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

  • skavi - Monday, January 22, 2018 - link

    nm, found it at ludashi.com.
    Does anyone know if you can set the language to English?
    Also, does Master Lu have any plans to support NNAPI? I'd like to see how the Pixel performs.
  • WorldWithoutMadness - Monday, January 22, 2018 - link

    Hmmm... I'm guessing next chip is a73 and a55, then next of it will be a75 and a55 just to give the user sense of 'upgrade' Reply
  • vladx - Tuesday, January 23, 2018 - link

    You can't pair A73 with A55 or A75 with A53, only A75/A55 and A73/A72/A57 with A53. Reply
  • zeeBomb - Monday, January 22, 2018 - link

    Holy crap...it's about time we got an Andrei post about SoCs!!! Reply
  • 69369369 - Tuesday, January 23, 2018 - link

    Inb4 huawei dropping driver support 1 year later Reply
  • lucam - Tuesday, January 23, 2018 - link

    ''I wouldn’t necessarily say that Apple is the SoC trend setter that other companies are trying to copy, as much as other vendors are coming to the same conclusion Apple has: to be able to evolve and compete in a mature ecosystem you need to be able to control the silicon roadmap yourself. Otherwise you fall into the risk of not being able to differentiate from other vendors using similar component stacks, or risk not being competitive against those vendors who do have vertical integration. Apple was early to recognize this, and to date Huawei has been the only other OEM able actually realize this goal towards quasi-silicon independence''

    LOL You must be joking...when did you try Apple products? I have seen NO A10X or A11 reviews from you guys...it's very difficult to believe what you can't prove...
  • SanX - Tuesday, January 23, 2018 - link

    Graphs are a worst mess, hard to comprehend
  • ZolaIII - Tuesday, January 23, 2018 - link

    This is one of the worst HiSilicon designs. So much black silicone & such a waist of it. Would be much better if they MP-d P6 DSP instead that horrible overgrown machine learning cluster block. Reply
  • haplo602 - Tuesday, January 23, 2018 - link

    I am normally not interested in phone ARM SoCs, but I wanted to just have a look at this one, so I looked at the rodmap table on the first page ... Now I am mighty confused ... The Kirin 970 especially ... So it does 2160p60 Decode and 2160p30 Encode ... not capable of 2160p30 Decode ? I guess the 950 is better in this regard ... the Camera/ISP line is also funny ... all 3 have Dual 14-bit ISP. no clue which is better ... again the 960 looks to be the best of the bunch ...

    Oh and the Interconect one takes tha crown ... just plain Arm CCI on the top model ? That CCI-550 on the 960 looks way superior ...

    For the future, either put up a reasonably detailed table that actually serves the purpose (provides at-a-glance overview and evaluation possibility) or don't. Doing a 1/2-assed job like this just lowers your credibility ...
  • haplo602 - Tuesday, January 23, 2018 - link

    oh ... not possible to edit comments ... I have to proof-read my own drivel next time :-) substitute the 950 with the 960 in that one appearance .... Reply

Log in

Don't have an account? Sign up now