Control

Control on the Intel Core Ultra 5 235 and NVIDIA GeForce GTX 1050 Ti pairing is a study in severe bottlenecking, where the data reveals a system struggling to maintain playable framerates at nearly every setting. The benchmark results place this combination at rank 2854 out of 2888 tested combos, firmly in the bottom percentile of all configurations. This is not a system built for smooth gameplay, but rather one where the user must make extreme compromises to achieve even marginal playability.

Settings Recommendations

The benchmark data offers a clear hierarchy of performance across the four tested presets, with the most playable experience found at the lowest settings. At 1920x1080, the Low preset delivers an average of 46 FPS, which is the only configuration in the entire dataset that approaches a standard of smoothness. This is a significant jump from the Medium preset, which averages 32 FPS at the same resolution, representing a 43.75% improvement in average framerate when dropping from Medium to Low.

The data suggests that the Low preset is the only viable option for an acceptable experience, but even this comes with caveats. The 46 FPS average is still below the 60 FPS target that many players consider ideal, and the lack of min/max FPS data for this setting leaves uncertainty about frame pacing and potential stutters. The Medium preset, while offering better visual fidelity, drops to an average of 32 FPS, which is borderline playable but likely to feel choppy in fast-paced action sequences.

At higher resolutions, the situation becomes more dire. At 2560x1440, Low settings yield 34 FPS, and at 3840x2160, Low drops to 23 FPS. The Ultra preset is effectively unplayable across all resolutions, with a maximum average of 25 FPS at 1080p and a dismal 13 FPS at 4K. The recommendation is unambiguous: for any semblance of playability, the Low preset at 1080p is the only choice. Medium at 1080p could be considered for slower-paced exploration, but the data indicates that combat scenarios would likely suffer from the sub-30 FPS average.

How This Combo Ranks

The comboRankInGame of 2854 out of 2888 places this system in the bottom 1.2% of all tested combinations for Control. This is a remarkably poor showing, underscoring the fundamental mismatch between the modern 14-core CPU and the aging 4 GB GPU. To contextualize this, the GPU alone ranks in the 25th percentile among all GPUs, while the CPU ranks in the 89th percentile among all CPUs. The massive gap between the CPU's high percentile and the GPU's low percentile highlights that the graphics card is the overwhelming limiting factor.

The nearest GPU rivals provide insight into the performance tier this system occupies. The GTX 1050 Ti sits within 1.8% of the AMD Radeon Vega 3 and within 1.1% of the NVIDIA Quadro K3000M. These are integrated or low-end mobile graphics solutions, indicating that this desktop GPU is performing at a level comparable to entry-level integrated graphics. The data shows that this combination is not merely underperforming; it is operating in a performance stratum where even modest visual demands in Control cause the framerate to collapse.

GPU Role

The GTX 1050 Ti is the clear performance bottleneck in this pairing, and its specifications paint a stark picture of its limitations. The GPU features 4 GB of GDDR5 memory on a 128-bit bus, providing a bandwidth of 112.1 GB/s. This is a modest amount of memory and bandwidth for a game like Control, which is known for its detailed environments and particle effects. The 768 shading units and 32 ROPs are equally limited, resulting in a pixel rate of 44.54 GPixel/s and a texture rate of 66.82 GTexel/s.

The GPU's clock speeds are also constrained, with a base clock of 1291 MHz and a boost clock of 1392 MHz. These are low figures compared to modern GPUs, and the data suggests that even at these low clocks, the card is overwhelmed by the game's demands. The FP32 performance of 2.138 TFLOPS places it in a range where it is fundamentally unsuited for running Control at modern standards. The 4 GB memory capacity is particularly problematic, likely causing texture thrashing and stuttering as the GPU attempts to load assets, especially at higher resolutions where memory pressure increases.

The GPU's benchmarks reinforce its limitations. Its Passmark G3D score of 6340 and 3DMark Steel Nomad score of 305 are indicative of a card that was entry-level at its launch and has not aged well. The DirectX 12 Passmark score of 26 is especially low, which is relevant since Control utilizes DX12 features. The data strongly implies that the GPU's architectural limitations, particularly its Pascal-era design and lack of modern features like hardware ray tracing, are the primary cause of the poor performance observed.

Measured FPS Breakdown

At 1920x1080, the measured FPS reveals a steep performance cliff as settings increase. The Low preset achieves an average of 46 FPS, which drops to 32 FPS at Medium, 29 FPS at High, and 25 FPS at Ultra. The Medium preset has min/max data of 27/37 FPS, showing a narrow band of performance that suggests the GPU is consistently hitting its limit. The difference between Low and Medium is a substantial 14 FPS, while the difference between High and Ultra is a mere 4 FPS, indicating diminishing returns at the higher settings.

Moving to 2560x1440, the framerates decrease across the board. Low settings average 34 FPS, Medium averages 24 FPS, High averages 22 FPS, and Ultra averages 19 FPS. The Medium preset at this resolution has min/max of 21/28 FPS, showing that even the average is barely above the minimum, indicating a very consistent but low performance level. The gap between Low and Medium at 1440p is 10 FPS, a slightly smaller absolute drop than at 1080p, but the relative impact is similar.

At 3840x2160, the performance becomes critically low. Low settings average 23 FPS, Medium averages 17 FPS with min/max of 14/19 FPS, High averages 15 FPS, and Ultra averages 13 FPS. The data shows a clear pattern: the GPU is unable to maintain even 24 FPS at 4K under any setting, making this resolution entirely unusable for this combination. The framerates at 4K are so low that even the Low preset, which is the best performer, fails to provide a consistently viewable experience.

Resolution Scaling

The FPS drop from 1080p to 4K is dramatic and reveals the GPU's inability to handle increased pixel throughput. At Low settings, the average FPS goes from 46 at 1080p to 34 at 1440p to 23 at 4K. This represents a 26% drop from 1080p to 1440p and a 50% drop from 1080p to 4K. While some scaling is expected, the magnitude of these drops suggests that the GPU is severely constrained by its pixel fill rate and memory bandwidth.

At Medium settings, the scaling is even more pronounced. The average FPS goes from 32 at 1080p to 24 at 1440p to 17 at 4K, representing a 25% drop from 1080p to 1440p and a 46.9% drop from 1080p to 4K. The data indicates that the GPU is not scaling efficiently with resolution, as the performance loss is roughly proportional to the increase in pixel count, if not worse. This is a classic sign of a GPU that is limited by its ROPs and memory bandwidth rather than its shader cores.

The resolution scaling data strongly suggests that the system is entirely GPU-bound. If the CPU were the limiting factor, we would expect to see less of a performance drop at higher resolutions, as the CPU's workload does not increase as much as the GPU's when raising resolution. The fact that the framerate drops so precipitously with resolution confirms that the GTX 1050 Ti is the primary bottleneck. The CPU's powerful 14 cores are essentially idle, waiting for the GPU to finish its work.

FAQ

Q: What is the best resolution and settings combo for playable FPS?

A: The data shows that 1920x1080 with Low settings is the only configuration that provides a reasonable average framerate, hitting 46 FPS. All other settings at this resolution fall below 32 FPS, and all higher resolutions fail to reach even 35 FPS at any setting.

Q: How does the GTX 1050 Ti compare to its nearest rivals?

A: The GTX 1050 Ti's average benchmark score of 4193 is 0.6% higher than the AMD Radeon R5 M330 and 1% higher than the NVIDIA Quadro K2100M. It is 1.1% lower than the NVIDIA Quadro K3000M and 1.8% lower than the AMD Radeon Vega 3.

Q: Is the CPU or GPU the limiting factor in this combination?

A: The data points overwhelmingly to the GPU. The CPU ranks in the 89th percentile of all CPUs, while the GPU ranks in the 25th percentile. The severe FPS drops from 1080p to 4K indicate that the GPU's pixel processing capabilities are the primary constraint.

Q: What is the average FPS difference between Low and Ultra settings at 1080p?

A: At 1920x1080, the Low preset averages 46 FPS, while the Ultra preset averages 25 FPS. This represents a 21 FPS difference, or a 45.7% decrease in average framerate when moving from Low to Ultra.

Q: Can this system run Control at 4K resolution?

A: The data indicates no. At 3840x2160, the highest average FPS is 23 at Low settings, which drops to 13 FPS at Ultra. These framerates are far below the threshold for playable performance.

Q: How does the combo rank compare to other tested systems?

A: The combination ranks 2854 out of 2888 tested combos for Control, placing it in the bottom 1.2% of all configurations. This is a very poor ranking, consistent with the low FPS measurements.

CPU Role

The Intel Core Ultra 5 235 has 14 cores and 14 threads, with a base clock of 3.40 GHz and a boost clock of 5.00 GHz. This is a high-end desktop processor, and its Cinebench R23 multi-core score of 14769 and single-core score of 2085 confirm its strength. The CPU's Passmark single-thread score of 4516 places it in the 89th percentile of all CPUs, making it a powerful processor that should have no trouble feeding frames to a GPU in most scenarios.

However, in this specific combination, the CPU's power is entirely irrelevant to the game's performance. The data shows that the CPU is operating at a fraction of its capability, held back by the GPU's inability to process frames quickly enough. The CPU's 24 MB of shared L3 cache and 102.4 GB/s memory bandwidth are more than sufficient for gaming, but these resources go unused when the GPU is the bottleneck.

The CPU's benchmark scores, such as the Passmark multi-thread score of 37816 and the Passmark physics score of 2570, indicate a processor capable of handling complex game logic and physics calculations. Yet, in Control, the framerate is determined almost entirely by the GPU. The measured FPS data shows no evidence of CPU limitation, as the framerates scale consistently with resolution and settings, which are GPU-dependent variables. The conclusion is that while the CPU is a top-tier performer, its capabilities are wasted in this pairing, and any attempt to improve performance would require a GPU upgrade.