Thief

Thief, a 2014 stealth title, responds predictably to modern hardware, but the measured data for the Intel Core i5-10400F and NVIDIA GeForce RTX 5060 reveals specific bottlenecks and scaling patterns. The data shows a system that is heavily GPU-limited at higher resolutions, but which transitions to a more balanced state at 1080p, where the CPU's capabilities become more relevant. This analysis walks through the measured FPS to understand how this combination behaves across the tested settings.

Resolution Scaling

The most significant finding in the measured data is the dramatic impact of resolution on performance. At 4K (3840x2160), the average frame rate under Ultra settings is 43.4 FPS, which is only marginally different from the High preset at 43.6 FPS. This near-identical performance between High and Ultra at 4K is a clear indicator that the rendering pipeline is saturated; the GPU is the limiting component. The difference between these two presets is a negligible 0.2 FPS, suggesting that the additional graphical load of Ultra over High is not the primary constraint at this resolution.

Moving down to 1440p (2560x1440), the situation changes notably. The Ultra preset yields 72.5 FPS, while High produces 75.8 FPS, a more pronounced gap of 3.3 FPS. This indicates that the GPU is still the primary bottleneck, but it has more headroom to process the extra effects of the Ultra preset. The scaling from 4K Ultra to 1440p Ultra shows a substantial 66.9% performance increase, underscoring how much pixel count affects this GPU’s workload.

At 1080p (1920x1080), the scaling pattern shifts again. The Ultra preset delivers 102.6 FPS, and the High preset achieves 104.2 FPS. The difference between these two settings is only 1.6 FPS, which is smaller than the gap seen at 1440p. This compression of the performance delta between presets at lower resolutions suggests that the CPU, the Intel Core i5-10400F, is beginning to assert its influence. With fewer pixels to render, the GPU can complete its work faster, and the frame rate becomes increasingly dependent on the CPU's ability to feed the graphics card with draw calls. The data shows the CPU's multi-threaded benchmark score of 10,297 in Cinebench R23, which is a mid-range result, indicating it can handle the game's logic but is not a high-end part.

The overall trajectory is clear: as resolution drops, the frame rate increases disproportionately, confirming a GPU-bound scenario at 4K that gradually becomes more balanced at 1080p. The jump from 1440p High (75.8 FPS) to 1080p High (104.2 FPS) is a 37.5% increase, which is a larger relative gain than the jump from 4K Ultra to 1440p Ultra. This non-linear scaling is typical for a game like Thief, where the pixel fill rate is a dominant factor at high resolutions.

Settings Recommendations

The measured rows provide a clear hierarchy of settings for this game and hardware combination. At 4K, the choice is stark. The Low preset is the only option that provides a playable experience, with an average of 78.5 FPS. Medium drops to 58.8 FPS, which is borderline, while High and Ultra are effectively unplayable at 43.6 and 43.4 FPS respectively. The data suggests that for a smooth 4K experience, one must accept the Low preset, which demonstrates a 34.9% performance improvement over Medium.

For 1440p, the recommendations are more flexible. The Ultra preset at 72.5 FPS is above the 60 FPS threshold and is viable for a high-fidelity experience. However, the High preset at 75.8 FPS offers a better balance, providing a 4.6% performance uplift over Ultra with presumably minor visual differences. The Low preset at 110.9 FPS is excessively high for this resolution, suggesting that the GPU is not being sufficiently stressed and the CPU is becoming a more significant factor. The Medium preset at 97.6 FPS is a solid middle ground for those who want a high refresh rate experience.

At 1080p, the data indicates that the system is capable of high refresh rates. The Ultra preset at 102.6 FPS and the High preset at 104.2 FPS are both excellent for a 100Hz+ display. The difference between these two is negligible, so users should opt for Ultra to maximize visual quality. The Low preset at 122.9 FPS and Medium at 115.9 FPS offer diminishing returns in terms of FPS for the loss in visual fidelity. The data shows that the CPU's single-threaded performance, with a Passmark score of 2,541, is sufficient to keep the frame rate high at this resolution, but the gains from lowering settings from High to Low are only 17.9%, a smaller relative increase than at higher resolutions.

The best experience per the measured rows is context-dependent. For 4K, it is Low. For 1440p, High provides the best combination of visual quality and performance. For 1080p, Ultra is the clear choice, as the hardware has ample headroom to run the game at its highest settings.

GPU Role

The NVIDIA GeForce RTX 5060 plays a central role in this game's performance, particularly at higher resolutions. The data shows that the GPU's 8 GB of GDDR7 memory on a 128-bit bus, with a bandwidth of 448.0 GB/s, is sufficient for Thief's demands, as the game is from 2014 and likely has modest VRAM requirements. The memory clock is listed as 1750 MHz, with an effective data rate of 28 Gbps. This memory configuration is not a limiting factor in the measured results.

The GPU's compute capabilities are substantial, with a FP32 performance of 19.18 TFLOPS and a pixel rate of 119.9 GPixel/s. The benchmark results show a Passmark G3D score of 20,891, which places it in the 73rd percentile of all GPUs. This level of performance is what drives the high frame rates at 1080p and the playable frame rates at 4K on the Low preset. The GPU's boost clock of 2497 MHz, with a base clock of 2280 MHz, provides the raw speed needed for the title's rendering tasks.

The relationship between the GPU and the game's settings is clear. At 4K, the GPU is the absolute bottleneck, as evidenced by the negligible difference between High (43.6 FPS) and Ultra (43.4 FPS). This suggests that the GPU is fully saturated and cannot process additional graphical features without dropping the frame rate. In contrast, at 1080p, the GPU has more headroom, and the difference between Low (122.9 FPS) and Ultra (102.6 FPS) is more pronounced, showing that the GPU can handle the extra work when it isn't overwhelmed by pixel count. The GPU's performance is the primary driver of the scaling patterns observed across the resolution and settings matrix.

How This Combo Ranks

The combination of the Intel Core i5-10400F and NVIDIA GeForce RTX 5060 ranks 980th out of 1,642 tested combinations in Thief. This places it in the 59.7th percentile of all tested combos in this game, meaning it outperforms a majority of systems but is far from the top tier. The rank is a holistic measure of the system's ability to run the game across all tested settings and resolutions.

The CPU, the Intel Core i5-10400F, has an average benchmark score of 14,168 and is in the 73rd percentile of all CPUs. Its nearest rivals in terms of average score are the AMD EPYC 7552 (14,115, a 0.4% delta) and the AMD Ryzen 3 7320C (14,277, a -0.8% delta). These scores are very close, indicating that the i5-10400F's compute performance is comparable to these parts. However, the CPU's multi-threaded performance, as seen in Cinebench R23 (10,297), is not exceptionally high, which may hold back the system in CPU-bound scenarios at lower resolutions.

The GPU, the NVIDIA GeForce RTX 5060, has an average benchmark score of 30,109 and is also in the 73rd percentile of all GPUs. Its nearest rivals are the AMD Radeon RX 7700 XT (30,097, a 0% delta) and the AMD Radeon RX 570 (30,158, a -0.2% delta). The performance parity between these GPUs is notable, but the RTX 5060's actual in-game performance is what determines the frame rates. The combo rank of 980 indicates that while the individual components are solid, the overall system is not optimized for Thief, likely due to the CPU limiting the GPU at lower resolutions.

FAQ

Q: What is the best resolution to play Thief on with this hardware?

A: The data suggests 1440p is the optimal resolution. At 1440p, the High preset achieves 75.8 FPS, which is well above the 60 FPS target, while 4K only offers playable frame rates on the Low preset (78.5 FPS). 1080p is also viable, with Ultra achieving 102.6 FPS, but 1440p provides a better balance of visual fidelity and performance.

Q: Is the Intel Core i5-10400F or the NVIDIA GeForce RTX 5060 the bottleneck?

A: The bottleneck shifts with resolution. At 4K, the GPU is the clear bottleneck, as shown by the minimal performance difference between High (43.6 FPS) and Ultra (43.4 FPS) presets. At 1080p, the CPU becomes more of a factor, as the performance difference between Low (122.9 FPS) and Ultra (102.6 FPS) is larger, indicating the GPU has headroom but the CPU cannot feed it data fast enough to maintain the highest frame rates.

Q: Can this system run Thief at 4K?

A: Yes, but only with significant graphical compromises. The measured data shows that the Low preset at 4K yields an average of 78.5 FPS, which is playable. However, the Medium preset at 58.8 FPS is borderline, and the High and Ultra presets are not playable at 43.6 and 43.4 FPS respectively.

Q: How does the RTX 5060's performance compare to its closest rivals?

A: Based on average benchmark scores, the RTX 5060 (30,109) is statistically tied with the AMD Radeon RX 7700 XT (30,097), with a 0% delta. It is also nearly identical to the AMD Radeon RX 570 (30,158), which has a -0.2% delta. These scores indicate comparable raw computing power among these GPUs.

Q: What is the frame rate difference between High and Ultra settings at 1080p?

A: The difference is minimal. The High preset averages 104.2 FPS, while the Ultra preset averages 102.6 FPS. This is a difference of only 1.6 FPS, suggesting that Ultra should be used at this resolution for the best visual quality without a meaningful performance penalty.

Measured FPS Breakdown

The following table details the exact measured average FPS for each resolution and settings combination.

| Resolution | Settings | Average FPS |

| :--------- | :------- | :---------- |

| 3840x2160 | High | 43.6 |

| 3840x2160 | Low | 78.5 |

| 3840x2160 | Medium | 58.8 |

| 3840x2160 | Ultra | 43.4 |

| 2560x1440 | High | 75.8 |

| 2560x1440 | Low | 110.9 |

| 2560x1440 | Medium | 97.6 |

| 2560x1440 | Ultra | 72.5 |

| 1920x1080 | High | 104.2 |

| 1920x1080 | Low | 122.9 |

| 1920x1080 | Medium | 115.9 |

| 1920x1080 | Ultra | 102.6 |

The data reveals a consistent pattern across resolutions. At 4K, the frame rate drops sharply as settings increase, with a 44.7% performance reduction from Low to High. At 1440p, the drop from Low to Ultra is 34.6%, indicating a less severe penalty. At 1080p, the reduction from Low to Ultra is 16.5%, showing that the system handles the graphical load much more efficiently at this resolution. The Ultra preset at 4K (43.4 FPS) is the lowest measured performance, while the Low preset at 1080p (122.9 FPS) is the highest. This spread of 79.5 FPS across the test matrix highlights how sensitive this game is to both resolution and quality settings.