Control

Resolution Scaling

The measured FPS data for Control with the Intel Core i7-12700KF and NVIDIA GeForce GTX 1660 SUPER reveals a clear pattern of GPU-bound scaling as resolution increases. At 1080p with Low settings, the combo delivers 91.9 FPS average. Moving to 1440p Low drops that to 67.7 FPS, a reduction of roughly 26%. Pushing to 4K Low further reduces performance to 48.5 FPS, which is about 28% lower than the 1440p result. This steady decline across each resolution step indicates the GTX 1660 SUPER is the limiting component; the CPU has ample headroom at these settings.

The scaling behavior becomes more pronounced at higher quality presets. At High settings, the frame rate falls from 56 FPS at 1080p to 42.4 FPS at 1440p, then to 30.3 FPS at 4K. The 4K High result is 46% below the 1080p High figure. Ultra settings show the steepest absolute drop: 47.8 FPS at 1080p, 37.6 FPS at 1440p, and 23.1 FPS at 4K. The 4K Ultra result sits only slightly above the threshold often considered playable, indicating that this preset is not viable at that resolution.

Notably, the gap between Low and Ultra narrows as resolution increases. At 1080p, the difference between Low and Ultra is 44.1 FPS (91.9 vs. 47.8). At 4K, that gap shrinks to 25.4 FPS (48.5 vs. 23.1). This compression suggests that at higher resolutions, the GPU’s pixel throughput becomes the dominant constraint, and settings changes have a relatively smaller impact on raw FPS. At 1080p, the GPU can still push enough frames that quality settings matter more, but the absolute ceiling is limited by the 6 GB memory and 192-bit bus of the GTX 1660 SUPER.

Data indicates that 1440p is the practical ceiling for this combo if the target is 60 FPS, and only at Low settings (67.7 FPS). At 1080p, Medium delivers 62.9 FPS, which is above the 60 FPS mark, while High falls just short at 56 FPS. For 4K, no preset reaches 60 FPS; the best case is Low at 48.5 FPS.

How This Combo Ranks

In the database of tested CPU-GPU combinations for Control, this pairing ranks 1463 out of 1636 combos. That places it in the lower portion of the distribution, roughly in the 11th percentile of tested configurations. The rank reflects the GTX 1660 SUPER’s position in the GPU hierarchy; the GPU’s benchmark percentile versus all GPUs is 55, meaning it sits near the middle of the pack overall, but in this specific game’s measured results, the combo underperforms relative to many others.

The CPU’s benchmark standing is much stronger. The Intel Core i7-12700KF has an average benchmark score of 35355 and ranks at the 89th percentile versus all CPUs. Its nearest rivals in the CPU benchmarks are tightly clustered: the Intel Core i5-14400 scores 35336 (a delta of 0.1% higher), the Intel Core i7-13800H scores 35416 (0.2% higher), the Intel Core i9-12900HX scores 35289 (0.2% lower), and the Intel Core i5-13600T scores 35256 (0.3% lower). These deltas are negligible, indicating that the 12700KF is performance-equivalent to a broad set of modern mid-range and high-end processors.

The GPU, by contrast, has an average benchmark score of 14286 and sits at the 55th percentile. Its nearest rivals are the AMD Radeon RX Vega 11 (14314, 0.2% higher), the Intel Iris Xe MAX Graphics (14315, 0.2% higher), the AMD Radeon Vega 11 (14352, 0.5% higher), and the AMD Radeon RX 5500 XT (14389, 0.7% higher). These small deltas show the GTX 1660 SUPER is positioned at a performance tier where small architectural differences produce minimal score changes. In Control, this GPU tier appears insufficient to push the combo higher in the rankings, even with a strong CPU.

The combination rank of 1463 out of 1636 indicates that most tested configurations outperform this one in Control. The data suggests that the GPU is the binding constraint; a more capable GPU paired with the same CPU would likely move the combo substantially higher, given the CPU’s high percentile ranking.

Settings Recommendations

For a target of 60 FPS average, the measured data shows that only two settings at 1080p meet that threshold: Low at 91.9 FPS and Medium at 62.9 FPS. High at 1080p produces 56 FPS, slightly below 60. At 1440p, only Low achieves 67.7 FPS; Medium falls to 46.5 FPS, which is below a 60 FPS target but still playable for many users. At 4K, no preset reaches 60 FPS, and the highest result is Low at 48.5 FPS.

If the goal is visual quality while maintaining 60 FPS, the data points to 1080p Medium as the optimal choice. It delivers 62.9 FPS, which is above the 60 FPS threshold, and provides a meaningful quality improvement over Low. The gap between Medium and High at 1080p is 6.9 FPS (62.9 vs. 56), suggesting that High is a modest step up in quality but costs more than 10% of the frame rate.

For users willing to accept a 30 FPS target, 1440p Ultra at 37.6 FPS or 4K Medium at 34.3 FPS become viable. However, 4K High at 30.3 FPS sits right at the edge of playability, and 4K Ultra at 23.1 FPS is likely too low for smooth gameplay. The data indicates that the best experience for a 1440p display is Low settings, since Medium drops to 46.5 FPS and High to 42.4 FPS, both of which are below 60 but above 30.

The measured rows show a consistent pattern: Low settings are always the highest FPS option at any resolution, and the delta between Low and Medium is larger than the delta between Medium and High. For example, at 1080p, Low-to-Medium is a 29 FPS drop (91.9 to 62.9), while Medium-to-High is only 6.9 FPS. At 1440p, Low-to-Medium is 21.2 FPS, and Medium-to-High is 4.1 FPS. At 4K, Low-to-Medium is 14.2 FPS, and Medium-to-High is 4 FPS. This suggests that the Medium preset captures most of the visual improvements with a disproportionate FPS cost, while High adds less overhead.

Measured FPS Breakdown

At 1920x1080, the average FPS figures are as follows: Low produces 91.9 FPS, Medium produces 62.9 FPS, High produces 56 FPS, and Ultra produces 47.8 FPS. The spread from Low to Ultra is 44.1 FPS, which is the largest across all resolutions. The step from Low to Medium is the most severe single drop at 29 FPS, representing a 31.6% reduction. Medium to High is a 6.9 FPS drop (10.9% lower), and High to Ultra is an 8.2 FPS drop (14.6% lower).

At 2560x1440, the figures are: Low at 67.7 FPS, Medium at 46.5 FPS, High at 42.4 FPS, and Ultra at 37.6 FPS. The Low-to-Medium drop is 21.2 FPS (31.3% lower), Medium-to-High is 4.1 FPS (8.8% lower), and High-to-Ultra is 4.8 FPS (11.3% lower). The total spread is 30.1 FPS. Compared to 1080p, every preset loses roughly 26-28% of its FPS when moving to 1440p, with Low losing 24.2 FPS and Ultra losing 10.2 FPS.

At 3840x2160, the figures are: Low at 48.5 FPS, Medium at 34.3 FPS, High at 30.3 FPS, and Ultra at 23.1 FPS. The Low-to-Medium drop is 14.2 FPS (29.3% lower), Medium-to-High is 4 FPS (11.7% lower), and High-to-Ultra is 7.2 FPS (23.8% lower). The total spread is 25.4 FPS. The 4K results show that Low is the only preset with an average above 30 FPS, and Ultra falls below 25 FPS.

The data also shows that the relative performance penalty for Ultra versus Low grows as resolution decreases. At 1080p, Ultra is 48% of Low’s FPS (47.8 vs. 91.9). At 1440p, Ultra is 55.5% of Low (37.6 vs. 67.7). At 4K, Ultra is 47.6% of Low (23.1 vs. 48.5). This non-linear relationship suggests that at 4K, the GPU is so heavily loaded that even Low cannot fully escape the resolution penalty, but the absolute FPS numbers are low enough that quality settings have a smaller absolute impact.

FAQ

Q: Can this combo run Control at 4K 60 FPS?

A: No. The highest 4K average FPS measured is 48.5 at Low settings. Medium produces 34.3, High produces 30.3, and Ultra produces 23.1. None approach 60 FPS.

Q: What is the best settings preset for a 60 FPS target at 1080p?

A: Medium is the only preset above 60 FPS at 1080p, delivering 62.9 FPS. Low delivers 91.9 FPS, but Medium offers a quality improvement while still meeting the target. High falls short at 56 FPS.

Q: How does the CPU compare to its nearest rivals?

A: The Intel Core i7-12700KF scores 35355 on average, which is 0.1% higher than the Intel Core i5-14400, 0.2% lower than the Intel Core i7-13800H, 0.2% higher than the Intel Core i9-12900HX, and 0.3% higher than the Intel Core i5-13600T. These differences are within 0.3%.

Q: Is the GPU or CPU the bottleneck in this game?

A: The data indicates the GPU is the limiting component. The CPU ranks at the 89th percentile versus all CPUs, while the GPU ranks at the 55th percentile. The FPS scaling with resolution shows steady declines that are characteristic of a GPU-bound scenario.

Q: What FPS can I expect at 1440p with High settings?

A: The measured average FPS at 1440p High is 42.4. This is below 60 FPS but above 30 FPS, making it playable for slower-paced action but not ideal for fast reactions.

Q: How does this combo rank among all tested configurations in Control?

A: It ranks 1463 out of 1636 combos, placing it in the lower portion of the distribution. This reflects the GTX 1660 SUPER’s mid-tier status rather than the CPU’s performance.

CPU Role

The Intel Core i7-12700KF provides 12 cores and 20 threads, with a base clock of 3.60 GHz and a boost clock of 5.00 GHz. This is an Alder Lake-S desktop processor built on Intel’s 10 nm process, with a 125 W TDP and support for DDR4 and DDR5 memory. The CPU’s benchmark scores show strong multi-threaded and single-threaded capabilities: Cinebench R23 multicore at 28979, Cinebench R23 singlecore at 4091, and Geekbench multicore at 13596 with singlecore at 2526.

In Control, the CPU’s high benchmark percentile (89th) does not translate into a high combo rank because the game’s FPS is constrained by the GPU. The data shows that at 1080p Low, the CPU is clearly not the limiting factor, since the combo achieves 91.9 FPS. At higher resolutions, the FPS drops consistently, which is a hallmark of GPU limitation rather than CPU limitation. The CPU’s 3DMark 16-thread score of 9282 and max-thread score of 9983 indicate it can handle many concurrent tasks, but Control’s rendering pipeline appears to rely more heavily on the GPU’s fill rate and memory bandwidth.

The GTX 1660 SUPER has 6 GB of GDDR6 memory on a 192-bit bus, with a bandwidth of 336.0 GB/s. Its FP32 performance is 5.027 TFLOPS, and it has 1408 shading units. The GPU’s 12 nm Turing architecture supports DirectX 12 (12_1), but its 55th percentile score of 14286 places it well below the CPU in relative performance. The CPU’s nearest rivals all score within 0.3% of the 12700KF, confirming that processor choice is not a differentiator in this combo; the GPU tier is what determines Control’s playability.

The measured results show that even at 4K Low, the CPU can feed the GPU enough frames to hit 48.5 FPS, which is a modest result that suggests the CPU is not causing stutter or frame drops. The CPU’s L3 cache is 25 MB shared, and its per-core L1 and L2 caches are 80 KB and 1.25 MB respectively, which are ample for game logic and physics workloads. In summary, the CPU is a high-end part that is effectively wasted in this pairing for Control, since the GPU cannot leverage its full potential at any resolution.