Unreal Engine: AMD Ryzen 5000 Series CPU PerformanceWritten on November 5, 2020 by Kelly Shipman
TL;DR: AMD Ryzen 5000 Series Performance in Unreal Engine
AMD's previous-gen Ryzen CPUs already provided a compelling option for Unreal Engine. The newly release Ryzen 5000 series CPUs have made it difficult to recommend anything else. They offer a 10-20% improvement over the previous generations and outperform any Intel CPU at similar price points or core counts.
AMD's Threadripper is still the top choice, taking the top three spots in every test. However, if you are looking for a lower-priced system or something in a smaller form factor, AMD's new Ryzen 5000 series is the way to go.
AMD's latest mainstream processor family, the Ryzen 5000 Series, has launched with direct upgrades to several (but not all) of their most popular Ryzen 3rd Gen chips. These new models cover from 6 to 16 cores and utilize the updated Zen 3 microarchitecture which brings substantial improvements to instructions per clock (IPC). Even though core counts and cache sizes are not increasing, the overall performance of these new CPUs is dramatically faster than both previous generations and competing processor designs.
For this article, we are looking at how the new AMD Ryzen 5000 Series processors compare to a host of other currently-available models: AMD's previous Ryzen 3000 Series, their high-end desktop (HEDT) Threadrippers, and both Intel's mainstream Core and HEDT Core X lines. We’ll look at major development tasks within the engine, as well as look at how the CPU impacts frame rates.
If you would like to skip over our test setup and benchmark sections, feel free to jump right to the Conclusion.
Looking for an Unreal Engine Workstation?
Puget Systems offers a range of powerful and reliable systems that are tailor-made for your unique workflow.
Listed below are the specifications of the systems we will be using for our testing:
|AMD Ryzen Test Platform|
|CPU||AMD Ryzen 9 5950X ($799)
AMD Ryzen 9 5900X ($549)
AMD Ryzen 7 5800X ($449)
AMD Ryzen 5 5600X ($299)
AMD Ryzen 9 3950X ($749)
AMD Ryzen 9 3900XT ($499)
AMD Ryzen 7 3800XT ($399)
AMD Ryzen 5 3600XT ($249)
|CPU Cooler||Noctua NH-U12S|
|Motherboard||Gigabyte X570 AORUS ULTRA|
|RAM||4x DDR4-3200 16GB (64GB total)|
|AMD Threadripper 3rd Gen Test Platform|
|CPU||AMD TR 3990X ($3,990)
AMD TR 3970X ($1,999)
AMD TR 3960X ($1,399)
|CPU Cooler||Noctua NH-U14S TR4-SP3|
|Motherboard||Gigabyte TRX40 AORUS PRO WIFI|
|RAM||4x DDR4-3200 16GB (64GB total)|
*All the latest drivers, OS updates, BIOS, and firmware applied as of October 26, 2020
One thing we do want to note is that the pre-launch BIOS that is available for Ryzen motherboards is using AGESA 1.0.8. Soon after launch, there should be an update that adds support for AGESA 1.1.0 which is supposed to increase the performance of each Ryzen CPU by another few percent.
Benchmark Analysis: Compiling Source Code
This test isn’t specific to Unreal Engine, but is common enough amongst game developers and machine learning development to include it in this round of testing. For this, we used the “Release” branch (version 4.25) from GitHub and compiled in Visual Studio 2019.
As you can see from the graph, compiling this sort of code scales very well based on available CPU cores. Interestingly, if we look at CPUs with the same core count, AMD comes out ahead in every instance. The new Ryzen 5000 series extends that lead even further. The 12 core Ryzen 9 5900X even manages to outperform Intel’s flagship 18 core i9 10980XE while being half the price. Naturally, AMD’s Threadripper takes the top three spots due to its massive core counts.
Benchmark Analysis: Build Lighting
While more and more Unreal users are turning to Unreal for its realtime lighting features, traditional rasterized rendering still has a lot of use. Anyone that has used Unreal for a while can attest to how much time is spent waiting for lighting to be rebuilt. For this test, we used “Medium” lighting quality.
This test still heavily leans towards multi-core performance. And will vary depending on the scale of your scene, the lighting complexity, etc. Clock speed can help (see 10900K outperforming 12 core CPUs), but generally more cores will speed up the process much more effectively. The Ryzen 5950X is roughly 9% faster than the previous generation and 12% faster than Intel’s 10980XE.
Benchmark Analysis: Recompile Shaders
Recompiling shaders is something that can take a lot of time depending on the workflow. Smaller projects or virtual production or architecture may not have to rebuild all their shaders very often, if at all. On the GameDev side, they may need to rebuild all their shaders quite often. Not everyone on the team, but someone will be doing this often
Once again, core count makes the biggest difference here. For this test, the scene contained roughly 6500 shaders. Both the Ryzen 5950X and 5900X beat out all of Intel’s offerings. The 5800X even pulls off a significant lead over the Intel 10900K. Once again, Threadripper takes the top three places.
Benchmark Analysis: Package Project
The final major CPU test is Packaging Projects. This means taking all the assets associated with the project and assembling them into a standalone executable. This combines several different tasks that need to be performed. Some heavily threaded, some lightly threaded. Unreal caches a lot of this data to make subsequent packaging much faster. So between each round, all that data had to be scrubbed. These results represent the first pass. Depending on how each individual’s workflow, this could be a big-time sink or not much time at all.
Yet again, the Ryzen 5950X outperforms the 10980XE. The 5900X trails Intel’s top offering, but barely. And the 5800X is neck and neck with the Intel 10900K. Threadripper is still at the top, but the margin has narrowed.
CPUs and FPS.
The relationship between CPUs and Frames Per Second is complicated. Exactly how much of an influence the CPU has on FPS depends on a lot of factors. For example, it is the CPU’s job to shuttle data from the hard drive to the video card, it's also controlling AI, calculating hit detection and projectiles, player input, etc. etc. For some use cases, such as Virtual Production or ArchViz, these really aren’t an issue. Also, as screens keep growing in pixel count, with 4k becoming more and more common, the burden of work keeps getting pushed more and more onto the GPU.
Let's take a look at the exact same scene, rendered at 1080p with no ray-tracing effects, and then again at 4k with full ray tracing. In our GPU tests, we saw this scene get limited by the CPU. Every GPU we tested came back with the exact same frame rates. So taking those settings at applying them to various CPUs, we can see what effect the CPU has on the max FPS possible.
In this instance, the clock speed of the CPU is the biggest factor. Among Intel CPUs, the 10900K has a big lead. However, the speed improvements that AMD has implemented in the 5000 series catapulted their new CPUs to the top of the charts. This improvement made me rerun the FPS tests on both AMD platforms because I was shocked by how well they did.
Now let's look at what happens when we up the resolution to 4k and enable ray-tracing effects. As you can see, the graph flattens out. Some of the AMD CPUs are still near the top, but all the results are within a couple of frames, essentially within the margin of error of each other.
Given all this, I’ve not included FPS in the overall CPU score. I do not know of a use case, outside of competitive gaming, that would benefit from these kinds of frame rates at such a low resolution. If that is a metric you use professionally, please let me know and I’ll look into adding it in future versions. I did want to include this section to show that at higher resolutions, and with more complex visuals, the CPU makes little difference. If you need the Threadripper 3990X for the other tasks reviewed above, you don't necessarily have to worry that you are sacrificing FPS.
Development within Unreal Engine wants more cores above all else. In every task, the high core count of AMD’s Threadripper CPUs chews through these processes. Just behind Threadripper is AMD’s new 5950X. This CPU is able to beat every Intel CPU including the i9 10980XE, which has 2 more cores. Just behind the 10980XE is the Ryzen 5900X, a very impressive performance for a CPU that costs half as much.
Continuing down the list, it becomes very difficult to recommend any Intel CPU for Unreal Engine, especially once we take cost into consideration. AMD’s CPUs manage to do more, with fewer cores, at a lower price point. While Threadripper is the undisputed winner in every test, the new Ryzen 5000 Series CPUs provide amazing performance.
Are the AMD Ryzen 5000 Series or Intel Core 10th Gen better for Unreal Engine?
in only a few short years, AMD has grown the Ryzen line from good budget CPUs that didn't quite compete on the top end, to a strong contender in many applications. Now with the Ryzen 5000 series, they have become the top option in most applications. We also tend to prefer hardware that has a track record of reliability and support in professional workloads. AMD has made huge strides with these applications, and we've sold enough of the previous generation to feel confident in their reliability.
Unreal Engine is an application that will leverage as many CPU cores as you throw at it. As such, Threadripper is the best option for power users, offering unmatched performance in any test. For those that want a system that isn't quite so expensive, or something physically smaller, AMD's new 5000 series CPUs offer more raw performance than Intel can provide.
Unreal Engine Workstations
Puget Systems offers a range of poweful and reliable systems that are tailor-made for your unique workflow.