Control

Settings Recommendations

The measured FPS rows for Control with this Intel Core Ultra 7 265K and NVIDIA GeForce RTX 5090 D combination reveal a clear hierarchy across the four settings presets. At 1920x1080, the Low preset delivers an average of 298 FPS, while Medium drops to 211 FPS, High to 188 FPS, and Ultra to 163 FPS. The step from Low to Medium represents a 87 FPS reduction, which is substantial, but the jump from Medium to High is only 23 FPS, and from High to Ultra another 25 FPS. This suggests that the visual gains from High to Ultra come at a relatively modest cost in frame rate compared to the massive leap from Low to Medium.

For the best experience, the data points toward the High preset as the optimal balance. At 2560x1440, High achieves 142 FPS, which remains well above the 60 FPS threshold for smooth gameplay, while Ultra still delivers 124 FPS. At 3840x2160, High produces 96 FPS, and Ultra drops to 83 FPS. The High preset consistently maintains a comfortable margin over the 60 FPS baseline at every resolution, whereas Ultra pushes closer to the edge at 4K. Given that Control is an Action Adventure title, the higher fidelity of Ultra might be tempting, but the measured numbers indicate that High preserves more headroom.

The Medium preset at 4K (108 FPS) actually outperforms High at 4K (96 FPS), which is counterintuitive, but the data shows this anomaly clearly. However, Medium also exhibits the only min FPS readings in the entire dataset: 92 FPS at 4K, 136 FPS at 1440p, and 180 FPS at 1080p. These min values suggest that Medium has more frame-time variance than the other presets, which could manifest as stutter in fast-paced scenes. The max FPS readings for Medium are 124 at 4K, 184 at 1440p, and 243 at 1080p, indicating a wide spread between minimum and maximum.

For players prioritizing visual fidelity with high refresh rates, the High preset at 1440p (142 FPS) seems ideal for 144Hz displays. For 60Hz 4K displays, High at 96 FPS provides sufficient overhead. The Low preset, while delivering the highest frame rates (298 FPS at 1080p, 225 FPS at 1440p, 152 FPS at 4K), likely sacrifices too much visual quality for a game like Control, which is known for its atmospheric environments. The recommendation is to start with High, then adjust downward only if the min FPS readings become problematic.

FAQ

Q: What is the highest average FPS recorded for this combination?

A: The highest average FPS is 298, achieved at 1920x1080 with the Low settings preset.

Q: How does the combo rank among other tested configurations?

A: This combination ranks 1st out of 2888 tested combos for Control, placing it at the very top of the database.

Q: What is the only settings preset that includes minimum and maximum FPS measurements?

A: The Medium preset is the only one with recorded min and max FPS values: 180/243 at 1080p, 136/184 at 1440p, and 92/124 at 4K.

Q: Does the Ultra preset ever drop below 80 FPS at any resolution?

A: At 3840x2160 Ultra, the average FPS is 83, which is above 80, but there is no min FPS recorded for Ultra to verify the lowest dip.

Q: What is the frame rate difference between 1080p and 4K at High settings?

A: At High settings, the average FPS drops from 188 at 1080p to 96 at 4K, a reduction of 92 FPS.

Q: Is the performance at 1440p High sufficient for high-refresh-rate gaming?

A: Yes, the 142 FPS average at 2560x1440 High exceeds the 120 FPS threshold commonly used for high-refresh displays.

How This Combo Ranks

The data places this Intel Core Ultra 7 265K with the NVIDIA GeForce RTX 5090 D at rank 1 out of 2888 tested combos for Control. This is the top position, meaning no other configuration in the database achieves higher measured performance in this game. The percentile ranking of the individual components reinforces this: the CPU sits at the 94th percentile versus all CPUs, and the GPU sits at the 92nd percentile versus all GPUs. While neither component is strictly the best ever recorded, their combination produces the highest observed frame rates in Control.

The CPU's nearest rivals in general benchmarks include the Intel Xeon 6511P (0.2% lower average score), the Intel Xeon Platinum 8270 (0.7% lower), the Intel Core i7-14700KF (1% higher), and the AMD Ryzen 7 9700F (1.3% higher). The Intel Core i7-14700KF and AMD Ryzen 7 9700F actually post slightly higher average benchmark scores, yet the Core Ultra 7 265K still leads in gaming performance for Control. This suggests that the game benefits from specific architectural traits of the Arrow Lake design rather than raw multi-threaded throughput.

The GPU's nearest rivals in general benchmarks include the AMD Radeon RX 6650M XT (1.1% higher average score), the AMD Radeon RX 6850M XT (1.6% lower), the NVIDIA Tesla P100 PCIe 12 GB (2.1% lower), and the NVIDIA Tesla P100 PCIe 16 GB (2.4% lower). The RTX 5090 D's average benchmark score of 77712 is competitive with these mobile and data-center parts, but in Control it achieves the top combo rank. The data implies that the RTX 5090 D's architecture excels in this specific game engine, regardless of how it fares in synthetic benchmarks.

The gap between the combo rank and the individual component percentiles is noteworthy. A rank of 1 out of 2888 means this pairing outperforms 2887 other combinations, many of which may feature more expensive or more powerful components on paper. This could indicate that Control is particularly well-optimized for the Blackwell 2.0 architecture, or that the CPU-GPU balance here avoids bottlenecks that plague other high-end pairings.

Measured FPS Breakdown

At 1920x1080, the full range of settings produces the following average FPS: Low at 298, Medium at 211, High at 188, and Ultra at 163. The Medium preset's min and max readings are 180 and 243, respectively, showing a 63 FPS spread. This resolution demonstrates the greatest absolute frame rates, with Low approaching the 300 FPS mark and Ultra still maintaining a solid 163 FPS.

Moving to 2560x1440, the averages shift to: Low at 225, Medium at 160, High at 142, and Ultra at 124. The Medium preset's min and max readings are 136 and 184, a 48 FPS spread. At this resolution, the gap between Low and Medium is 65 FPS, while the gap between High and Ultra narrows to 18 FPS. The scaling from 1080p to 1440p reduces frame rates by roughly 24-27% across settings, with Low dropping from 298 to 225, Medium from 211 to 160, High from 188 to 142, and Ultra from 163 to 124.

At 3840x2160, the averages are: Low at 152, Medium at 108, High at 96, and Ultra at 83. The Medium preset's min and max readings are 92 and 124, a 32 FPS spread. The reduction from 1440p to 4K is more severe: Low drops by 73 FPS, Medium by 52 FPS, High by 46 FPS, and Ultra by 41 FPS. The diminishing absolute differences as resolution increases suggest that the GPU is becoming the primary constraint at higher pixel counts.

The Ultra preset at 4K, with an average of 83 FPS, is the only configuration that approaches the 60 FPS minimum for comfortable gaming. All other settings and resolutions maintain averages above 96 FPS. The Low preset at 4K still achieves 152 FPS, which is remarkable for a 4K resolution and indicates the sheer headroom this combination possesses.

Resolution Scaling

The data reveals a consistent pattern of frame rate degradation as resolution increases from 1080p to 1440p to 4K. For the High preset, the progression is 188 FPS at 1080p, 142 FPS at 1440p, and 96 FPS at 4K. This represents a 24.5% drop from 1080p to 1440p, and a 32.4% drop from 1440p to 4K. The Low preset shows 298 to 225 to 152 FPS, with drops of 24.5% and 32.4% respectively. The Medium preset goes from 211 to 160 to 108 FPS, with drops of 24.2% and 32.5%. The Ultra preset goes from 163 to 124 to 83 FPS, with drops of 23.9% and 33.1%.

The percentage drops are remarkably consistent across all settings presets, hovering around 24% for the first step and 32-33% for the second step. This consistency suggests that the scaling is driven by the GPU's pixel throughput capabilities rather than by any settings-specific bottleneck. The increase in pixel count from 1080p to 1440p is 1.78x, and from 1440p to 4K is 2.25x, yet the frame rates do not drop proportionally. This indicates that the RTX 5090 D's 1.79 TB/s memory bandwidth and 104.8 TFLOPS of FP32 performance are sufficient to handle the increased load with better-than-linear scaling.

The fact that the FPS drops are smaller than the pixel count increases implies that the limiting factor at lower resolutions is not the GPU but rather the CPU or game engine's draw call processing. At 1080p, the CPU must feed frames to the GPU at rates up to 298 FPS, which likely approaches the engine's maximum frame rate cap or the CPU's single-threaded limit. The Intel Core Ultra 7 265K's boost clock of 5.50 GHz and single-core benchmark scores (Cinebench R23 single-core score of 2020, Geekbench single-core score of 2713) suggest it is capable, but the near-linear scaling from 1080p to 4K indicates that the GPU becomes the dominant constraint only at higher resolutions.

CPU Role

The Intel Core Ultra 7 265K features 20 cores and 20 threads, with no hyperthreading, based on the Arrow Lake architecture. Its base clock is 3.90 GHz with a boost clock of 5.50 GHz. The 20 cores are split between performance and efficient cores, though the fact pack does not detail the exact configuration. The L3 cache is 30 MB shared, with 3 MB L2 per core and 192 KB L1 per core. This cache hierarchy is designed to reduce memory latency, which is critical for game engines that frequently access the same data structures.

In Control, the CPU's role is evident from the 1080p results. At 1080p Low, the 298 FPS average likely approaches the engine's CPU-bound limit. The Cinebench R23 multi-core score of 35850 and single-core score of 2020 indicate strong performance, but the game's frame rates at lower resolutions suggest that the CPU is still the bottleneck before the GPU takes over. The 20-thread configuration might not be fully utilized by a game from 2019, which likely scales well to 8-16 threads but may not benefit from all 20.

The CPU's benchmark rivals provide context: the Intel Core i7-14700KF scores 1% higher on average, and the AMD Ryzen 7 9700F scores 1.3% higher, yet the Core Ultra 7 265K achieves the top combo rank. This implies that Control does not heavily rely on raw multi-threaded performance but rather on single-threaded responsiveness and memory subsystem efficiency. The 102.4 GB/s memory bandwidth and DDR5 support likely contribute to faster data feeding compared to DDR4-equipped rivals.

The CPU's integrated graphics (Arc Xe-LPG Graphics 64EU) are irrelevant for this discrete GPU configuration, but the 20 PCIe Gen 5 lanes from the CPU ensure the RTX 5090 D's PCIe 5.0 x16 interface receives full bandwidth. The 125W TDP and 3nm process node from TSMC indicate an efficient chip, though power draw is not directly measured in the game results. The CPU's percentile of 94 versus all CPUs places it in the top 6% of all processors, which is consistent with its performance class.

GPU Role

The NVIDIA GeForce RTX 5090 D is built on the Blackwell 2.0 architecture with the GB202 chip, manufactured on a 5nm process by TSMC. It features 32 GB of GDDR7 memory on a 512-bit bus, providing 1.79 TB/s of bandwidth. The base clock is 2017 MHz with a boost clock of 2407 MHz, and memory runs at 1750 MHz (28 Gbps effective). The GPU has 21760 shading units, 680 TMUs, and 176 ROPs, along with 170 RT cores and 680 tensor cores. Its FP32 performance is 104.8 TFLOPS, and the pixel rate is 423.6 GPixel/s with a texture rate of 1,636.8 GTexel/s.

In Control, the GPU's dominance becomes apparent at 4K resolutions. At 3840x2160, the frame rates range from 83 (Ultra) to 152 (Low), which are high for 4K but still show the GPU working harder. The 32 GB VRAM is far more than Control requires, as the game's textures are not likely to exceed 8-12 GB even at Ultra settings. This VRAM headroom means there is no memory swapping or texture thrashing, which could explain the smooth frame times at higher resolutions.

The GPU's benchmark rivals include mobile and data-center parts: the AMD Radeon RX 6650M XT scores 1.1% higher, while the AMD Radeon RX 6850M XT scores 1.6% lower. The RTX 5090 D's 3DMark Steel Nomad DX12 score of 14326 and Geekbench Vulkan score of 376915 are strong, but its PassMark G3D score of 44065 is notably lower than its rivals' average scores. This discrepancy suggests that the RTX 5090 D's architecture may excel in specific workloads like real-time ray tracing or modern API features, which Control uses, even if older DirectX benchmarks do not fully capture its capabilities.

The GPU's 92nd percentile versus all GPUs indicates it is in the top 8% of all graphics cards. The 575W TDP and suggested 950W PSU highlight its power requirements, but the measured FPS results justify this power draw. The 1.79 TB/s memory bandwidth is likely the key factor in the consistent resolution scaling, as it provides ample bandwidth for 4K texture streaming. The RTX 5090 D's 170 RT cores are particularly relevant for Control, which is known for its ray-traced reflections and lighting effects, though the measured data does not specify whether ray tracing was enabled in these tests.