Downfall

Resolution Scaling

The measured frames per second across the three resolutions reveal a classic GPU-bound scaling pattern, with the CPU headroom only becoming apparent at the lowest settings. At 1080p with Low settings, the AMD Ryzen 9 7900X + NVIDIA GeForce RTX 4090 combination delivers 442 FPS. Moving to 1440p Low drops the average to 317.4 FPS, a 28.2% reduction, while 4K Low lands at 199.3 FPS, which is a further 37.2% decline from the 1440p figure. This steep, consistent falloff from 1080p to 4K indicates that the rendering workload is scaling directly with pixel count, a hallmark of a system where the GPU is the primary limiting factor.

The pattern holds across higher settings, though the percentage drops between resolutions become less severe. At High settings, the system produces 299.4 FPS at 1080p, 210.8 FPS at 1440p, and 133.7 FPS at 4K. The transition from 1080p to 1440p costs 29.6% of the frame rate, while the jump from 1440p to 4K costs another 36.6%. This suggests that the RTX 4090's 24 GB of GDDR6X memory and 1.01 TB/s bandwidth are being utilized progressively more as resolution increases, but the GPU's compute resources—82.58 TFLOPS FP32—remain the bottleneck even at 4K.

The most telling data point for CPU involvement comes from the Low settings tier. At 1080p Low, the 442 FPS average is extraordinarily high, but the 12-core, 24-thread Ryzen 9 7900X with a boost clock of 5.60 GHz is clearly keeping pace. The fact that 1440p Low still achieves 317.4 FPS—a 28.2% drop—while 4K Low falls to 199.3 FPS shows that even at the lowest graphical load, the GPU is still the dominant constraint. If the CPU were the bottleneck, the 1080p-to-1440p delta would be far smaller, as the CPU load would remain nearly constant while the GPU work increases.

Ultra settings present a different story. At 4K Ultra, the average FPS collapses to 83.5, a 58.3% reduction from 4K High's 133.7 FPS. This massive penalty from High to Ultra at the same resolution indicates that the Ultra preset introduces effects that heavily tax the GPU's shading units and ROPs (176 ROPs, 512 TMUs). The 1080p Ultra figure of 188.9 FPS versus 4K Ultra's 83.5 FPS shows a 55.8% loss, which is proportionally similar to the Low setting's resolution scaling, reinforcing that the RTX 4090's architecture—Ada Lovelace on a 5 nm process—is the performance ceiling regardless of preset.

FAQ

Q: What is the highest average FPS this combo achieves?

A: The highest measured average is 442 FPS at 1920x1080 with Low settings. This is the only configuration in the data exceeding 400 FPS.

Q: How much performance is lost when moving from 1080p to 4K at Ultra settings?

A: At Ultra, the 1080p average is 188.9 FPS, while 4K drops to 83.5 FPS. This represents a 55.8% reduction in frame rate, the largest proportional loss across any settings tier.

Q: Is the Ryzen 9 7900X holding back the RTX 4090 in this game?

A: The data does not indicate CPU limitation. At 1080p Low, the system delivers 442 FPS, and the scaling from 1080p to 1440p (a 28.2% drop) mirrors the pixel count increase, suggesting the GPU is the limiting component even at low settings.

Q: Which resolution provides the most balanced experience?

A: 2560x1440 at High settings yields 210.8 FPS, which is 57.6% higher than the 4K High figure of 133.7 FPS while still maintaining a high frame rate. This resolution offers a substantial margin over 60 FPS with no compromise from the Ultra preset's heavy penalties.

Q: How does the Ultra preset compare to High at the same resolution?

A: At 4K, Ultra averages 83.5 FPS versus High's 133.7 FPS, a 37.5% penalty. At 1440p, Ultra is 131.8 FPS versus High's 210.8 FPS, a 37.5% penalty as well. The relative cost of Ultra is consistent across resolutions.

Q: What is the minimum FPS recorded?

A: The dataset does not include minimum FPS values; only average FPS figures are provided for each resolution and settings combination.

CPU Role

The AMD Ryzen 9 7900X is a 12-core, 24-thread processor built on the Zen 4 architecture with a 5 nm process. Its base clock of 4.70 GHz and boost clock of 5.60 GHz provide the single-thread performance necessary for high-FPS gaming, as evidenced by its 3DMark single-thread score of 1093 and Cinebench R23 single-core score of 6168. In Downfall, a horror game that does not appear to be heavily multi-threaded based on the FPS data, the CPU's 12 cores are more than sufficient to feed the RTX 4090.

The benchmark results for this combo show that the CPU is not the bottleneck at any resolution or settings level. The 1080p Low figure of 442 FPS is the strongest evidence, as this setting places minimal load on the GPU while demanding maximum frame submission from the CPU. The Ryzen 9 7900X's 64 MB of shared L3 cache and dual-channel DDR5 memory support with 83.2 GB/s bandwidth contribute to its ability to sustain such high frame rates.

However, the data also reveals that the CPU's advantage diminishes as resolution increases. At 4K Ultra, the 83.5 FPS average is so far below the CPU's capability that the processor's performance characteristics—such as its 3DMark max-threads score of 12536—become irrelevant. The game's engine appears to scale with GPU load rather than CPU thread count, which is typical for a title from 2016. The Ryzen 9 7900X's 24 threads are never fully utilized; the passmark multithread score of 51406 suggests headroom that Downfall cannot exploit.

The CPU's percentile ranking of 93 among all CPUs, with an average benchmark score of 50556, places it in the top tier of processors. Its nearest rival, the Intel Core Ultra 5 235, scores 50598 (a delta of -0.1%), indicating that the 7900X is essentially tied with that chip in synthetic benchmarks. This parity in raw CPU performance, combined with the GPU-bound nature of Downfall, means that the CPU choice has minimal impact on the measured FPS beyond ensuring the GPU is not starved.

How This Combo Ranks

In Downfall, this combination of AMD Ryzen 9 7900X and NVIDIA GeForce RTX 4090 ranks 23rd out of 1,741 tested combos. This places it in the top 1.3% of all configurations in the database, a strong showing for a desktop CPU paired with a flagship GPU. The rank reflects the data as measured, with no min or max FPS values to skew the average.

The 23rd position out of 1,741 might seem low for a 12-core CPU and the top-tier RTX 4090, but it is important to note that the ranking is based on average FPS across all tested settings and resolutions. The RTX 4090's performance in Downfall is hampered by the game's age (released in 2016), which likely does not utilize the GPU's full feature set—DirectX 12 Ultimate (12_2), ray tracing cores (128), or tensor cores (512). The GPU's passmark G3D score of 38194 and percentile rank of 91 indicate it is a very fast card, but the game's engine may not scale perfectly with such modern hardware.

Comparatively, the CPU's nearest rivals in synthetic benchmarks—like the Intel Core i9-13980HX with a delta of -0.8%—would likely produce nearly identical FPS in this game, given that the GPU is the bottleneck. The combo rank of 23 suggests that the few combos ranked higher likely pair the RTX 4090 with CPUs that have even stronger single-thread performance, though the 7900X's 3DMark single-thread score of 1093 is already excellent. The data shows that this combo sits in the upper echelon, with the 1,718 combos below it all delivering lower average FPS in Downfall.

Settings Recommendations

Based on the measured FPS rows, the optimal settings depend on the target display resolution. At 4K, the High preset delivers 133.7 FPS, which is 60.2% higher than Ultra's 83.5 FPS. The jump from High to Ultra costs 50.2 FPS, a penalty that most users would find difficult to justify unless they have a 120 Hz or higher refresh rate monitor that can take advantage of the extra frames. The Medium preset at 4K offers 158.4 FPS, which is 18.5% above High, but the graphical quality reduction may not be worth the gain.

At 1440p, the High preset at 210.8 FPS is the sweet spot. It provides 60% more performance than Ultra's 131.8 FPS while still maintaining a frame rate that exceeds any current high-refresh-rate display's capabilities. The Low preset at 1440p hits 317.4 FPS, but the visual degradation is likely too severe for a horror game where atmosphere matters. Medium at 1440p (262 FPS) is a viable alternative if the user prefers a balance between quality and speed.

For 1080p, the data shows that even Ultra at 188.9 FPS is remarkably playable. The High preset at 299.4 FPS is within 6.7% of Medium's 362.4 FPS, making High the recommended choice for 1080p due to its superior visual fidelity at a still-very-high frame rate. The Low preset at 442 FPS is only relevant for competitive play, which is not typical for a horror game. Across all resolutions, the High preset offers the best balance of image quality and performance, avoiding the steep penalty of Ultra while delivering well above 60 FPS.

Measured FPS Breakdown

The complete dataset provides 12 measured configurations, covering three resolutions and four settings tiers. At 3840x2160, the average FPS ranges from 83.5 (Ultra) to 199.3 (Low), with High at 133.7 and Medium at 158.4. The progression from Low to Ultra at 4K shows a 58.1% reduction in frame rate, with the largest single drop occurring between High and Ultra (50.2 FPS).

At 2560x1440, the figures are consistently higher: Low averages 317.4 FPS, Medium 262 FPS, High 210.8 FPS, and Ultra 131.8 FPS. The delta between Low and Ultra at 1440p is 185.6 FPS, a 58.5% decrease, nearly identical to the 4K scaling. This consistency suggests that the settings penalties are resolution-independent in percentage terms.

At 1920x1080, the performance ceiling is evident: Low hits 442 FPS, Medium 362.4 FPS, High 299.4 FPS, and Ultra 188.9 FPS. The gap between 1080p and 1440p at Low settings is 124.6 FPS, while the gap between 1440p and 4K at Low is 118.1 FPS. These nearly equal drops indicate a linear scaling with pixel count, reinforcing that the RTX 4090's rendering pipeline is the sole limiter. The highest single measurement is 1080p Low at 442 FPS, while the lowest is 4K Ultra at 83.5 FPS, a 5.3x spread across the entire test matrix.