Benchmarking Performance: CPU System Tests

Our first set of tests is our general system tests. These set of tests are meant to emulate more about what people usually do on a system, like opening large files or processing small stacks of data. This is a bit different to our office testing, which uses more industry standard benchmarks, and a few of the benchmarks here are relatively new and different.

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

PDF Opening

First up is a self-penned test using a monstrous PDF we once received in advance of attending an event. While the PDF was only a single page, it had so many high-quality layers embedded it was taking north of 15 seconds to open and to gain control on the mid-range notebook I was using at the time. This put it as a great candidate for our 'let's open an obnoxious PDF' test. Here we use Adobe Reader DC, and disable all the update functionality within. The benchmark sets the screen to 1080p, opens the PDF to in fit-to-screen mode, and measures the time from sending the command to open the PDF until it is fully displayed and the user can take control of the software again. The test is repeated ten times, and the average time taken. Results are in milliseconds.

System: PDF Opening with Adobe Reader DC

Our PDF test relies mostly on single core frequency, but memory frequency can also help. The 2400G pips the 1400, and the older AMD processors take a back seat. This is the sort of test that Intel's 4.2 GHz chips can take advantage off, as shown by the Core i3-8350K.

FCAT Processing: link

One of the more interesting workloads that has crossed our desks in recent quarters is FCAT - the tool we use to measure stuttering in gaming due to dropped or runt frames. The FCAT process requires enabling a color-based overlay onto a game, recording the gameplay, and then parsing the video file through the analysis software. The software is mostly single-threaded, however because the video is basically in a raw format, the file size is large and requires moving a lot of data around. For our test, we take a 90-second clip of the Rise of the Tomb Raider benchmark running on a GTX 980 Ti at 1440p, which comes in around 21 GB, and measure the time it takes to process through the visual analysis tool.

System: FCAT Processing ROTR 1440p GTX980Ti Data

The FCAT program is single threaded, so again Intel's chips take a win here. The Ryzen 5 2400G takes another chunk out of the Ryzen 5 1400, due to its higher frequency.

Dolphin Benchmark: link

Many emulators are often bound by single thread CPU performance, and general reports tended to suggest that Haswell provided a significant boost to emulator performance. This benchmark runs a Wii program that ray traces a complex 3D scene inside the Dolphin Wii emulator. Performance on this benchmark is a good proxy of the speed of Dolphin CPU emulation, which is an intensive single core task using most aspects of a CPU. Results are given in seconds, where the Wii itself scores 1,052 seconds (17.53 minutes).

System: Dolphin 5.0 Render Test

3D Movement Algorithm Test v2.1: link

This is the latest version of the self-penned 3DPM benchmark. The goal of 3DPM is to simulate semi-optimized scientific algorithms taken directly from my doctorate thesis. Version 2.1 improves over 2.0 by passing the main particle structs by reference rather than by value, and decreasing the amount of double->float->double recasts the compiler was adding in. It affords a ~25% speed-up over v2.0, which means new data.

System: 3D Particle Movement v2.1

3DPM is our first multi-threaded test, and the Ryzen 5 2400G powers ahead over the 1400 due to frequency, and ahead the Core i3-8350K due to thread count. This is a benchmark that can take advantage of multithreading, so the quad-core APU with eight threads pushes ahead of the six-core Intel Core i5-8400.

DigiCortex v1.20: link

Despite being a couple of years old, the DigiCortex software is a pet project for the visualization of neuron and synapse activity in the brain. The software comes with a variety of benchmark modes, and we take the small benchmark which runs a 32k neuron/1.8B synapse simulation. The results on the output are given as a fraction of whether the system can simulate in real-time, so anything above a value of one is suitable for real-time work. The benchmark offers a 'no firing synapse' mode, which in essence detects DRAM and bus speed, however we take the firing mode which adds CPU work with every firing.

System: DigiCortex 1.20 (32k Neuron, 1.8B Synapse)

Agisoft Photoscan 1.3.3: link

Photoscan stays in our benchmark suite from the previous version, however now we are running on Windows 10 so features such as Speed Shift on the latest processors come into play. The concept of Photoscan is translating many 2D images into a 3D model - so the more detailed the images, and the more you have, the better the model. The algorithm has four stages, some single threaded and some multi-threaded, along with some cache/memory dependency in there as well. For some of the more variable threaded workload, features such as Speed Shift and XFR will be able to take advantage of CPU stalls or downtime, giving sizeable speedups on newer microarchitectures.

System: Agisoft Photoscan 1.3.3 (Large) Total Time

For Photoscan, certain elements of the algorithms require full cores to get the best performance, hence why the six-core CPU comes top and the Ryzen 5 2400G and Core i3-8350K are matched. That being said, the multithreading of the 2400G outweighs the extra frequency of the 8350K.

iGPU Gaming Performance, Cont Benchmarking Performance: CPU Rendering Tests
POST A COMMENT

177 Comments

View All Comments

  • Gideon - Monday, February 12, 2018 - link

    BTW Octane 2.0 is retired for Google (just check their github), and even they endorse using Mozillas Speedometer 2.0 (darn can't find the relevant blog post). Reply
  • Ian Cutress - Monday, February 12, 2018 - link

    I know; in the same way we have legacy benchmarks up, some people like to look at the data.

    Not directed to you in general, but don't worry if 100% of the benchmarks aren't important to you: If there's 40 you care about, and we have 80 that include those 40, don't worry that the other 40 aren't relevant for what you want. I find it surprising how many people want 100% of the tests to be relevant to them, even if it means fewer tests. Optane was easy to script up and a minor addition, just like CB11.5 is. As time marches on, we add more.
    Reply
  • kmmatney - Monday, February 12, 2018 - link

    In this case, a few 720p gaming benchmarks would have been useful, or even 1080p at medium or low settings. Reply
  • III-V - Tuesday, February 13, 2018 - link

    Who uses 720p and is in the market for this? Reply
  • PeachNCream - Tuesday, February 13, 2018 - link

    I'm happy with 1366x768 and I'm seriously considering the 2400G because it looks like it can handle max detail settings at that resolution. I'm not interested in playing at high resolutions, but I do like having all the other non-AA eye candy turned on. Reply
  • atatassault - Tuesday, February 13, 2018 - link

    People who buy sub $100 monitors. Reply
  • WorldWithoutMadness - Tuesday, February 13, 2018 - link

    Just google GDP per capita and you'll find huge market for 720p budget gaming pc. Reply
  • Sarah Terra - Wednesday, February 14, 2018 - link

    Wow, i just came here after not visiting in ages, really sad to see how far this site has fallen.

    Ian Cutress was the worst thing that ever happened to Anandtech.

    At one point AT was the defacto standard for tech news on the web, but now it has simply become irrelevant.

    Unless things change i see AT slowly but surely dying
    Reply
  • mikato - Thursday, February 15, 2018 - link

    Me. My TV is 720p and still kicking after many years. These CPUs would make for a perfect high end HTPC with some solid gaming ability. Awesome. Reply
  • nevcairiel - Tuesday, February 13, 2018 - link

    Some more realistic gaming settings might be nice. Noone is going to play on settings that result in ~20 fps, and the GPU/CPU scaling can tilt quite a bit if you reduce the settings.

    I can see why you might not like it, because it takes the focus away from the GPU a bit and makes comparisons against a dGPU harder (unless you run it on the exact same hardware, which might mean you have to re-run it every time), but this is a combined product, so testing both against other iGPU products would be useful info.
    Reply

Log in

Don't have an account? Sign up now