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

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 GTX1080 Data

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 minutes, where the Wii itself scores 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

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 1.0: 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 1.0 Total Time

Benchmark Overview Benchmarking Performance: CPU Rendering Tests
Comments Locked

140 Comments

View All Comments

  • Gavin Bonshor - Thursday, July 27, 2017 - link

    One of the hardest working men in the industry! :D
  • edlee - Thursday, July 27, 2017 - link

    I dont understand the point of making a $100 cpu without an integrated gpu if you wanted to attract the lower end market, this is really silly mistake. Sort of like intel including an integrated gpu with i7-7700k, it doesnt make sense, 95% of those with a 7700k will buy a gpu, but someone who is looking for a lowend cpu is not going to buy a discrete graphics cards, its just silly
  • phoenix_rizzen - Thursday, July 27, 2017 - link

    It really depends on the use case.

    For example, are there any integrated GPUs that support 3 monitors? I know a lot of them support dual monitors, but haven't come across any that support 3 (although I haven't looked that hard). My work PC is a low-profile desktop running an AMD Athlon-II x4 CPU and an Nvidia 730 GT GPU for tri-monitor setup. Upgrading the CPU/motherboard/RAM to a Ryzen 3 1300X would be a huge upgrade for this system.

    90-odd % of the desktops in the schools here use AMD Athlon-II CPUs (graphics integrated into the chipset), with the rest using Intel Pentium CPUs (graphics integrated into the CPU). And we add Nvidia 210 or 730 GPUs to those that need better multi-monitor support or better 3D performance. Why do we do it that way? Cost. We try to keep the complete desktop system (case, motherboard,
    CPU, at least 2 GB RAM, no storage of any kind) to under $200 CDN (they're diskless Linux stations). We have just shy of 5000 of those in the district right now.

    We've avoided the Bulldozer-based APUs so far as the price/performance just wasn't there compared to the Pentium line (from our suppliers). But the Ryzen 3 looks like a decent upgrade. Will be interesting to see what the prices are like for it from our suppliers this winter/spring. Will also be interesting to see what the GPU side of the Zen-based APUs will be like next year.

    The other important bit is driver support. We are a mostly Linux-using school district, so we tend to use hardware that's at least 2 steps back from the bleeding edge. That way, we get better prices, and better driver support.
  • edlee - Thursday, July 27, 2017 - link

    i understand when upgrading from integrated to gpu like you stated in your use case, but from the low end price standpoint, a i3-7100 is cheaper because they dont need to add a gpu like the ryzen 3 needs, so its not competing on a performance standpoint or a price standpoint when you add the price of the cheapest gpu
  • Outlander_04 - Friday, July 28, 2017 - link

    Using an integrated gpu is usually a poor choice. Intels drivers are so dumbed down they are worse than hopeless.
    Factor in that using integrated means less system RAM available as well so performance can be reduced
  • Ratman6161 - Tuesday, August 1, 2017 - link

    Many people may be starting out from the position of knowing that the integrated graphics on any of the Intel CPU's in the test are not good enough for them. If you know that from the start then the argument that AMD doesn't have an IGPU is meaningless. I'm also somewhat interested in seeing overclocking tests with the R3 as that is one thing you just don't get with Intel at this level short of the 7350K. I sort of suspect that an OC'd 1200 could but just as fast or faster than a 1300X (though at only a $20 difference I'm not sure how much it matters).
    Also, in more computationally intense tasks, the 1300x really doesn't do badly against the i5 that costs $53 more so once again, if you don't care about integrated graphics it could be a good choice for some people.

    On the other hand, for someone for whom MS Office, email, and web browsing are their main uses, then something like the i3-7100 suddenly looks very attractive - or even the Pentium G.
    In this segment, AMD really needs to get a Ryzen Based APU on the market. If they did a single CCX, 4 core and used the empty space vacated by the second CCX for a decent IGPU they could definitely have an i3 killer.
  • renw0rp - Thursday, July 27, 2017 - link

    I had HP Folio 9470m with core i5-3437U and it was driving 3 * 1920x1200 screens without an issue. And it's ~2013 processor...

    3rd gen of Core processors was the first to support 3 displays. The 2nd gen supported just 2.
  • stuartlew - Thursday, July 27, 2017 - link

    AMD Kaveri does three monitors
  • serendip - Friday, July 28, 2017 - link

    Are there motherboards with integrated chipset graphics for Ryzen?

    I understand the good thing about adding a discrete GPU only to PCs that need one but not having an integrated GPU is nuts, for the mass market at least.
  • silverblue - Friday, July 28, 2017 - link

    No, but Bristol Ridge launched yesterday, so there are now APUs that use AM4.

Log in

Don't have an account? Sign up now