Benchmarking Performance: CPU Office Tests

The office programs we use for benchmarking aren't specific programs per-se, but industry standard tests that hold weight with professionals. The goal of these tests is to use an array of software and techniques that a typical office user might encounter, such as video conferencing, document editing, architectural modelling, and so on and so forth.

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

Chromium Compile (v56)

Our new compilation test uses Windows 10 Pro, VS Community 2015.3 with the Win10 SDK to combile a nightly build of Chromium. We've fixed the test for a build in late March 2017, and we run a fresh full compile in our test. Compilation is the typical example given of a variable threaded workload - some of the compile and linking is linear, whereas other parts are multithreaded.

Office: Chromium Compile (v56)

This is another case where I think our improvised testbed is playing a bigger part, and I'd like to eventually re-run this on my standard testbed. Especially as compiling heavily hits more than just the CPU.

GeekBench4: link

Due to numerous requests, GeekBench 4 is now part of our suite. GB4 is a synthetic test using algorithms often seen in high-performance workloads along with a series of memory focused tests. GB4’s biggest asset is a single-number output which its users seem to love, although it is not always easy to translate that number into real-world performance comparisons.

Office: Geekbench 4 - Single Threaded Score (Overall)

Office: Geekbench 4 - MultiThreaded Score (Overall)

Like CineBench, the Core i7-8086K does will on the synthetic single threaded test.

PCMark8: link

Despite originally coming out in 2008/2009, Futuremark has maintained PCMark8 to remain relevant in 2017. On the scale of complicated tasks, PCMark focuses more on the low-to-mid range of professional workloads, making it a good indicator for what people consider 'office' work. We run the benchmark from the commandline in 'conventional' mode, meaning C++ over OpenCL, to remove the graphics card from the equation and focus purely on the CPU. PCMark8 offers Home, Work and Creative workloads, with some software tests shared and others unique to each benchmark set.

Office: PCMark8 Home (non-OpenCL)

Here the 8086K does eek out a win over the 8700K, although just barely.

Benchmarking Performance: CPU Encoding Tests Benchmarking Performance: CPU Legacy Tests


View All Comments

  • AsParallel - Monday, June 11, 2018 - link

    Addition. The 8087 was the floating point coprocessor for the 8086/88 Reply
  • 29a - Monday, June 11, 2018 - link

    You didn't put 2mb of RAM in an original IBM PC it supported 256kb max. Reply
  • HStewart - Monday, June 11, 2018 - link

    I had a special card in the PC - it was EMS memory - that could also fill up the main system memory to 640kb - instead of normal cache mode use by the card - I configured it as ram drive. Memory above 640Kb was directly accessible by the system. Reply
  • peevee - Monday, June 11, 2018 - link

    8086 being slower than 8700 just indicates an error in your methodology.
    For example, one has updated microcode for exploits and another does not.
  • TheinsanegamerN - Monday, June 18, 2018 - link

    OOORrrrr....its a different motherboard, not the usual test bed. The motherboard used for this is an asrock board, which explains the difference in performance. Reply
  • Memo.Ray - Monday, June 11, 2018 - link

    As I mentioned in my comment in the other article a couple of days ago:

    Intel managed to give away 8086 "binned" 8700K (AKA 8086K) and still make some money on top of it. win-win situation :D
  • Xenphor - Monday, June 11, 2018 - link

    How did they get such a lower score on the Dolphin benchmark with a 5ghz overclock on the 8086k? Isn't the benchmark single core only and considering the 8086 already turbos to 5ghz on a single core, why would there be that much of a difference? I tried it on my 8700k at 5ghz and only get a score of about 265-270 with 2666mhz ram. Reply
  • Ian Cutress - Monday, June 11, 2018 - link

    The 5.0 GHz turbo, at stock, doesn't kick in that often. Depends on how the software sets its own affinity, and most do not. This is the danger with only single core turbo - with all the modern software in the background, even with Windows and scheduling, you rarely hit single core Turbo. Reply
  • Xenphor - Monday, June 11, 2018 - link

    I suppose but even on the Dolphin forums spreadsheet the highest score is a 249 which is a 7700k at 5.2ghz. Reply
  • Ian Cutress - Tuesday, June 12, 2018 - link

    I'll retest when I'm back home at the end of the week and recovered from jet lag Reply

Log in

Don't have an account? Sign up now