Thief

Thief, a 2014 stealth title, runs on a modern high-end mobile platform with a clear performance profile: the Intel Core Ultra 9 275HX and NVIDIA GeForce RTX 5080 combination delivers high frame rates at 1080p and 1440p, but the 4K Ultra setting is the only point where the GPU truly struggles. The measured data shows a system that is heavily GPU-limited at higher resolutions, while the CPU has ample headroom even at its lowest settings. This analysis breaks down the scaling across resolutions, settings, and what the numbers mean for a practical gaming experience.

Resolution Scaling

The most striking trend in the measured data is how average frame rates collapse as resolution increases, which points directly to the RTX 5080 as the primary bottleneck at 4K. At 1080p with High settings, the combo produces 145 FPS, but moving to 1440p drops that to 99 FPS, a 31.7% reduction. Scaling further to 4K, the average falls to 58 FPS, which is a 60% decrease from the 1080p baseline. This steep drop-off is classic GPU-bound behavior; the CPU has to feed fewer frames as pixel count rises, but the GPU's workload increases dramatically.

The pattern repeats across all settings. For Low settings, the progression is 232 FPS at 1080p, 172 FPS at 1440p (a 25.9% drop), and 101 FPS at 4K (a 56.5% drop from 1080p). Medium settings show 190 FPS at 1080p, 130 FPS at 1440p, and 77 FPS at 4K. Even the most demanding Ultra preset follows suit: 138 FPS at 1080p, 95 FPS at 1440p, and 56 FPS at 4K. The consistency of this scaling across all presets confirms that resolution, not the settings themselves, is the primary driver of performance loss.

What this means for the limiting component is clear: the Intel Core Ultra 9 275HX is never the constraint in this game. At 1080p Low, the system hits 232 FPS, which is a very high frame rate, but the CPU’s performance headroom is still not fully tested because Thief’s engine is not particularly demanding on modern processors. The 24-core, 24-thread CPU with a 5.40 GHz boost clock is far more capable than what this 2014 title requires. The data suggests that if the GPU were swapped for a weaker model, the frame rates would drop proportionally at all resolutions, but the CPU would still be waiting. The RTX 5080, with its 16 GB of GDDR7 memory and 960.0 GB/s bandwidth, is the component that dictates the final FPS, especially beyond 1080p.

The 4K results are particularly telling. At 4K High, the combo averages 58 FPS, which is just below the 60 FPS target many players consider smooth. At 4K Ultra, it drops to 56 FPS. These numbers are not dramatic failures, but they are not ideal for a high-refresh-rate monitor. The 4K Low setting, at 101 FPS, shows that the GPU can handle the resolution if the graphical load is reduced, but the gap between Low and Ultra at 4K is only 45 FPS. This suggests that the RTX 5080’s raw power is sufficient for 4K, but Thief’s specific rendering paths, likely its lighting and shadow effects, place a heavy burden on the GPU at higher presets.

FAQ

Q: Can this combo maintain 60 FPS at 4K Ultra in Thief?

A: No. The measured data shows an average of 56 FPS at 3840x2160 with Ultra settings, which falls short of a consistent 60 FPS target. Lowering to High settings brings the average to 58 FPS, still slightly below 60 FPS, while Medium and Low settings exceed it with 77 FPS and 101 FPS, respectively.

Q: What is the best resolution for a high-refresh-rate monitor?

A: For 144 Hz or higher displays, 1080p is the only resolution that consistently exceeds 144 FPS across all settings. The lowest measured average at 1080p is 138 FPS on Ultra, while High, Medium, and Low all exceed 145 FPS. At 1440p, only the Low preset (172 FPS) and Medium preset (130 FPS) come close to or exceed 144 FPS, while High and Ultra fall to 99 FPS and 95 FPS, respectively.

Q: How does the RTX 5080's performance in Thief compare to its nearest GPU rivals?

A: The RTX 5080’s average benchmark score is 56083, which places it 0.6% ahead of the AMD Radeon 8060S and 0.7% ahead of the AMD Radeon RX 6750 GRE 12 GB. It trails the AMD Radeon RX 9070 GRE by 2.2% in average score. In terms of percentile, the RTX 5080 ranks in the 87th percentile of all GPUs, indicating strong overall performance, but these rival deltas are small, suggesting that in-game frame rates would be very similar.

Q: Is the Intel Core Ultra 9 275HX a bottleneck in this game?

A: No, the data does not indicate a CPU bottleneck. The CPU’s benchmark scores are very high (e.g., 35589 in Cinebench R23 multi-core, 55759 in Passmark multithread), and it ranks in the 94th percentile of all CPUs. In Thief, the frame rates scale predictably with resolution and GPU load, which is the signature of a GPU-bound scenario. The CPU has 24 cores and a 5.40 GHz boost clock, far more than needed for a 2014 title.

Q: What is the frame time consistency like at different settings?

A: The only settings with measured minimum and maximum FPS data are Medium at all three resolutions. At 1080p Medium, the average FPS is 190, with a minimum of 161 and a maximum of 218. At 1440p Medium, the average is 130 FPS, with a minimum of 111 and a maximum of 150. At 4K Medium, the average is 77 FPS, with a minimum of 65 and a maximum of 88. These ranges indicate that frame pacing is relatively stable, with the minimum FPS staying above 60 in all cases.

Q: How does this combo rank among all tested configurations for Thief?

A: This specific combination of the Intel Core Ultra 9 275HX and RTX 5080 ranks 584th out of 3720 tested combos for this game. This places it in the top 15.7% of all tested systems, which is a strong result, but not the absolute top tier, likely due to the high demands of 4K Ultra settings.

Settings Recommendations

For the best experience, the optimal setting depends entirely on the target resolution and refresh rate. At 1080p, the data shows that Ultra settings are perfectly playable, with an average of 138 FPS. This is a high frame rate that will satisfy most gamers, and the visual upgrade from High (145 FPS) to Ultra is minimal in terms of FPS cost, only a 7 FPS difference. Therefore, at 1080p, Ultra is the recommended preset, as it delivers excellent performance without sacrificing visual fidelity.

Moving to 1440p, the recommendation shifts. Ultra settings produce 95 FPS, which is good, but High settings are nearly identical at 99 FPS. The difference between High and Ultra is only 4 FPS, so choosing High is a safe bet for a small performance gain with a negligible visual difference. However, for players with a 144 Hz monitor, neither High nor Ultra will hit the refresh rate ceiling. The Medium preset at 130 FPS is the sweet spot for high-refresh-rate 1440p gaming, offering a 31 FPS improvement over High while still looking substantially better than Low (172 FPS). The measured minimum FPS for Medium at 1440p is 111, which ensures smooth gameplay even in demanding scenes.

At 4K, the situation is more delicate. The Ultra preset averages 56 FPS, which is below the 60 FPS target, and even High only reaches 58 FPS. These are not ideal for a smooth experience on a standard 60 Hz display, as they will be close to the refresh rate limit but with potential for dips. The Medium preset, at 77 FPS, is the recommended choice for 4K. It provides a comfortable margin above 60 FPS, and the measured minimum of 65 FPS ensures that frame drops are rare. The visual step down from High to Medium is noticeable but acceptable for the significant performance gain. The Low preset at 101 FPS is an option for players who prioritize frame rate above all else, but it defeats the purpose of 4K resolution by reducing detail levels significantly.

In summary, the best experience per the measured rows is a balance of visual quality and frame rate. For most players, 1440p Medium is the recommended sweet spot, as it offers a high average FPS (130) with a solid minimum (111), making it suitable for both 60 Hz and 144 Hz displays. For 4K, Medium is the only preset that consistently exceeds 60 FPS, making it the practical choice. At 1080p, Ultra is the default recommendation due to its high average FPS.

Measured FPS Breakdown

The complete set of measured average frame rates for the Intel Core Ultra 9 275HX and RTX 5080 in Thief is as follows. At 1920x1080, the Low preset delivers an average of 232 FPS, which is the highest frame rate recorded for this combo. The Medium preset averages 190 FPS, with a minimum of 161 FPS and a maximum of 218 FPS. The High preset averages 145 FPS, and the Ultra preset averages 138 FPS. This shows a clear hierarchy, with Low being the fastest and Ultra the slowest, but all settings at 1080p are above 138 FPS, indicating a very playable experience regardless of choice.

At 2560x1440, the average frame rates are lower but still robust. Low settings produce 172 FPS, while Medium settings average 130 FPS, with a minimum of 111 FPS and a maximum of 150 FPS. High settings average 99 FPS, and Ultra settings average 95 FPS. The transition from 1080p to 1440p costs approximately 35-40% of the frame rate at each setting, which is consistent with the increased pixel count. The Medium preset at 1440p is the first point where the minimum FPS drops below 120, but it stays above 100, which is still smooth.

At 3840x2160, the frame rates drop significantly. Low settings average 101 FPS, which is the only 4K setting that exceeds 100 FPS. Medium settings average 77 FPS, with a minimum of 65 FPS and a maximum of 88 FPS. High settings average 58 FPS, and Ultra settings average 56 FPS. This data shows that the RTX 5080 is capable of 4K gaming, but only at lower presets. The High and Ultra presets are borderline for 60 FPS, and the Medium preset is the only one that provides a comfortable margin above 60 FPS. The minimum FPS values for Medium at 4K (65) and 1440p (111) and 1080p (161) demonstrate that the game does not have severe frame-time spikes, which is a positive sign for gameplay stability.

How This Combo Ranks

In the database of tested hardware combinations for Thief, this specific pairing of the Intel Core Ultra 9 275HX and the NVIDIA GeForce RTX 5080 achieves a rank of 584 out of 3720 total combos. This puts the system in the top 15.7% of all configurations tested, which is a strong showing. It is not a top-100 result, but it is well above the median, indicating that this is a high-performance setup for this game. The rank is likely influenced by the fact that the RTX 5080 is a very powerful GPU, but the CPU also contributes to the overall score, as a weaker processor might bottleneck the system in other games.

The GPU’s individual standing is notable. The RTX 5080 has an average benchmark score of 56083, which places it in the 87th percentile of all GPUs. Its nearest rivals are very close in average score: the AMD Radeon 8060S is 0.6% faster, the AMD Radeon RX 6750 GRE 12 GB is 0.7% faster, and the AMD Radeon RX 9070 GRE is 2.2% faster. These are marginal differences, suggesting that in Thief, the RTX 5080 would perform almost identically to these other cards, and the rank would likely shift by only a few positions if the GPU were swapped. The CPU, meanwhile, is in the 94th percentile of all CPUs, with an average benchmark score of 67469. Its nearest rival, the Intel Xeon w5-3525, is only 0.3% faster, and the AMD EPYC 4484PX is 0.5% faster. This means the CPU is not a differentiator in this particular game, as it is far more powerful than required.

The combo rank of 584 out of 3720 is a useful metric for a builder. It shows that this is a very capable system for Thief, capable of high frame rates at 1080p and 1440p, but it also indicates that there are many other configurations that can achieve similar or better results. The data suggests that a user with a high refresh rate 1080p monitor will be extremely satisfied, while a user with a 4K 60 Hz display will need to adjust settings to Medium to get a consistent experience. The rank is a reflection of the hardware’s overall power, not a specific optimization for Thief, which is an older title.

CPU Role

The Intel Core Ultra 9 275HX is a 24-core, 24-thread processor based on the Arrow Lake architecture, with a base clock of 2.70 GHz and a boost clock of 5.40 GHz. It has a 36 MB shared L3 cache and supports DDR5 memory with a dual-channel bus. In synthetic benchmarks, this CPU scores 35589 in Cinebench R23 multi-core and 2204 in single-core, which places it in the 94th percentile of all CPUs. These numbers indicate a very high level of performance, far exceeding the requirements of a 2014 game like Thief.

In the context of Thief’s measured FPS, the CPU’s role is minimal. The frame rates scale almost entirely with resolution and GPU settings, which is the expected behavior when the CPU is not a bottleneck. At 1080p Low, the system produces 232 FPS, which is a very high frame rate that would test many CPUs, but the Core Ultra 9 275HX handles it without issue. The fact that the frame rate drops to 138 FPS at 1080p Ultra, a 40% reduction, is due to the GPU’s increased workload, not any CPU limitation. The CPU’s high single-thread performance (2204 in Cinebench R23) ensures that the game’s main thread, which is often a limiting factor in older titles, is never starved.

The data shows that the CPU has a significant performance headroom. Even at 4K Low, where the GPU is working harder, the frame rate is 101 FPS, which is limited by the GPU’s ability to render the scene, not the CPU’s ability to issue draw calls. The 24 cores and 24 threads are largely unused by Thief, which is a game from an era when dual-core and quad-core processors were common. The high boost clock of 5.40 GHz is more relevant for gaming than the core count, as it helps with the single-threaded performance that many older games rely on. In this case, the CPU is a non-factor in the final FPS numbers, and the RTX 5080 is the sole determinant of performance. This is a positive result for builders, as it means the CPU will not need an upgrade for this game, and it also indicates that the platform has plenty of headroom for more modern, multi-threaded titles.