Control

Control, Remedy’s 2019 action-adventure, remains a demanding showcase for both GPU and CPU scaling. This analysis pairs the Intel Core Ultra 5 235HX with the NVIDIA GeForce RTX 5080, using measured frame-rate data across three resolutions and four quality presets. The dataset provides a clear picture of where this mobile-platform combination excels and where its bottlenecks emerge.

How This Combo Ranks

The combination of the Intel Core Ultra 5 235HX and NVIDIA GeForce RTX 5080 places 730th out of 2888 tested combos for Control, putting it in the top 25.3% of all configurations. This rank reflects a strong overall pairing, but the percentile spread between the CPU and GPU components suggests a slight imbalance. The CPU sits at the 91st percentile among all processors, while the GPU ranks at the 87th percentile among all graphics cards. The combo’s mid-pack position within the top quartile indicates that neither component is a weak link, but the system is not at the absolute frontier of performance for this title.

Looking at the CPU’s nearest rivals, the Ultra 5 235HX’s average benchmark score of 52073 is virtually tied with the AMD EPYC 8124P (52121, -0.1%) and the Intel Core i7-14700 (52301, -0.4%). It edges out the AMD Ryzen 9 5950X (51947, +0.2%) and the Intel Core i9-13900F (51730, +0.7%). These deltas are all within a single percentage point, meaning the processor’s raw compute capability is squarely in a well-populated performance tier. The GPU’s nearest rivals show a wider spread: the RTX 5080’s average score of 56083 is 0.6% ahead of the AMD Radeon 8060S (55757), 0.7% ahead of the AMD Radeon RX 6750 GRE 12 GB (55698), and 2.3% ahead of the AMD Radeon Pro W5700X (54828). However, it trails the AMD Radeon RX 9070 GRE (57367) by 2.2%. This GPU performance class is competitive, but Control’s specific engine workload will determine whether the combo’s rank is driven more by the graphics card or the processor.

GPU Role

The NVIDIA GeForce RTX 5080 is built on the Blackwell 2.0 architecture with a 5 nm process and a 378 mm² die containing 45,600 million transistors. Its 16 GB of GDDR7 memory on a 256-bit bus delivers 960.0 GB/s of bandwidth, which is a decisive factor for Control at high resolutions. The GPU’s base clock of 2295 MHz and boost clock of 2617 MHz provide substantial headroom for sustained rendering workloads. With 10752 shading units, 336 texture mapping units, and 112 ROPs, the card is well-equipped to handle the game’s physics-heavy environments and particle effects.

In Control, the GPU’s role is dominant at higher settings and resolutions. The measured data shows that at 1080p Low, the combo produces 190 FPS, but at 1080p Ultra, it drops to 104 FPS. This 86 FPS delta (a 45% reduction) is entirely attributable to the GPU’s increased workload from higher-quality textures, lighting, and post-processing effects. The RTX 5080’s 84 RT cores are relevant here, as Control features ray-traced reflections and indirect lighting that can be enabled at higher presets. The 112 ROPs handle the pixel fill rate, and the GPU’s pixel rate of 293.1 GPixel/s ensures that even at 4K Ultra, the card can push 53 FPS average, which is playable for a slower-paced action title.

The GPU’s benchmark scores reinforce its position. Its 3DMark Steel Nomad DX12 score of 8637 and Passmark G3D score of 36565 place it in the upper echelon of graphics hardware. However, the Passmark DirectX 12 score of 151 is notably lower than its DirectX 11 score of 324, which may indicate driver-level variance or workload-specific optimization. For Control, which uses DX11 and DX12 render paths, the card’s raw compute (56.28 TFLOPS FP32) and memory bandwidth are the primary contributors to the frame rates observed.

CPU Role

The Intel Core Ultra 5 235HX is a 14-core, 14-thread processor based on the Arrow Lake-HX architecture, fabricated on a 3 nm process. Its base clock of 2.90 GHz and boost clock of 5.10 GHz provide strong single-thread performance, which is critical for Control’s game logic and physics simulation. The CPU’s 24 MB of shared L3 cache and 102.4 GB/s of dual-channel DDR5 memory bandwidth ensure that data flows efficiently to the GPU. The processor’s 20 PCIe Gen 5 lanes (CPU only) provide a high-bandwidth connection to the RTX 5080, minimizing data transfer bottlenecks.

The CPU’s role in Control becomes apparent when examining the FPS scaling at lower settings. At 1080p Low, the combo achieves 190 FPS, which is a high frame rate that still leaves headroom for the GPU to be the limiting factor. However, the CPU’s Cinebench R23 multi-core score of 34731 and single-core score of 4903 indicate that it can keep up with the GPU’s output at these settings. The Passmark single-thread score of 4683 is strong, and the CPU’s percentile rank of 91st among all processors reflects its capability in both lightly-threaded and multi-threaded workloads.

In Control, the CPU’s 14 threads are sufficient to handle the game’s background AI, physics, and streaming systems without becoming a bottleneck at lower resolutions. The delta between 1080p Low (190 FPS) and 1080p High (120 FPS) is 70 FPS, which is GPU-driven as settings increase. But the CPU’s headroom is evident in the fact that 1080p Low does not exceed 200 FPS, suggesting a platform-level limit—likely the CPU’s single-thread ceiling or memory latency—that caps the absolute frame rate. The CPU’s nearest rival deltas (all within 0.7%) show that it is a solid mid-range performer, but it is not a top-tier processor like the Core i9-13900F, which it only beats by 0.7% in average benchmark score.

Measured FPS Breakdown

The measured frame rates reveal a consistent pattern across all resolutions: performance scales predictably with settings, and the GPU is the dominant constraint at higher quality presets.

At 1920x1080, the combo delivers:

• Low: 190 FPS average
• Medium: 135 FPS average (with a minimum of 115 FPS and maximum of 155 FPS)
• High: 120 FPS average
• Ultra: 104 FPS average

The gap between Low and Ultra is 86 FPS, indicating that the GPU is heavily taxed by the Ultra preset’s additional effects. The Medium preset’s 20 FPS range (115-155) shows some frame-time variance, but it remains smooth.

At 2560x1440, the results are:

• Low: 144 FPS average
• Medium: 102 FPS average (minimum 87 FPS, maximum 117 FPS)
• High: 91 FPS average
• Ultra: 79 FPS average

The drop from 1080p Low (190 FPS) to 1440p Low (144 FPS) is 46 FPS (a 24% reduction), which is purely resolution-driven. The Ultra preset at 1440p (79 FPS) is still well within playable territory, with no minimum FPS data points suggesting stutter-free operation.

At 3840x2160, the data shows:

• Low: 97 FPS average
• Medium: 69 FPS average (minimum 59 FPS, maximum 79 FPS)
• High: 61 FPS average
• Ultra: 53 FPS average

The 4K Ultra result of 53 FPS is the lowest measured value, but it remains above 50 FPS, which is acceptable for a game like Control. The Medium preset’s minimum of 59 FPS indicates that the 16 GB VRAM buffer is not being exceeded, and the 960.0 GB/s bandwidth is sufficient for 4K textures.

FAQ

Q: Is the RTX 5080 the bottleneck at 1080p Low?

A: No. The CPU’s single-thread performance (Passmark single-thread score of 4683) and the platform’s memory bandwidth (102.4 GB/s) are likely limiting the 190 FPS result, as the GPU’s raw compute (56.28 TFLOPS) is underutilized at such a low resolution and preset.

Q: How does the combo perform at 4K Ultra compared to 1080p Ultra?

A: The 4K Ultra average FPS of 53 is exactly half of the 1080p Ultra result of 104 FPS. This linear scaling indicates that the GPU is the sole limiting factor at these settings, as the resolution increase quadruples the pixel count but the frame rate halves due to memory bandwidth and shading unit limits.

Q: Which rival GPU is closest in performance to the RTX 5080?

A: The AMD Radeon RX 9070 GRE (avg score 57367) is 2.2% faster, while the AMD Radeon 8060S (55757) is 0.6% slower. The RTX 5080’s average benchmark score of 56083 places it between these two, with the AMD Radeon RX 6750 GRE 12 GB (55698) being 0.7% slower.

Q: Does the CPU’s 14-thread count limit Control’s performance?

A: No. Control’s engine is not heavily multi-threaded, and the CPU’s Cinebench R23 multi-core score of 34731 is more than sufficient. The 14 threads handle background tasks without issue, as evidenced by the 1440p Low result of 144 FPS, which is still GPU-bound.

Q: What is the best resolution for a 144Hz display?

A: At 1080p, the Low preset (190 FPS) and Medium preset (135 FPS) can exceed or approach 144Hz, but the High preset (120 FPS) falls short. At 1440p, only the Low preset (144 FPS) exactly matches a 144Hz refresh rate.

Q: Is the 16 GB VRAM sufficient for Control at 4K?

A: Yes. The 4K Medium preset’s minimum FPS of 59 and maximum of 79 show no VRAM-related stutter, and the 16 GB capacity with 960.0 GB/s bandwidth is ample for the game’s highest texture quality at 3840x2160.

Settings Recommendations

Based on the measured FPS rows, the High preset offers the best balance of visual fidelity and performance across all resolutions. At 1080p, High delivers 120 FPS, which is smooth for most displays and a 16 FPS improvement over Ultra (104 FPS) while retaining most of the visual quality. At 1440p, High hits 91 FPS, which is comfortably above 60 FPS and preserves the game’s atmospheric lighting and particle effects. At 4K, High produces 61 FPS, which is right at the threshold for a consistent 60Hz experience. The Ultra preset’s gains are marginal in Control’s engine, but the 8 FPS difference between High and Ultra at 4K (61 vs 53 FPS) is not worth the performance hit.

For users with a 144Hz monitor, the Medium preset at 1080p (135 FPS) is the optimal choice, as it gets closer to the refresh rate than High (120 FPS) without the frame-time variance seen at Low. At 1440p, Medium (102 FPS) is the sweet spot for high-refresh gaming, as it stays above 100 FPS. The Low preset is only recommended for competitive play at 1080p (190 FPS), but it sacrifices too much visual quality for the modest gain over Medium.

Resolution Scaling

The FPS drop from 1080p to 4K is consistent and reveals the GPU as the primary scaling constraint. At Low settings, the frame rate falls from 190 FPS at 1080p to 144 FPS at 1440p (a 24% reduction) and then to 97 FPS at 4K (a 49% total reduction from 1080p). This is a near-linear relationship with pixel count: 4K has four times the pixels of 1080p, and the frame rate drops by roughly half, indicating that the GPU’s fill rate and memory bandwidth are the limiting factors.

At Ultra settings, the scaling is even steeper. The frame rate drops from 104 FPS at 1080p to 79 FPS at 1440p (a 24% reduction) and then to 53 FPS at 4K (a 49% total reduction). This consistent percentage drop across all presets confirms that the CPU is not a bottleneck at any resolution. The Ultra 4K result of 53 FPS is exactly 51% of the 1080p Ultra result, which is a textbook example of GPU-limited scaling. The RTX 5080’s 960.0 GB/s memory bandwidth is the primary constraint at 4K, as the 256-bit bus must move significantly more texture data per frame. The CPU’s role remains static, with its 5.10 GHz boost clock and 24 MB L3 cache providing consistent game logic processing regardless of resolution. This analysis indicates that users seeking higher 4K frame rates would need to lower settings to High (61 FPS) or Medium (69 FPS), as the GPU is the definitive performance ceiling for this combo in Control.