Control

The AMD Ryzen 9 9900X3D and AMD Radeon RX 5700 XT combination presents a fascinating study in bottleneck dynamics within Control, a game that stresses both raw compute and memory bandwidth. The data shows a pairing of a top-tier modern processor with a previous-generation graphics card, resulting in performance that is heavily dictated by the GPU at higher resolutions, yet surprisingly robust at 1080p.

CPU Role — cores, clocks, and how they relate to this game's results

The Ryzen 9 9900X3D is a 12-core, 24-thread processor based on the Zen 5 architecture, built on a 4 nm process. Its base clock is 4.40 GHz, boosting up to 5.50 GHz, which provides substantial headroom for game logic and physics. The processor’s 128 MB of L3 cache is a significant asset, likely aiding in scenarios where the game engine needs rapid access to frequently used data. In synthetic benchmarks, the CPU demonstrates its strength with a Cinebench R23 multicore score of 47737 and a single-core score of 6739, alongside a Geekbench multicore score of 22884 and single-core score of 3041.

Given these impressive compute figures, it is unlikely that the CPU becomes the primary limiting factor in this specific configuration. In Control, which relies on demanding physics and AI routines, the 12 cores and 24 threads should provide ample processing power to feed frames to the GPU. The data suggests that at lower resolutions, such as 1920x1080, the CPU’s performance is sufficient to allow the GPU to reach its maximum output, as evidenced by the significant FPS jump from Low to High settings. The processor's percentile ranking of 91 among all CPUs further underscores its high-end positioning; however, the benchmark results indicate that its performance is closely matched by rivals like the Intel Core Ultra 5 245KF, which scores only 0.6% lower on average. This indicates that while the 9900X3D is powerful, it is not dramatically ahead of its immediate competitors in general processing tasks, making the GPU the more decisive component in this pairing.

GPU Role — VRAM, clocks, and how they relate to this game's results

The AMD Radeon RX 5700 XT is the clear performance bottleneck in most scenarios tested. This GPU, based on the RDNA 1.0 architecture, features 8 GB of GDDR6 memory on a 256-bit bus, providing 448.0 GB/s of bandwidth. Its boost clock is 1905 MHz, and it delivers 9.754 TFLOPS of FP32 compute. The card’s 64 ROPs are responsible for final pixel output, which becomes a critical factor at higher resolutions.

The measured FPS data clearly illustrates this limitation. At 3840x2160, the GPU is heavily strained, with average FPS dropping to 37 on High settings and 32 on Ultra. This suggests that the 8 GB VRAM and the overall pixel throughput of the RX 5700 XT are insufficient to maintain high frame rates at 4K. Even at 1440p, Ultra settings only yield an average of 48 FPS, indicating that the GPU is still the primary constraint. The card’s benchmark percentile of 59 places it in the mid-range, and its average benchmark score of 16361 is virtually identical to rivals like the AMD Radeon Pro 5600M (0.1% difference) and the NVIDIA GeForce RTX 5090 D V2 (-0.9% difference). This comparison underscores that while the CPU is a modern flagship, the GPU is a mid-range part from a previous generation, creating a significant performance imbalance that defines the overall experience.

Resolution Scaling — how FPS drops from 1080p to 4K and what it says about the limiting component

The FPS scaling across resolutions provides a textbook example of GPU-bound behavior. Starting at 1920x1080 with Low settings, the combo achieves an average of 115 FPS. Moving to 2560x1440 at Low settings, the average drops to 87 FPS, a 24% reduction. At 3840x2160 at Low settings, the average plummets to 59 FPS, a further 32% drop from 1440p. This steep decline as resolution increases is a clear indicator that the GPU’s rendering capabilities are being saturated.

The pattern holds across all settings presets. For High settings, the FPS goes from 73 at 1080p to 55 at 1440p (a 25% drop) and then to 37 at 4K (a 33% drop from 1440p). This consistent scaling behavior points directly to the RX 5700 XT as the limiting component. If the CPU were the bottleneck, we would expect to see less significant FPS changes when altering resolution, as the CPU’s workload would remain relatively constant. The data shows that the CPU is capable of delivering high frame rates at lower resolutions, but the GPU simply cannot sustain them as pixel count increases. The gap between Low and Ultra settings also widens at 4K (59 FPS vs 32 FPS) compared to 1080p (115 FPS vs 63 FPS), further confirming that the GPU’s fill rate and memory bandwidth are exhausted at higher resolutions.

How This Combo Ranks — use comboRankInGame vs other tested combos

This specific combination of the Ryzen 9 9900X3D and RX 5700 XT ranks 2767 out of 3720 tested combos for Control. This places it in the 74th percentile of all tested configurations, meaning approximately 74% of all CPU and GPU pairings tested by the database perform better in this game. This ranking is surprisingly low for a system with such a powerful CPU, and it directly reflects the performance drag of the older GPU.

The high-end processor is wasted in this pairing from a gaming perspective, as the GPU prevents the system from achieving the high frame rates that the CPU could otherwise support. This is not a reflection of the CPU’s capability, but rather evidence of how heavily Control relies on GPU performance. The rank suggests that users would see a substantial improvement by pairing the 9900X3D with a more modern and powerful graphics card, as the CPU has ample headroom to support it. The current ranking is a direct consequence of the GPU’s mid-range status, and it highlights the importance of balanced component selection for gaming workloads.

FAQ

Q: Is the AMD Ryzen 9 9900X3D or the AMD Radeon RX 5700 XT the main bottleneck in this system for Control?

A: The benchmark data clearly indicates that the AMD Radeon RX 5700 XT is the primary bottleneck. This is evident from the FPS scaling, where performance drops significantly more when increasing resolution than when increasing CPU load. The CPU’s high benchmark scores and performance percentile suggest it is not the limiting factor.

Q: What is the best resolution to play Control with this hardware configuration?

A: For a consistently smooth experience, 1920x1080 is the most appropriate resolution. At this resolution, the system achieves an average of 63 FPS on Ultra settings and 73 FPS on High, which are playable frame rates. At 2560x1440, performance becomes more variable, with Ultra dipping to an average of 48 FPS.

Q: How much of a performance drop should one expect when moving from 1080p High settings to 4K High settings?

A: The average frame rate drops from 73 FPS at 1920x1080 High to 37 FPS at 3840x2160 High. This represents a 49% reduction in average FPS, illustrating the severe strain that 4K resolution places on the RX 5700 XT.

Q: Is the Ryzen 9 9900X3D's performance significantly better than its close rivals in general benchmarks?

A: The data shows that the Ryzen 9 9900X3D is only marginally ahead of its nearest rivals. For instance, it is just 0.6% faster than the Intel Core Ultra 5 245KF and only 1.3% faster than the Intel Core Ultra 5 245K in average benchmark scores, indicating a very tight competitive field at this performance level.

Q: Can this system achieve 60 FPS on Ultra settings at any resolution?

A: No. The measured FPS data shows that the maximum average FPS on Ultra settings is 63 at 1920x1080. At 2560x1440, the Ultra setting average drops to 48 FPS, and at 3840x2160, it falls to 32 FPS. The system cannot maintain a 60 FPS average on Ultra at any resolution above 1080p.

Q: At which settings and resolution does the RX 5700 XT's 8 GB VRAM appear to be a limiting factor?

A: While not explicitly measured, the significant performance drop at 3840x2160 suggests the GPU’s resources, including its 8 GB VRAM and 448.0 GB/s bandwidth, are overwhelmed. The average FPS at 4K is below 60 on all settings, including Low, which indicates a fundamental hardware limitation at that resolution.

Settings Recommendations

Based on the measured FPS data, the optimal experience balances visual fidelity with a stable frame rate. For a smooth and responsive experience, the High preset at 1920x1080 appears to be the sweet spot, delivering an average of 73 FPS with no reported minimum FPS data indicating stability. This provides a significant visual upgrade over Low settings while maintaining a frame rate well above 60 FPS.

For users with a high refresh rate monitor, the Low preset at 1920x1080 offers the highest performance, averaging 115 FPS. This is the only configuration that comes close to maximizing a 120Hz display. At 2560x1440, the Medium preset provides a reasonable compromise at 62 FPS average, but it dips to a minimum of 53 FPS, which may be noticeable in demanding scenes. Playing at 3840x2160 is not recommended for a smooth experience, as even the Low preset only averages 59 FPS, and visual quality settings have a large impact on performance, with High dropping to 37 FPS. The data suggests that 1080p High is the most balanced choice for this hardware.

Measured FPS Breakdown

The following data represents the exact measured average FPS for each resolution and settings preset combination.

1920x1080:

• Low: 115 FPS average
• Medium: 82 FPS average (70 min, 94 max)
• High: 73 FPS average
• Ultra: 63 FPS average

2560x1440:

• Low: 87 FPS average
• Medium: 62 FPS average (53 min, 71 max)
• High: 55 FPS average
• Ultra: 48 FPS average

3840x2160:

• Low: 59 FPS average
• Medium: 42 FPS average (36 min, 48 max)
• High: 37 FPS average
• Ultra: 32 FPS average

These figures illustrate a clear and consistent performance hierarchy. The system is most capable at 1080p, where it can handle even the most demanding presets. As resolution increases, the average FPS drops substantially, and the gap between preset levels narrows in relative terms, confirming the GPU as the definitive performance limiter.