article the NDK version used in r16rc1 and Clang/LLVM were used as the compilers with just the –Ofast optimization flags (alongside applicable test portability flags). Clang was chosen over of GCC because Google has deprecated GCC in the NDK toolchain and will be removing the compiler altogether in 2018, making it unlikely that we’ll revisit GCC results in the future. It should be noted that in my testing GCC 4.9 still produced faster code in some SPEC subtests when compared to Clang. Nevertheless the choice of Clang should in the future also facilitate better Androids-to-Apples comparisons in the future. While there are arguments that SPEC scores should be published with the best compiler flags for each architecture I wanted a more apples-to-apples approach using identical binaries (Which is also what we expect to see distributed among real applications). As such for this article the I’ve chosen to pass to the compiler the –mcpu=cortex-a53 flag as it gave the best average overall score among all tested CPUs. The only exception was the Exynos M2 which profited from an additional 14% performance boost in perlbench when compiled with its corresponding CPU architecture target flag.

As the following SPEC scores are not submitted to the SPEC website we have to disclose that they represent only estimated values and thus are not officially validated submissions.

Alongside the full suite for CINT2006 we are also running the C/C++ subtests of CFP2006. Unfortunately 10 out of the 17 tests in the CFP2006 suite are written in Fortran and can only be compiled with hardship with GCC on Android and the NDK Clang lacks a Fortran front-end.

As an overview of the various run subtests, here are the various application areas and descriptions as listed on the official SPEC website:

SPEC2006 C/C++ Benchmarks
Suite Benchmark Application Area Description

(Complete Suite)
400.perlbench Programming Language Derived from Perl V5.8.7. The workload includes SpamAssassin, MHonArc (an email indexer), and specdiff (SPEC's tool that checks benchmark outputs).
401.bzip2 Compression Julian Seward's bzip2 version 1.0.3, modified to do most work in memory, rather than doing I/O.
403.gcc C Compiler Based on gcc Version 3.2, generates code for Opteron.
429.mcf Combinatorial Optimization Vehicle scheduling. Uses a network simplex algorithm (which is also used in commercial products) to schedule public transport.
445.gobmk Artificial Intelligence: Go Plays the game of Go, a simply described but deeply complex game.
456.hmmer Search Gene Sequence Protein sequence analysis using profile hidden Markov models (profile HMMs)
458.sjeng Artificial Intelligence: chess A highly-ranked chess program that also plays several chess variants.
462.libquantum Physics / Quantum Computing Simulates a quantum computer, running Shor's polynomial-time factorization algorithm.
464.h264ref Video Compression A reference implementation of H.264/AVC, encodes a videostream using 2 parameter sets. The H.264/AVC standard is expected to replace MPEG2
471.omnetpp Discrete Event Simulation Uses the OMNet++ discrete event simulator to model a large Ethernet campus network.
473.astar Path-finding Algorithms Pathfinding library for 2D maps, including the well known A* algorithm.
483.xalancbmk XML Processing A modified version of Xalan-C++, which transforms XML documents to other document types.

(C/C++ Subtests)
433.milc Physics / Quantum Chromodynamics A gauge field generating program for lattice gauge theory programs with dynamical quarks.
444.namd Biology / Molecular Dynamics Simulates large biomolecular systems. The test case has 92,224 atoms of apolipoprotein A-I.
447.dealII Finite Element Analysis deal.II is a C++ program library targeted at adaptive finite elements and error estimation. The testcase solves a Helmholtz-type equation with non-constant coefficients.
450.soplex Linear Programming, Optimization Solves a linear program using a simplex algorithm and sparse linear algebra. Test cases include railroad planning and military airlift models.
453.povray Image Ray-tracing Image rendering. The testcase is a 1280x1024 anti-aliased image of a landscape with some abstract objects with textures using a Perlin noise function.
470.lbm Fluid Dynamics Implements the "Lattice-Boltzmann Method" to simulate incompressible fluids in 3D
482.sphinx3 Speech recognition A widely-known speech recognition system from Carnegie Mellon University

It’s important to note one extremely distinguishing aspect of SPEC CPU versus other CPU benchmarks such as GeekBench: it’s not just a CPU benchmark, but rather a system benchmark. While benchmarks such as GeekBench serve as a good quick view of basic workloads, the vastly greater workload and codebase size of SPEC CPU stresses the memory subsystem to a much greater degree. To demonstrate this we can see the individual subtest performance differences when solely limiting the memory controller frequency, in this case on the Mate 10 Pro with the Kirin 970.

An increase in main memory latency from just 80ns to 115ns (Random access within access window) can have dramatic effects on many of the more memory access sensitive tests in SPEC CPU. Meanwhile the same handicap essentially has no effect on the GeekBench 4 single-threaded scores and only marginal effect on some subtests of the multi-threaded scores.

In general the benchmarks can be grouped in three categories: memory-bound, balanced memory and execution-bound, and finally execution bound benchmarks. From the memory latency sensitivity chart it’s relatively easy to find out which benchmarks belong to which category based on the performance degradation. The worst memory bound benchmarks include the infamous 429.mcf but alongside we also see 433.milc, 450.soplex, 470.lbm and 482.sphinx3. The least affected such as 400.perlbench, 445.gobmk, 456.hmmer, 464.h264ref, 444.namd, 453.povray and with even 458.sjeng and 462.libquantum slightly increasing in performance pointing out to very saturated execution units. The remaining benchmarks are more balanced and see a reduced impact on the performance. Of course this is an oversimplification and the results will differ between architectures and platforms, but it gives us a solid hint in terms of separation between execution and memory-access bound tests.

As well as tracking performance (SPECspeed) I also included a power tracking mechanisms which relies on the device’s fuel-gauge for current measurements. The values published here represent only the active power of the platform, meaning it subtracts idle power from total absolute load power during the workloads to compensate for platform components such as the displays. Again I have to emphasize that the power and energy figures don't just represent the CPU, but the SoC system as a whole, including interconnects, memory controllers, DRAM, and PMIC overhead.

Alongside the current generation SoCs I also included a few predecessors to be able to track the progress that has happened over the last 2 years in the Android space and over CPU microarchitecture generations. Because the runtime of all benchmarks is in excess of 5 hours for the fastest devices we are actively cooling the phones with an external fan to ensure consistent DVFS frequencies across all of the subtests and that we don’t favour the early tests.

The Kirin 970 - Overview SPEC2006 - The Results


View All Comments

  • Wardrive86 - Monday, January 22, 2018 - link

    Also really a testament to the Adreno 500 series of GPUs..great performance with good energy consumption and good temps. Can't wait to see how the 600 series does Reply
  • arvindgr - Monday, January 22, 2018 - link

    Nice article. Can someone highlight if chipset supports USB v3.x?? GsmArena lists USBv2 which is scarry for a flagship chip! Reply
  • hescominsoon - Monday, January 22, 2018 - link

    Samsung does not use exynos in the US due to a license agreement with...Qualcomm. https://www.androidcentral.com/qualcomm-licensing-...

    I prefer exynos to the QC SOC's....
  • Wardrive86 - Monday, January 22, 2018 - link

    Why do you prefer exynos over snapdragon? Not being smart, just curious Reply
  • tuxRoller - Monday, January 22, 2018 - link

    The 3% higher integer IPC? Reply
  • lilmoe - Monday, January 22, 2018 - link

    I understand that it's out of your educational level to understand what makes an SoC better, since it has been explained to you by myself and others, so please stop using abbreviations like you're some sort of expert. Do you even know what IPC means? SMH... Reply
  • tuxRoller - Tuesday, January 23, 2018 - link

    I'm assuming you've confused me with another.

    "The Exynos 8895 shows a 25% IPC uplift in CINT2006 and 21% uplift in CFP2006 whilst leading the A73 in overall IPC by a slight 3%."

    Yes, that's simply referencing the CPU, but that's a pretty important component and one whose prowess fans of Sam have enjoyed trumpeting.
  • Wardrive86 - Monday, January 22, 2018 - link

    Ok maybe...though I don't know what workload you would even be able to see "3% higher integer IPC" on a phone. The only workloads I'm running that even remotely tax these monsters, really come down to how well the Vulkan drivers pan out, it's ability to not thermally throttle all of its performance away and actually do this for awhile away from a charger. For these workloads Snapdragon is King as the Mali Vulkan/OpenGLes 3.x drivers are terrible in comparison. Again I was just curious Reply
  • tuxRoller - Tuesday, January 23, 2018 - link

    @Wardrive86 I was being a bit facetious. I assume the person either prefers Samsung because of an association that's developed between the success of the company and their own sense of self-worth, or they like watching YouTube videos of proper opening a bunch of apps while a timer runs on the screen:) Reply
  • Space Jam - Monday, January 22, 2018 - link

    >We’ve seen companies such as Nvidia try and repeatedly fail at carving out meaningful market-share.

    Don't think i'd call Nvidia's strategy for mobile SoCs as of the Shield Portable 'pursuing market-share' and I think their actual intentions have been more long-term with emphasis around the Drive CX/PX. The Shield devices were just a convenient way to monetize exploration into ARM and their custom Denver cores. Hence why we saw the Shield Portable and Tablet more or less die after one iteration; the SoCs were more or less there as an experiment. They weren't really prepared I think for the success the Shield TV has had and so that's gotten to see some evolution; the Nintendo Switch win is also nice for them but not really the focus. As much as I want to see a more current Tegra for a Shield Tablet (A73, Pascal cores, <=16nm) the Shield Tablet 2 was cancelled and doesn't look to be getting an update.

    >Meanwhile even Samsung LSI, while having a relatively good product with its flagship Exynos series, still has not managed to win over the trust of the conglomorate's own mobile division. Rather than using Exynos as an exclusive keystone component of the Galaxy series, Samsing has instead been dual-sourcing it along with Qualcomm’s Snapdragon SoCs. It’s therefore not hard to make the claim that producing competitive high-end SoCs and semiconductor components is a really hard business.

    We did see the Exynos 7420 with its Samsung sourced Exynos modem 333 which further adds to the questions of *why* Samsung bothers to source Snapdragons for the US. That's just extra development complexity on multiple levels. I always thought it had something to do with the cost of CDMA patent licensing, so they'd just opt to use Qualcomm's products and the Galaxy S6 was a special situation as Snapdragon was hot garbage.

    There has to be some reason that Samsung bothers with Snapdragon when their Exynos offerings perform pretty similarly.

Log in

Don't have an account? Sign up now