Control

Control pairs an eight-core Zen 3 processor with NVIDIA’s flagship Blackwell GPU, and the measured frame rates across three resolutions and four settings presets reveal a system that is overwhelmingly GPU-bound at higher fidelity levels, yet surprisingly CPU-sensitive at the lowest settings. The data set contains 12 measured FPS points, each an average, with no minimum or maximum frametime information provided, so the analysis focuses on average throughput and scaling behavior. This combination sits at rank 25 out of 1,636 tested combos in this title, placing it in the top 1.5% of all configurations cataloged for Control.

Resolution Scaling

The most striking pattern in the data is how frame rate collapses as resolution climbs, but only when settings are high. At 1080p Low, the system delivers 228.4 FPS; at 4K Low, it drops to 142.7 FPS, a reduction of 37.5% from the 1080p figure. That is a significant loss, but it is far less severe than the scaling seen at Ultra settings, where 1080p yields 159.1 FPS and 4K falls to 77.2 FPS — a 51.5% drop. This divergence suggests that at Low settings, something other than raw GPU throughput is beginning to limit performance at higher resolutions, while at Ultra settings, the GPU is the clear bottleneck, and its workload scales almost linearly with pixel count.

The intermediate resolutions reinforce this interpretation. At 1440p High, the system produces 137.3 FPS, which is 75.8% of the 4K High result (90.3 FPS) and 75.9% of the 1080p High result (181.1 FPS) — wait, that math is off. Let’s re-examine: 1440p High is 137.3, 1080p High is 181.1, so 1440p delivers 75.8% of 1080p. 4K High is 90.3, which is 65.8% of 1440p. So the scaling from 1440p to 4K is steeper (losing 34.2%) than from 1080p to 1440p (losing 24.2%). This non-linear drop indicates that the GPU’s memory subsystem or shading resources are being taxed progressively harder as pixel count increases, with the 4K Ultra result of 77.2 FPS representing the heaviest load in the entire matrix.

What does this imply about the limiting component? At 1080p Low, the 228.4 FPS figure is high enough that the Ryzen 7 5800X’s single-thread performance (3DMark single-thread score of 943) could be a factor, but the more telling comparison is between Low and Medium at the same resolution. At 1080p, the gap from Low (228.4) to Medium (199.3) is only 12.8%, whereas at 4K, the same settings change produces a 26.0% drop (142.7 to 105.7). The larger relative penalty at 4K confirms that higher settings amplify GPU workload disproportionately as resolution grows, leaving the CPU with relatively more headroom at lower resolutions.

CPU Role

The AMD Ryzen 7 5800X brings 8 cores and 16 threads, built on the Zen 3 architecture with a base clock of 3.80 GHz and a boost clock of 4.70 GHz. Its 3DMark scores show a pattern typical of this generation: single-thread score of 943, 2-thread score of 1,846, and 8-thread score of 6,035, scaling up to 7,578 at max threads. This near-linear scaling from 8 to 16 threads (6,035 to 7,578, a 25.6% gain) suggests the processor has ample multi-threaded headroom for a game like Control, which is not known for heavy CPU utilization across many cores.

The benchmark results place this CPU at the 84th percentile among all tested processors, with an average benchmark score of 27,606. Its nearest rivals are extremely close: the AMD Ryzen 7 5800 scores 27,535 (0.3% higher), the Ryzen 5 8500GE scores 27,708 (0.4% higher), and the Intel Core i7-13700H scores 27,716 (0.4% higher). The Intel Core i5-12600KF is 0.8% ahead at 27,827. These sub-1% differences mean that for Control’s frame rates, swapping to any of these CPUs would produce virtually identical results, reinforcing that the CPU is not the primary constraint in most measured scenarios.

However, the 1080p Low result of 228.4 FPS deserves scrutiny. At this resolution and settings level, the GPU is underutilized, and the CPU’s boost clock of 4.70 GHz becomes more relevant. The Cinebench R23 single-core score of 3,327 indicates strong per-thread performance, which likely helps maintain high frame pacing in less demanding scenes. Yet the fact that 1080p Low (228.4) is only 9.1% higher than 1080p Medium (199.3) — while 4K Low (142.7) is 35.0% higher than 4K Medium (105.7) — suggests that even at 1080p Low, the GPU still has enough work to keep the CPU from becoming the sole bottleneck. The CPU’s role is supportive: it provides sufficient single-thread speed to avoid capping frame rates, but it does not dominate the performance envelope.

GPU Role

The NVIDIA GeForce RTX 5090 is a massive GPU by any measure: 32 GB of GDDR7 memory on a 512-bit bus, yielding 1.79 TB/s of bandwidth, with a boost clock of 2,407 MHz. Its 21,760 shading units and 176 ROPs deliver a pixel rate of 423.6 GPixel/s and a texture rate of 1,636.8 GTexel/s. These specifications explain why the card sits at the 94th percentile among all GPUs, with an average benchmark score of 84,306. Its nearest rivals are essentially identical in synthetic scores: the RTX 5090 D is 0.1% ahead, the RTX 5050 Mobile is 0.2% ahead, and the AMD Radeon PRO W6600 is 1.3% ahead. The NVIDIA CMP 40HX trails by 1.6%.

In Control, the GPU’s role becomes dominant as settings increase. At 4K Ultra, the system produces 77.2 FPS, which is 45.9% lower than the 4K Low result of 142.7 FPS. This 65.5 FPS penalty from Low to Ultra at 4K is the largest absolute drop in the entire data set, and it directly reflects the GPU’s workload from higher-quality textures, shadows, and post-processing effects. The VRAM capacity of 32 GB is far beyond what Control requires, even at 4K Ultra, so memory capacity is not a limiting factor; instead, the GPU’s compute and memory bandwidth are being saturated.

The scaling from Medium to High to Ultra at each resolution shows consistent incremental penalties. At 4K, the steps are 142.7 (Low) → 105.7 (Medium) → 90.3 (High) → 77.2 (Ultra). Each step costs roughly 15-25% performance, with the largest relative jump from Low to Medium (26.0% loss). At 1080p, the same steps are 228.4 → 199.3 → 181.1 → 159.1, with losses of 12.8%, 9.1%, and 12.2% respectively. The GPU’s headroom at 1080p means it can absorb higher settings with smaller relative penalties, whereas at 4K, every setting increase pushes the GPU closer to its limits, producing larger percentage losses.

Measured FPS Breakdown

At 1080p, the system delivers 228.4 FPS on Low, 199.3 FPS on Medium, 181.1 FPS on High, and 159.1 FPS on Ultra. The spread from lowest to highest settings is 69.3 FPS, or 30.3% of the Low result. This is the tightest spread across all resolutions, indicating that the GPU has enough compute headroom at 1080p to render even Ultra settings at playable frame rates.

Moving to 1440p, the results are 209.3 FPS (Low), 154.0 FPS (Medium), 137.3 FPS (High), and 119.9 FPS (Ultra). The Low-to-Ultra spread is 89.4 FPS, a 42.7% drop from Low. This widening gap shows the GPU beginning to feel the pressure of higher pixel counts, but the 119.9 FPS Ultra result remains smooth for most displays.

At 4K, the numbers are 142.7 FPS (Low), 105.7 FPS (Medium), 90.3 FPS (High), and 77.2 FPS (Ultra). The spread is 65.5 FPS, and the Ultra result is the lowest in the entire matrix. Notably, the 4K Low result of 142.7 FPS is still higher than the 1080p Ultra result of 159.1 FPS — actually, it’s lower, so that comparison is inverted. The 4K Low figure is 10.3% below 1080p Ultra, which shows that resolution alone can cost more performance than moving from Low to Ultra at a fixed resolution.

FAQ

Q: What is the highest average FPS recorded in this data set?

A: The highest average FPS is 228.4, achieved at 1920x1080 with Low settings.

Q: How does the RTX 5090 compare to its closest rival in synthetic benchmarks?

A: The RTX 5090 D scores 0.1% higher in average benchmarks, while the RTX 5050 Mobile is 0.2% higher, and the AMD Radeon PRO W6600 is 1.3% higher. The NVIDIA CMP 40HX is 1.6% lower.

Q: Is the Ryzen 7 5800X likely to bottleneck the RTX 5090 in Control?

A: The data suggests minimal CPU bottlenecking. At 1080p Low, the 228.4 FPS result is high, but the CPU’s nearest rival (Ryzen 7 5800) scores only 0.3% higher, and the CPU sits at the 84th percentile, indicating it has sufficient single-thread performance to feed the GPU.

Q: What is the largest performance drop between consecutive settings at any resolution?

A: The largest single-step drop is from Low to Medium at 4K, where performance falls from 142.7 FPS to 105.7 FPS, a 26.0% reduction.

Q: How many tested combos does this system rank above in Control?

A: This combo ranks 25th out of 1,636 tested combos, meaning it outperforms 1,611 other configurations.

Q: Does the GPU’s memory bandwidth appear to be a limiting factor at 4K?

A: The 1.79 TB/s bandwidth is substantial, but the 4K Ultra result of 77.2 FPS suggests the GPU’s shading resources, not just memory bandwidth, are saturated, as the drop from High to Ultra (90.3 to 77.2 FPS, a 14.5% loss) reflects compute-heavy effects.

Settings Recommendations

For the best balance of visual fidelity and frame rate, the data points to 1440p High as the sweet spot. At 137.3 FPS, this setting delivers a frame rate that exceeds 120 FPS on most high-refresh displays, while the visual quality is only one step below Ultra. The 1440p Ultra result of 119.9 FPS is only 12.7% lower, so players with 120 Hz monitors might still prefer Ultra, but the High preset offers a safer margin above refresh-rate thresholds.

At 1080p, Ultra at 159.1 FPS is the clear choice for competitive play, as it provides over 144 FPS with maximum settings. The gap from High (181.1 FPS) to Ultra (159.1 FPS) costs 12.2% performance, but the absolute frame rate remains high enough that Ultra is justifiable for most users. At 4K, the situation is different: High at 90.3 FPS is the most reasonable option, as Ultra drops to 77.2 FPS, which may introduce noticeable dips on 60 Hz displays but is still playable. The Medium preset at 4K (105.7 FPS) offers a 17.1% improvement over High and might be preferred for smoother motion, but it sacrifices visual quality.

The Low presets are not recommended for any resolution, as the frame rate gains over Medium are modest at 1080p (14.6%) and 1440p (35.9%) but come with substantial visual degradation. The only scenario where Low might be considered is 4K, where it reaches 142.7 FPS, but this is 58.8% higher than Ultra — a huge gain that could matter for esports-style play, though Control’s atmosphere benefits from higher settings.

How This Combo Ranks

This combination of the AMD Ryzen 7 5800X and NVIDIA GeForce RTX 5090 ranks 25th out of 1,636 tested combos in Control, placing it in the top 1.5% of all configurations. Given that the GPU sits at the 94th percentile among all GPUs and the CPU at the 84th percentile among all CPUs, this high rank is expected — the GPU’s dominance in the performance equation is clear, as its percentile ranking is 10 points higher than the CPU’s.

The rank implies that only 24 combos achieve higher average frame rates in Control, which likely include newer or more specialized processors paired with the same or faster GPUs. The CPU’s nearest rivals, all within 0.8% in synthetic benchmarks, would not meaningfully alter this ranking, as the delta in Control’s frame rates would be negligible. Conversely, the GPU’s nearest rivals — the RTX 5090 D and RTX 5050 Mobile — are within 0.2% in average scores, suggesting that swapping to those cards would produce nearly identical ranking positions.

The data shows a system that excels at high-refresh 1080p and 1440p gaming, and remains capable at 4K with settings compromises. The 25th rank out of 1,636 is a strong result, but the wide gap between 1080p Ultra (159.1 FPS) and 4K Ultra (77.2 FPS) highlights that even this top-tier GPU cannot brute-force maximum settings at 4K in this title. The CPU, while not the star, provides enough single-thread performance to avoid becoming a limiting factor, as evidenced by the tight clustering of its nearest rivals in synthetic tests.