Control

Resolution Scaling

The measured frame rates across the three resolutions tell a clear story about where this combination's bottleneck lies. At 1920x1080 with Low settings, the system delivers 100.1 FPS average, but this drops to 73 FPS at 2560x1440 and 50.7 FPS at 3840x2160. That represents a 27% reduction from 1080p to 1440p, and a further 30.5% reduction from 1440p to 4K. These are substantial scaling penalties that indicate the graphics card is the limiting factor at higher resolutions.

Moving through the preset ladder at 4K reinforces this conclusion. The Low preset achieves 50.7 FPS, Medium drops to 38.1 FPS, High falls to 30.3 FPS, and Ultra bottoms out at 26.5 FPS. The performance gap between Low and Ultra at 4K is 24.2 FPS, or roughly 47.7% of the Low preset's performance. This steep drop-off across presets at 4K suggests that the GPU's shading resources and memory bandwidth are being saturated as quality settings increase.

At 1080p, the scaling pattern changes character. The Low preset hits 100.1 FPS, but High only reaches 61.8 FPS, and Ultra manages 56 FPS. The difference between Low and Ultra at 1080p is 44.1 FPS, which is a wider absolute gap than at any other resolution. This indicates that at 1080p, the CPU has enough headroom to feed the GPU at lower settings, but the GPU still becomes the constraint as presets increase. The 9800X3D's strong single-thread performance keeps the frame rate from collapsing at lower resolutions, but the GTX 1070's Pascal architecture simply cannot sustain high frame rates with heavier graphical loads.

The resolution scaling data shows a system that is GPU-limited across the board. If the CPU were the primary constraint, we would expect smaller performance gaps between 1080p and 1440p, since the CPU workload does not change with resolution. Instead, the consistent frame rate drops at every resolution step confirm that the GTX 1070 is the component determining playability here.

GPU Role

The NVIDIA GeForce GTX 1070 is built on the Pascal architecture using the GP104 chip, manufactured on a 16 nm process at TSMC with 7,200 million transistors. Its base clock runs at 1506 MHz with a boost clock of 1683 MHz, and it carries 8 GB of GDDR5 memory on a 256-bit bus, yielding 256.3 GB/s of memory bandwidth. These specifications place the card in the 51st percentile among all GPUs in the database, with an average benchmark score of 12331.

The nearest rivals in the GPU database provide context for this card's position. The Quadro K4200 scores 12241, which is 0.7% lower than the GTX 1070's average. The GTX 880M scores 12490, representing a 1.3% advantage over the 1070. The GTX 1650 SUPER scores 12504, which is 1.4% higher, and the Tesla K20Xm scores 12547, a 1.7% edge. These are very tight margins, indicating that the GTX 1070 sits in a crowded performance band where even small architectural differences can shift rankings.

In Control, the GPU's 8 GB VRAM capacity appears sufficient for the measured presets, as the frame rates scale consistently with resolution and settings rather than showing sudden drops that would indicate memory exhaustion. The 256-bit memory bus and 256.3 GB/s bandwidth provide the data throughput needed for 4K rendering, though the card's 6.463 TFLOPS of FP32 compute and 107.7 GPixel/s pixel rate become limiting factors at higher presets. The boost clock of 1683 MHz is modest by modern standards, and the measured frame rates reflect that the card cannot maintain high performance when pixel count and shading complexity increase simultaneously.

CPU Role

The AMD Ryzen 7 9800X3D is a Zen 5 architecture processor codenamed Granite Ridge, built on TSMC's 4 nm process with 8,315 million transistors on a 70.6 mm² die. It features 8 cores and 16 threads, with a base clock of 4.70 GHz and a boost clock of 5.20 GHz. The 96 MB of shared L3 cache is a defining characteristic of this chip, and it supports DDR5 memory in a dual-channel configuration with 89.6 GB/s of bandwidth. These specifications place the CPU in the 90th percentile among all processors, with an average benchmark score of 37780.

The CPU's nearest rivals show how close the top of the benchmark charts can be. The Xeon 6353P scores 37811, which is 0.1% higher than the 9800X3D. The Core Ultra 5 225 scores 37831, also 0.1% higher. The Core i9-13900HK scores 38020, a 0.6% advantage, and the Core i5-13600KF scores 38028, 0.7% higher. These delta percentages are minimal, meaning the 9800X3D's performance is statistically indistinguishable from these rivals in aggregate benchmarks.

For Control specifically, the CPU's capabilities are only partially utilized. The 3DMark 16-thread score of 10081 and max-thread score of 10067 indicate strong multi-threaded performance, but the single-thread score of 1212 and 2-thread score of 2394 show that the chip also excels at lightly threaded workloads. Control is not a particularly CPU-heavy game at the resolutions tested, so the 9800X3D's role is to ensure that the GTX 1070 never starves for draw calls or physics data. The high boost clock of 5.20 GHz and large L3 cache help maintain consistent frame delivery, but the measured data shows the GPU remains the performance ceiling.

Measured FPS Breakdown

At 1920x1080, the GTX 1070 with the 9800X3D produces a wide spread of results. Low settings yield 100.1 FPS average, Medium achieves 68.4 FPS, High reaches 61.8 FPS, and Ultra manages 56 FPS. The gap between Low and Medium is 31.7 FPS, which is the largest single-step drop across all settings at this resolution. From Medium to High, the drop is 6.6 FPS, and from High to Ultra, it narrows to 5.8 FPS. This pattern suggests that the Low preset removes enough graphical work to let the GPU stretch, while the higher presets compress the performance range as the card approaches its limits.

At 2560x1440, the same settings produce 73 FPS on Low, 52.6 FPS on Medium, 46 FPS on High, and 38.3 FPS on Ultra. The Low-to-Medium drop is 20.4 FPS, Medium-to-High is 6.6 FPS, and High-to-Ultra is 7.7 FPS. The 1440p results show a more uniform scaling pattern than 1080p, with each preset step costing between 6.6 and 20.4 FPS. The Low preset at 1440p still clears 70 FPS, which is playable, but the higher presets fall below 60 FPS.

At 3840x2160, the results become more constrained. Low settings average 50.7 FPS, Medium drops to 38.1 FPS, High falls to 30.3 FPS, and Ultra bottoms out at 26.5 FPS. The Low-to-Medium gap is 12.6 FPS, Medium-to-High is 7.8 FPS, and High-to-Ultra is 3.8 FPS. The 4K results show a compression of the performance envelope, with the difference between Low and Ultra being only 24.2 FPS compared to 44.1 FPS at 1080p. This indicates that at 4K, even the Low preset saturates the GPU to a significant degree, leaving less headroom for higher settings to make a large impact.

How This Combo Ranks

This specific combination of the Ryzen 7 9800X3D and GTX 1070 ranks 1375 out of 1636 tested combos in Control. That places it in the bottom 16% of all combinations in the database, which is expected given that the GPU sits at the 51st percentile while the CPU sits at the 90th percentile. The pairing creates a significant imbalance: the CPU is capable of far more than the GPU can deliver in this game.

The rank of 1375 out of 1636 means there are 261 combos that perform worse than this pairing. This is not a bottom-tier result, but it is far from competitive. The data suggests that most tested systems achieve better frame rates in Control, likely because they pair more modern GPUs with high-end CPUs. The 9800X3D's strong benchmark scores provide no benefit in this game when the GTX 1070 is the rendering component, as the measured FPS values are entirely consistent with a GPU-bound scenario.

The delta between the CPU's 90th percentile ranking and the combo's bottom-16% ranking illustrates the magnitude of the GPU bottleneck. This is a case where the whole is significantly less than the sum of its parts, and the measured data reflects that the GPU determines the experience. The 9800X3D could support much higher frame rates if paired with a more capable GPU, but within this combo, its performance headroom goes unused.

Settings Recommendations

The measured FPS rows provide clear guidance for the best experience at each resolution. At 1920x1080, the High preset at 61.8 FPS offers the best balance of visual quality and smoothness, though the Low preset at 100.1 FPS is the only option that approaches high-refresh-rate territory. Medium at 68.4 FPS also clears 60 FPS, but High adds meaningful visual improvements for a relatively small 6.6 FPS cost. Ultra at 56 FPS is marginally playable, but the 5.8 FPS gain from High to Ultra is not worth the additional graphical load.

At 2560x1440, the Low preset at 73 FPS is the only setting that comfortably exceeds 60 FPS. Medium at 52.6 FPS is playable but noticeably below the 60 FPS threshold, and High at 46 FPS and Ultra at 38.3 FPS are increasingly compromised. The data suggests that 1440p users should prioritize the Low preset for smooth gameplay, or accept Medium if visual fidelity is more important than frame rate consistency. The 20.4 FPS drop from Low to Medium is steep, making Low the most sensible default.

At 3840x2160, no preset reaches 60 FPS. The Low preset at 50.7 FPS is the most playable option, with Medium at 38.1 FPS and High at 30.3 FPS falling below acceptable thresholds for fast-paced action. Ultra at 26.5 FPS is effectively unplayable for most users. The recommendation at 4K is to use the Low preset, as it is the only setting that maintains a frame rate near the 50 FPS range, and the visual differences between Low and Medium at this resolution are less pronounced than at lower resolutions due to pixel density.

Across all resolutions, the Low preset provides the most consistent experience, with frame rates of 100.1 FPS at 1080p, 73 FPS at 1440p, and 50.7 FPS at 4K. Users seeking higher visual quality should target 1080p with High settings, where the 61.8 FPS result represents the best trade-off between image quality and playability for this combination.