Launching the #CPUOverload Project: Testing Every x86 Desktop Processor since 2010
by Dr. Ian Cutress on July 20, 2020 1:30 PM ESTGaming Tests: Chernobylite
Despite the advent of recent TV shows like Chernobyl, recreating the situation revolving around the 1986 Chernobyl nuclear disaster, the concept of nuclear fallout and the town of Pripyat have been popular settings for a number of games – mostly first person shooters. Chernobylite is an indie title that plays on a science-fiction survival horror experience and uses a 3D-scanned recreation of the real Chernobyl Exclusion Zone. It involves challenging combat, a mix of free exploration with crafting and non-linear story telling. While still in early access, it is already picking up plenty of awards.
I picked up Chernobylite while still in early access, and was impressed by its ingame benchmark, showcasing complex building structure with plenty of trees and structures where aliasing becomes important. The in-game benchmark is an on-rails experience through the scenery, covering both indoor and outdoor scenes – it ends up being very CPU limited in the way it is designed. We have taken an offline version of Chernobylite to use in our tests, and we are testing the following settings combinations:
- 360p Low
- 1440p Low,
- 4K Low
- 1080p Max
For automation purposes, the game has no flags to initiate benchmark mode. We delete the movies from the install directory to speed up entering the game, and use timers and keypresses to start the benchmark mode. The game puts out a benchmark results file, however this only shows average frame rates, not frame times. In-game settings are controlled by copying pre-arranged .ini files into the relevant location. We do as many runs within 10 minutes per resolution/setting combination, and then take averages.
| AnandTech | IGP | Low | Medium | High |
| Average FPS |  |
 |
 |
 |
All of our benchmark results can also be found in our benchmark engine, Bench.
Facebook
Twitter
RSS
110 Comments
View All Comments
PeachNCream - Tuesday, July 21, 2020 - link
You don't get what it means to perform a controlled test do you?Aspernari - Wednesday, July 22, 2020 - link
It's important to note that the environment is not actually well-controlled.https://twitter.com/IanCutress/status/128480609693...
We don't know temperature for the operating conditions for these tests, which matters more and more for boost behavior for CPUs and GPUs. He says 36c when he got into the office, we'll never know what the temperature peaked at, nor how often similar conditions were reached.
A standard platform is a good choice, but a controlled environment is also important. Unfortunately, the results aren't as reliable as they otherwise might have been.
PeterCollier - Wednesday, July 22, 2020 - link
And that's why this entire test is a complete waste of time. Something like Geekbench or especially Userbench is much, much better because it gives you a range of scores. Instead of trying to create false precision by saying that a AMD 4700U scored, say, a "979" on a benchmark, Userbench will say that all the 4700U's tested scored from 899 to 1008, and break it down into percentiles. This way, you have a range of expected performance in mind instead of being fixated on that "979" number, which could have been obtained in an unrealistic scenario.Rudde - Saturday, July 25, 2020 - link
Isn't userbench a synthetic together with geekbench? What exactly are they testing? Instead of knowing which of Intel i7 10700k and AMD ryzen 7 3800X is better at rendering, video encoding, number crunching or whatever your use case is, you'll get a distribution based on a largely unknown test. The Intel and AMD processors might end up being within error margins of each other in your use case, but that in itself tells something too. All benchmarks are inherently bad; there is not a single benchmark that captures every use case while not being affected by its environment (ram speeds, temperatures, etc). I prefer tests that I understand, over tests that I do not understand.bananaforscale - Wednesday, July 22, 2020 - link
One could ask what the point of Userbenchmark is in these days of quadcores being basically entry level while the benchmark has DECREASED its multicore weighting.A5 - Monday, July 20, 2020 - link
For my own personal test, getting an i7-4770K in the list would be a big help.Once you have a compile test, a Xeon E5-1680v3 would be nice to see so that I can sell my corp on newer workstations...
Shmee - Wednesday, July 22, 2020 - link
Those are great Haswell EP CPUs, and they OC too! I have an E5-1660v3 in my X99 rig.Mockingtruth - Monday, July 20, 2020 - link
I have a 3570k and a E8600 spare with respective motherboards and ram if useful?CampGareth - Monday, July 20, 2020 - link
Personally I'd like to see a Xeon E5-2670 v1 benchmarked. I'm still running a pair of them as my workstation but these days AMD can beat the performance on a single socket and halve the power consumption.Samus - Tuesday, July 21, 2020 - link
Do you run them in an HP Z620? I ran the same system with the same CPU’s for years at one of my clients. What a beast.