We always try to offer you a test course with the most current content possible. With a total of 16 games, the new game course therefore consists of more content than ever before, as not only the core optimization but also the graphical interfaces have evolved. Especially when a low-level API like Vulkan is used, the CPU performance increases unimaginably. In the new index you will find games that respond very well to many CPU cores and that scale them very well with a very high clock frequency. Some of them generally run better on Intel processors, others on their respective AMD counterparts. We want to do justice to the two chip giants and show a fair comparison. In addition, four new applications have been introduced that allow us to test the processors under extreme conditions. The number of cores plays a much larger role in programs than in games and the power consumption is also much higher because of the strong parallelization.
Before the benchmark runs, each CPU is subjected to a warm-up period during which the core clock (boost) drops to a certain level that varies from model to model. We configure a TDP limit in the UEFI: Each CPU is configured to the parameters that the manufacturer provides for the respective model. The speed of the built-in memory is also forced to meet the manufacturer's specifications, we specify the amount with 32 GiB and the timings (XMP). In this way we avoid unrealistically high values and enable a fair comparison. If possible, we always use the latest version of Windows including all (security) updates. All measurements are made using CapFrameXcarried out. Each benchmark run is repeated at least three times. The average value from all three runs is entered in the index. We reserve the right to ignore measurements that are out of line. The resolution for the games is usually 1,280 x 720 pixels (720p) , the details are at the highest level, but we do without (MS) AA and other purely graphic effects.
As usual, we test highly demanding scenes in current but stable patched games. All index values are standardized, i.e. set in relation to the respective maximum. This is necessary so that tests with high results do not influence the index more than tests with low Fps values.
Why tests in "unrealistic" 720p?
Dear colleague Raffael Vötter explained it aptly
in his comment , here is an excerpt:
"There are good reasons for the 1,280 × 720 pixels. Most games have the lowest possible resolution in the 16: 9 aspect ratio. The latter is by an overwhelming margin the most common column-to-line ratio in computer and console games The aspect ratio determines the visible section in games, so it is by no means irrelevant. You could also reactivate a classic from the 90s and measure it in 640 × 480 instead. This would, however, lead to an aspect ratio of 4: 3 and thus a reduced peripheral field of view If a more important role is assigned to our perception and current image sizes, one remains at 16: 9 and thus the same section as with 1,920 × 1,080, 2,560 × 1,440, 3,840 × 2,160 and many other popular resolutions.
If the latter resolutions are so popular and practical, why do we test processors so hard at 1,280 × 720? Because that's the only way to put the load on what is to be tested: the CPU (and the attached infrastructure around RAM). Yes, we are actually pursuing the crazy idea of only testing the processor and not the graphics card in a processor test. At this point we should all take a short break and think about what has been said. Does this make sense? You nod, I'm sure of that. So further in the text.
A CPU test should reveal how strong the CPU is. Makes sense doesn't it? If we run Cinebench or Handbrake, we use them to test the CPU performance. Nobody complains, everyone is happy about the numbers and bars obtained with them. But nobody asks about the resolution. Have you noticed? This is due to the expectations: application benchmarks are those weird, rather ugly applications in which you don't move around interactively. What do we care about the resolution? This very mindset should become the standard when we talk about processor testing. Even with games! "
Test platform and driver
• CPU cooler: NZXT Kraken X62 + "case fan" for cooling the voltage converter
• Graphics card: MSI RTX 2080 Ti Trio Z (overclocked)
• Power supply: Seasonic Prime Titanium 850W (80-Plus Titanium)
• 16 GiByte RAM per available memory channel. Speed depending on CPU manufacturer specification, timings CL16-18-18-36-1T (XMP)
Driver and operating system
• Current Geforce driver in the standard configuration
• Windows 10, 64-bit version in the latest version
