Control

The Intel Core i9-14900K paired with the Intel Arc B370 produces a striking performance profile in Control, one defined by severe GPU limitation. The data shows a combo that ranks 2657th out of 2888 tested combinations in this game, placing it in the bottom percentile of all recorded systems. This pairing leverages a top-tier 24-core, 32-thread processor with a 95th percentile standing among all CPUs, yet the graphics solution holds the entire system back to 5th percentile territory among all GPUs. The result is a benchmark set where frame rates hover near playability thresholds at best, and the resolution scaling pattern reveals a component bottleneck so pronounced that even the most powerful processor available cannot compensate for the graphics subsystem's limitations.

Resolution Scaling

The measured FPS data for this combo in Control displays a textbook case of GPU-bound scaling across the three tested resolutions. At 1920x1080 with Low settings, the system achieves an average of 59 FPS, which drops to 45 FPS at 2560x1440 and further to 30 FPS at 3840x2160. This represents a 24% reduction from 1080p to 1440p, and then a 33% reduction from 1440p to 4K. The total drop from 1080p Low to 4K Low is 49%, meaning the system loses nearly half its frame rate when rendering four times the pixels.

Looking at High settings, the pattern intensifies. The 1080p High average of 37 FPS falls to 28 FPS at 1440p (a 24% drop) and then to 19 FPS at 4K (a 32% drop from 1440p). Again, the total reduction from 1080p to 4K is 49%, nearly identical to the Low preset scaling. This consistency across presets is telling. When a system scales almost identically regardless of quality settings, it indicates the bottleneck is not related to shader complexity, texture filtering, or other settings-dependent workloads. Instead, the limiting factor is raw pixel throughput and fill rate.

The Ultra preset tells the same story. At 1080p Ultra, the average is 32 FPS. This drops to 25 FPS at 1440p and 17 FPS at 4K. The percentage drops are 22% and 32% respectively, with a total 47% reduction from 1080p to 4K. The Medium preset, which has min/max data available, shows 42 FPS average at 1080p (with a range of 36 to 48), 32 FPS at 1440p (range 27 to 37), and 21 FPS at 4K (range 18 to 25). The Medium preset's drop from 1080p to 4K is exactly 50%.

What this scaling pattern reveals is that the Intel Arc B370's 20 ROPs and 48.00 GPixel/s pixel rate are the fundamental constraining resources. Moving from 1080p to 4K quadruples the pixel count, and the FPS drops by roughly half in every preset configuration. This is not a linear relationship, but it is close to what one would expect from a GPU that is completely saturated at every resolution. The CPU, which boasts a Cinebench R23 multi-core score of 39399.5 and a single-core score of 2262.5, is never the limiting factor here. Even at 1080p Low, where the CPU would have the most headroom to flex, the system only reaches 59 FPS average, suggesting the GPU is already at its ceiling.

The fact that the 1080p Low preset only achieves 59 FPS — just one frame shy of 60 — while the same resolution at Ultra delivers 32 FPS, shows that settings do matter for absolute performance, but they do not change the fundamental bottleneck. The spread between Low and Ultra at 1080p is 27 FPS (59 vs 32), while at 4K the spread is 13 FPS (30 vs 17). This narrowing spread at higher resolutions further confirms GPU-bound behavior; as pixel count rises, the relative cost of quality settings diminishes because the pixel fill rate dominates the workload.

GPU Role

The Intel Arc B370 is an integrated graphics solution based on the Xe3-LPG architecture, manufactured on a 3 nm process. Its clock speeds are listed as a 300 MHz base and 2400 MHz boost, with memory designated as "System Shared" and bandwidth described as "System Dependent." This means the GPU relies on the system's main memory rather than dedicated VRAM, and its performance characteristics are intrinsically tied to the memory subsystem of the host platform. The GPU's 1280 shading units, 40 TMUs, and 20 ROPs produce a fill rate of 48.00 GPixel/s and a texture rate of 96.00 GTexel/s. Compute throughput is rated at 6.144 TFLOPS for FP32 and 12.29 TFLOPS for FP16 (2:1).

In the context of Control, a game that features heavy particle effects, physics-based destruction, and complex lighting, these specifications translate to modest performance. The GPU's 3DMark Steel Nomad DX12 score of 1184 places it in the 5th percentile among all GPUs. Its nearest rivals in benchmark scores are the ATI Mobility Radeon HD 5570 (scoring 1186, a -0.2% delta), the ATI Radeon HD 5770 (scoring 1190, -0.5% delta), and the AMD Radeon HD 7650M (scoring 1192, -0.7% delta). These are all legacy discrete GPUs from over a decade ago, and the Arc B370 essentially matches their raw benchmark output.

The measured frame rates in Control align closely with this benchmark positioning. At 1080p Medium, the average of 42 FPS with a minimum of 36 and maximum of 48 suggests the GPU can handle the game at moderate settings but lacks the headroom for consistent high refresh rate play. The 4K Ultra result of 17 FPS average is essentially unplayable, while 4K Low at 30 FPS is marginally interactive at best. The GPU's 10 RT cores and DirectX 12 Ultimate support are present, but with only 6.144 TFLOPS of FP32 compute, ray tracing workloads in Control would likely push performance even lower than the already constrained rasterized results.

An important consideration is the GPU's TDP of 25 W and its IGP slot width with no power connectors. This is not a discrete graphics card but an integrated processor graphics solution, likely embedded in a mobile or compact platform. The "System Shared" memory configuration means the GPU is borrowing bandwidth from the CPU's DDR4 or DDR5 memory, which, while the i9-14900K supports both, cannot match the dedicated bandwidth of a discrete GPU's VRAM. The "System Dependent" bandwidth specification is a key limiting factor; in a system with fast DDR5 memory, the Arc B370 might perform slightly better, but the data shows it remains critically constrained regardless.

The contrast with the CPU is stark. The Intel Core i9-14900K has a 95th percentile ranking among all CPUs, with an average benchmark score of 79097. Its nearest rivals include the Intel Core i9-14900KF (scoring 79371, -0.3% delta), the Intel Core Ultra 9 290HX Plus (scoring 79574, -0.6% delta), and the AMD EPYC 7413 (scoring 80041, -1.2% delta). The i9-14900K's multi-threaded performance is exceptional, but in Control, none of that power translates to frame rate because the GPU is the sole bottleneck. The data shows that even at 1080p Low, the system cannot exceed 59 FPS, which would be a trivial load for this CPU if paired with a capable discrete GPU.

FAQ

Q: What is the best average frame rate this combo achieves in Control?

A: The highest average FPS recorded is 59 at 1920x1080 with Low settings. This is just one frame shy of 60 FPS and represents the only preset/resolution combination that approaches a playable 60 FPS target.

Q: Is this system capable of 4K gaming in Control?

A: The data shows 4K averages of 30 FPS on Low, 21 FPS on Medium, 19 FPS on High, and 17 FPS on Ultra. None of these reach the 30 FPS threshold for acceptable console-style play, and the Ultra setting at 17 FPS is effectively a slideshow. 4K is not viable for a smooth experience with this combo.

Q: How does the Intel Arc B370 compare to its nearest GPU rivals in synthetic benchmarks?

A: The Arc B370 scores 1184 in 3DMark Steel Nomad DX12. Its nearest rival, the ATI Mobility Radeon HD 5570, scores 1186 (a -0.2% delta), while the ATI Radeon HD 5770 scores 1190 (-0.5% delta) and the AMD Radeon HD 7650M scores 1192 (-0.7% delta). The AMD FirePro M2000 scores 1168 (a 1.4% delta below the Arc B370).

Q: What is the combo's ranking among all tested systems in Control?

A: The combo ranks 2657th out of 2888 tested combinations. This places it in the bottom 8% of all systems tested for this game, reflecting the severe GPU limitation.

Q: Does increasing resolution affect all settings presets equally?

A: Yes. From 1080p to 4K, the Low preset drops from 59 to 30 FPS (49% reduction), Medium drops from 42 to 21 FPS (50% reduction), High drops from 37 to 19 FPS (49% reduction), and Ultra drops from 32 to 17 FPS (47% reduction). The consistency of these percentages confirms a pixel-bound bottleneck.

Q: What is the i9-14900K's percentile standing among all CPUs?

A: The i9-14900K is in the 95th percentile among all CPUs, with an average benchmark score of 79097. Its closest rival, the i9-14900KF, scores 79371 with a -0.3% delta, meaning the two are effectively tied in synthetic CPU benchmarks.

How This Combo Ranks

The combo's rank of 2657 out of 2888 in Control places it in the bottom 92% of all tested systems. This is a remarkably low position for a system featuring the Intel Core i9-14900K, which by itself ranks in the 95th percentile among all CPUs. The discrepancy between CPU and GPU percentile standings (95 vs 5) illustrates the extreme imbalance of this pairing. To contextualize the rank, consider that 231 combinations out of 2888 perform worse, meaning there are very few systems that this combo can beat in this game. The data indicates that any system with a discrete GPU from the past decade would likely outperform this pairing, given that the Arc B370's nearest rivals are the ATI Radeon HD 5770 and similar legacy parts.

The 2657th rank is driven entirely by the GPU's 5th percentile standing. The Arc B370's 3DMark Steel Nomad score of 1184 is indicative of a GPU that is not designed for modern gaming at playable frame rates. In Control, this translates to the measured FPS data: the best case scenario is 59 FPS at 1080p Low, and every other configuration falls below 45 FPS. The rank also reflects the fact that most other tested combos include discrete GPUs with dedicated VRAM, which is a critical advantage in a game like Control that demands significant texture bandwidth and fill rate.

Comparing to the CPU's rivals, the i9-14900K's nearest competitor, the i9-14900KF, has a -0.3% delta in average benchmark score, meaning they are functionally identical in processing power. The AMD EPYC 7413 scores 80041 with a -1.2% delta, while the AMD Ryzen AI Max+ 395 scores 77740 with a 1.7% delta. None of these CPU differences matter in Control for this combo, because the GPU is the definitive bottleneck. Even if the CPU were swapped for a lower-tier part, the measured FPS would remain nearly identical, as the GPU cannot be fed fast enough to become the limiting factor.

The combo rank of 2657 also has implications for the data's reliability. The "dataIsMeasured" field is true, indicating that these are actual measured results rather than estimates. This adds credibility to the finding that a 95th percentile CPU paired with a 5th percentile GPU produces a bottom-tier gaming experience. The rank is not an artifact of poor CPU performance; it is a direct reflection of the GPU's inability to render the game at acceptable frame rates.

Measured FPS Breakdown

The complete measured FPS dataset for this combo in Control is as follows, organized by resolution and settings:

1920x1080:

• Low: 59 FPS average
• Medium: 42 FPS average, with a minimum of 36 and maximum of 48
• High: 37 FPS average
• Ultra: 32 FPS average

2560x1440:

• Low: 45 FPS average
• Medium: 32 FPS average, with a minimum of 27 and maximum of 37
• High: 28 FPS average
• Ultra: 25 FPS average

3840x2160:

• Low: 30 FPS average
• Medium: 21 FPS average, with a minimum of 18 and maximum of 25
• High: 19 FPS average
• Ultra: 17 FPS average

Analyzing the 1080p results, the progression from Low to Ultra shows a steady decline: 59, 42, 37, and 32 FPS. The gap between Low and Medium is 17 FPS, while the gap between Medium and High is 5 FPS, and between High and Ultra is another 5 FPS. This suggests that the Low preset removes enough graphical features to significantly lighten the GPU load, but the subsequent presets have diminishing differences as the GPU becomes more saturated.

At 1440p, the same pattern holds: 45, 32, 28, and 25 FPS for Low through Ultra. The delta between Low and Medium is 13 FPS, between Medium and High is 4 FPS, and between High and Ultra is 3 FPS. The overall reduction from 1080p to 1440p is 14 FPS on Low, 10 FPS on Medium, 9 FPS on High, and 7 FPS on Ultra. This narrowing delta at higher settings again confirms that the GPU's pixel throughput is the primary constraint.

The 4K results are uniformly poor: 30, 21, 19, and 17 FPS for Low through Ultra. The Low preset just barely hits 30 FPS, which could be considered minimally playable for a slow-paced game, but the Medium preset drops to 21 FPS, and Ultra falls to 17 FPS. The min/max data for Medium at 4K shows a range of 18 to 25 FPS, indicating significant frame time variance even at this low performance level. At 1440p Medium, the range is 27 to 37 FPS, and at 1080p Medium, the range is 36 to 48 FPS. These ranges show that the frame time consistency improves as resolution decreases, which is expected given the GPU's limited fill rate.

The data also reveals that the difference between Low and Ultra at any given resolution is relatively small compared to the difference between resolutions. At 1080p, the Low-to-Ultra spread is 27 FPS. At 1440p, it is 20 FPS. At 4K, it is 13 FPS. This trend indicates that the GPU's pixel fill rate dominates over settings-dependent effects; as pixel count rises, the relative cost of Ultra settings diminishes because the GPU is already maxed out on pixel output.

Settings Recommendations

Given the measured data, the only configuration that approaches a playable experience is 1080p Low, which delivers 59 FPS average. This is the sole setting that allows the game to run near 60 FPS, and it represents the best possible experience for this combo in Control. The 1080p Medium preset at 42 FPS is the next best option, offering a smoother visual quality while maintaining a frame rate that is acceptable for a single-player action-adventure title. However, the 17 FPS drop from Low to Medium (59 to 42) suggests that the Medium preset's additional visual features come at a significant performance cost.

For players who prioritize visual fidelity over frame rate, 1080p High at 37 FPS is a viable option, though it sits below the 40 FPS threshold that many consider the minimum for comfortable play. The Ultra preset at 1080p (32 FPS) is not recommended, as it provides marginal visual improvements over High while sacrificing 5 FPS. The data shows that the gap between High and Ultra is the same as the gap between Medium and High (5 FPS), so the cost-benefit ratio of Ultra is poor.

At 1440p, the Low preset at 45 FPS is the only configuration that exceeds 30 FPS. The Medium preset at 32 FPS is technically above 30 FPS but leaves little headroom for demanding scenes, as indicated by the 27 FPS minimum. The High (28 FPS) and Ultra (25 FPS) presets at 1440p fall below the playability threshold and should be avoided. 4K is not recommended at any settings, as even the Low preset at 30 FPS is borderline, and the Medium preset's 21 FPS average with an 18 FPS minimum would result in a frustrating experience during combat or heavy particle effects.

The overall recommendation is to run Control at 1080p with Low settings for the most consistent frame rate, or 1080p Medium if the user can tolerate occasional dips toward 36 FPS. The game's action-adventure genre benefits from smooth frame pacing, and the 1080p Low configuration's 59 FPS average provides the best chance of maintaining acceptable performance during intense sequences. For users with a display that supports variable refresh rate, 1080p Medium at 42 FPS might be preferable for the improved visual quality, but the data suggests that the GPU's fundamental limits will cause noticeable stutter in complex scenes. The 1440p and 4K resolutions are only suitable for static screenshots or very undemanding exploration, as the frame rates in those configurations are too low for reactive gameplay.