Downfall

The benchmark data for Downfall pairs the Intel Core i3-12100F with the AMD Radeon RX 570, a combination that lands near the bottom of the database’s tested pool. The measured frame rates across resolutions and settings reveal a system that is consistently constrained by the graphics card, though the CPU’s four cores and eight threads prove sufficient for this horror title. The following sections break down the results step by step.

Resolution Scaling

The frame rate drop from 1080p to 4K is steep across every preset, which points directly to the RX 570 as the limiting component. At Low settings, the average FPS falls from 92 at 1920x1080 to 68 at 2560x1440, a reduction of 24 frames, and then to 43 at 3840x2160, which is less than half of the 1080p figure. This pattern intensifies at High settings, where the average drops from 59.5 at 1080p to 45.8 at 1440p and finally to 25.6 at 4K. The scaling is not linear; the gap between 1440p and 4K is larger than the gap between 1080p and 1440p, indicating that the GPU’s pixel throughput is exhausted as resolution climbs.

At Ultra settings, the situation becomes more pronounced. The 1080p average of 36.5 FPS falls to 28.1 at 1440p and then to 18.6 at 4K. The 4K Ultra result is the lowest in the entire measured set, and it sits well below playable thresholds for a horror game that relies on atmosphere and reaction time. The data suggests that the RX 570’s 4 GB of GDDR5 memory and 224.0 GB/s bandwidth are simply overwhelmed by the pixel count at 3840x2160, regardless of the preset. Even at Low settings, 4K only reaches 43 FPS, which is playable but far from smooth.

The scaling behavior also shows that the i3-12100F is not the bottleneck in this pairing. If the CPU were the limiting factor, the FPS would remain relatively flat as resolution increases, because the same number of frames would be processed regardless of pixel count. Instead, the averages drop consistently with each resolution step, which is the classic signature of a GPU-bound workload. The CPU’s single-thread score of 893 in 3dmark and a PassMark single-thread score of 3444 are adequate for this game’s demands, leaving the graphics card to determine the ceiling.

How This Combo Ranks

In the overall database of tested combinations, this pairing ranks 1656 out of 1741 combos. That places it in the bottom 5% of all tested systems, which is consistent with the hardware’s age and tier. The RX 570 is an end-of-life product from the Polaris generation, while the i3-12100F is a modern budget chip, yet the combination still struggles to reach high frame rates in Downfall. The rank is driven almost entirely by the GPU, as the CPU’s percentile of 72 against all tested CPUs is respectable, while the GPU’s percentile of 74 against all GPUs is marred by the fact that the RX 570’s average benchmark score of 30158 is barely above its nearest rivals.

Looking at the nearest rivals for the CPU, the i3-12100F posts an average benchmark score of 13516, which is only 0.1% below the AMD Ryzen Threadripper PRO 3975WX and 1.8% above the Intel Core i7-7700K. This tells us that the processor is not the weak link in this pairing; it performs on par with much more expensive chips in synthetic tests. The GPU, however, sits in a strange position. Its average benchmark score of 30158 is just 0.2% above the NVIDIA GeForce RTX 5060 and 0.8% above the NVIDIA GeForce RTX 5070 Mobile, which are far more modern parts. This suggests that the RX 570’s raw compute is still competitive in synthetic workloads, but its 4 GB memory capacity and older architecture hold it back in real games like Downfall.

GPU Role

The RX 570’s specifications explain why it struggles at higher resolutions. It has 2048 shading units, 128 texture mapping units, and 32 render output units, with a base clock of 1168 MHz and a boost clock of 1244 MHz. The memory runs at 1750 MHz with a 256-bit bus, yielding 224.0 GB/s of bandwidth across 4 GB of GDDR5. These numbers are modest by modern standards, and the 4 GB capacity is particularly limiting for a game released in 2016, which may have texture pools that exceed that amount at higher presets.

The pixel rate of 39.81 GPixel/s and texture rate of 159.2 GTexel/s are direct indicators of how many pixels the card can fill each second. At 4K, the card must push over 8 million pixels per frame, and the data shows that even at Medium settings, the average FPS drops to 34.2, which is a 26% reduction from the 1440p Medium result of 54.9. The FP32 performance of 5.095 TFLOPS is the same as the FP16 figure, meaning there is no half-rate advantage to exploit. The card supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3, so it can run the game’s modern APIs, but the hardware simply lacks the throughput for high resolution.

Settings Recommendations

The measured data points to Low settings as the only reliable choice for smooth gameplay at 1080p. At 1920x1080 with Low settings, the average FPS is 92, which is the highest result in the entire dataset. This provides ample headroom for a horror game where sudden movements and dark scenes require consistent frame pacing. Medium settings at 1080p deliver 74.2 FPS, which is still playable, but the drop from Low to Medium is 17.8 FPS, suggesting that the preset’s added effects strain the GPU significantly.

For users with a 1440p monitor, Low settings are the only option that approaches smoothness, with an average of 68 FPS. Medium at 1440p falls to 54.9, which is borderline, and High drops to 45.8, which is noticeably choppy. At 4K, no preset is truly playable; even Low only reaches 43 FPS, and the experience will feel sluggish. The data suggests that the best experience is 1080p with Low settings, as it maximizes the frame rate while keeping the GPU within its comfort zone. Ultra settings should be avoided entirely, as the 1080p average of 36.5 FPS is below the threshold for responsive gameplay.

Measured FPS Breakdown

The complete set of measured results shows a clear hierarchy across resolutions and presets. At 1920x1080, the averages are 92 FPS for Low, 74.2 for Medium, 59.5 for High, and 36.5 for Ultra. The gap between Medium and High is 14.7 FPS, while the gap between High and Ultra is 23 FPS, indicating that the Ultra preset introduces effects that are disproportionately expensive for this GPU.

At 2560x1440, the averages are 68 FPS for Low, 54.9 for Medium, 45.8 for High, and 28.1 for Ultra. The step from Low to Medium costs 13.1 FPS, and the step from High to Ultra costs 17.7 FPS. The pattern is consistent: each preset tier reduces performance by roughly 15-20% at this resolution. At 3840x2160, the averages are 43 FPS for Low, 34.2 for Medium, 25.6 for High, and 18.6 for Ultra. The drop from Low to Ultra at 4K is 24.4 FPS, which is the largest absolute difference in the dataset. This confirms that the GPU’s rendering pipeline is saturated at the highest resolution, and the CPU has little influence on the outcome.

CPU Role

The Intel Core i3-12100F has four physical cores and eight threads, with a base clock of 3.30 GHz and a boost clock of 4.30 GHz. It draws 58 W of power and uses the Alder Lake architecture on a 10 nm process. The cache hierarchy includes 80 KB of L1 per core, 1.25 MB of L2 per core, and 12 MB of shared L3, which is a solid configuration for a budget desktop part. In synthetic benchmarks, the CPU scores 11912 in Cinebench R23 multicore and 1681 in single-core, with a Geekbench multicore score of 7413 and a single-core score of 2224. These numbers are competitive with much more expensive chips, as shown by its nearest rivals.

For Downfall, the CPU’s role is secondary. The game’s genre is horror, which typically involves limited on-screen actors and simple physics, so the four cores are unlikely to be saturated. The 3dmark results show a 4-thread score of 2859 and an 8-thread score of 4026, which indicates that the CPU scales well up to its thread count. However, the measured FPS data shows no evidence of CPU limitation, as the frame rates are uniformly low at high resolutions regardless of the CPU’s capabilities. The combination of a strong budget CPU and an older mid-range GPU results in a system where the GPU is always the bottleneck, and the CPU has plenty of headroom to spare.