Thief

The Intel Core Ultra 9 290K Plus paired with the NVIDIA GeForce RTX 5050 delivers a playable, though not flawless, experience in Thief. The data shows a system that is heavily GPU-limited at higher resolutions, while the processor’s substantial lead in synthetic benchmarks does not translate to dominance in this 2014 stealth title. At 1080p, the combination produces excellent frame rates, but 4K Ultra demands compromises.

FAQ

*Q: What is the best resolution for this CPU/GPU combo in Thief?*

A: Based on the measured FPS, 2560x1440 with Medium settings offers the best balance, delivering an average of 85 FPS. This is significantly smoother than the 38 FPS average at 4K High and still maintains a high level of visual fidelity.

Q: How does the RTX 5050 perform at 1080p with Ultra settings?

A: The RTX 5050 achieves an average of 90 FPS at 1920x1080 with Ultra settings. This is a solid result, indicating that the GPU can handle the game's maximum preset at the most common resolution.

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

A: The data indicates the GPU is the primary bottleneck. For example, at 4K, the average FPS drops to 36 on Ultra, while at 1080p on Low, the average is 164 FPS. This wide variance tied to resolution points to the graphics card struggling at higher pixel counts.

Q: What is the performance difference between Low and Ultra settings at 1440p?

A: At 2560x1440, the average frame rate drops from 112 FPS on Low to 62 FPS on Ultra. This represents a significant 45% performance cost for the high-fidelity preset.

*Q: Does the CPU's 24-core configuration provide any advantage in Thief?*

A: While the Intel Core Ultra 9 290K Plus has 24 cores and 24 threads, its benchmark scores are impressive, the game's performance scaling is tied to the GPU. The CPU's strength is evident in its 96th percentile ranking, but it does not prevent the frame rate from dropping at higher resolutions, which is a GPU workload.

Q: How does the RTX 5050 compare to its direct rivals in synthetic benchmarks?

A: The GPU’s average benchmark score of 21035 is nearly identical to its closest rivals. It is 1.6% ahead of the AMD Radeon RX 5600 XT and 0.6% behind the AMD Radeon RX Vega M GL, indicating performance parity in this tier.

GPU Role

The NVIDIA GeForce RTX 5050 is the decisive component in this configuration's Thief performance. The data shows a clear correlation between the graphics card's workload and the frame rates. The GPU features 8 GB of GDDR6 memory on a 128-bit bus, providing 320.0 GB/s of bandwidth. This memory configuration is sufficient for 1080p and 1440p gameplay, but the 4K results suggest it is a limiting factor.

The GPU's clock speeds are set to a base of 2317 MHz and a boost of 2572 MHz. These clocks, combined with 2560 shading units, produce a peak FP32 throughput of 13.17 TFLOPS. This raw compute power is what drives the game's rendering. The benchmark results show that at 1080p, the GPU has ample headroom, producing an average of 164 FPS on Low settings. This indicates that the GPU is not fully stressed at this resolution and setting.

However, as the resolution increases, the GPU's limitations become clear. At 4K Ultra, the average FPS plummets to 36. This is a 78% drop from the 1080p Low result, demonstrating that the RTX 5050's rendering capabilities are strained by the pixel count and shader complexity of 4K. The GPU's 66th percentile ranking among all GPUs and its proximity to rivals like the AMD Radeon RX 5600 XT in synthetic tests confirm its position as a mid-range performer. The 3DMark Steel Nomad DX12 score of 2502 is a modern metric, but in Thief, the older engine's demands are what we see reflected in the measured FPS.

The RTX 5050’s 32 ROPs are a likely constraint at higher resolutions, as the fill-rate demands of 4K can overwhelm this unit count. The performance difference between 1080p and 4K at the same Ultra preset is stark: 90 FPS versus 36 FPS. This 60% reduction is a classic sign of a GPU that is reaching its limits, as the CPU has already proven its capability to feed high frame rates at lower resolutions.

CPU Role

The Intel Core Ultra 9 290K Plus is a high-end processor that provides more than enough computational power for Thief. It has 24 cores and 24 threads, with a base clock of 3.70 GHz and a boost clock of 5.80 GHz. This configuration is far beyond what a 2014 stealth game requires. The CPU’s benchmark scores are exceptional, with a 96th percentile ranking among all CPUs and an average benchmark score of 84003.

The data shows that the CPU is not the performance bottleneck in this pairing. At 1080p with Low settings, the system achieves 164 FPS. If the CPU were the limiting factor, we would expect to see a cap on frame rates regardless of resolution or settings. Instead, we see a massive variation in FPS based on GPU load, which confirms the processor is keeping pace.

The CPU’s synthetic performance is head and shoulders above its rivals. It is 0.2% ahead of the Intel Core Ultra 9 285K, 1.1% ahead of the AMD EPYC 4584PX, and 1.2% ahead of the AMD EPYC 9135. In single-threaded tasks like Cinebench R23, it scores 7303, which is critical for gaming. However, this strength does not translate into a proportional advantage in Thief because the game's frame rate is largely dictated by the GPU. The CPU's role is to handle the game's logic and physics, which it does with ease, leaving the RTX 5050 as the sole arbiter of final performance.

The processor's 36 MB of shared L3 cache and dual-channel DDR5 memory support with 115.2 GB/s bandwidth ensure that data is fed to the GPU efficiently. The CPU's high boost clock of 5.80 GHz is more than sufficient for the game's single-threaded tasks. In essence, the Core Ultra 9 290K Plus ensures that the RTX 5050 is never waiting for data, which allows the GPU to work at its full potential, as seen in the 1080p results.

Measured FPS Breakdown

The measured FPS data provides a complete picture of the RTX 5050's performance across three resolutions and four settings presets in Thief.

At 1920x1080, the results are strong across the board. On Low settings, the average FPS is 164. On Medium, it drops to 124, with a minimum of 106 and a maximum of 143. High settings yield an average of 95 FPS. The Ultra preset produces an average of 90 FPS. This shows a smooth scaling curve, with each settings tier costing a predictable amount of performance.

Moving to 2560x1440, the performance takes a noticeable hit. The Low preset average is 112 FPS. The Medium preset produces an average of 85 FPS, with a minimum of 72 and a maximum of 98. High settings average 65 FPS. The Ultra preset brings the average down to 62 FPS. The gap between Low and Ultra at 1440p is 50 FPS, which is a significant difference.

At 3840x2160, the RTX 5050 begins to struggle. The Low preset manages an average of 66 FPS. The Medium preset averages 50 FPS, with a minimum of 43 and a maximum of 58. High settings drop to an average of 38 FPS. The Ultra preset is the most demanding, producing an average of only 36 FPS. This is the only resolution where even the Low preset cannot maintain a 60 FPS average. The data shows that 4K is only playable at Low settings, and even then, it is borderline.

The frame time consistency also degrades at higher resolutions. At 1440p Medium, the difference between the min and max FPS is 26 FPS. At 4K Medium, this difference is 15 FPS. This suggests that while the average FPS drops, the frame pacing may be more consistent at lower frame rates, but the overall experience is less fluid.

Resolution Scaling

The performance scaling from 1080p to 4K reveals a system that is clearly GPU-bound. When comparing the Ultra preset across all resolutions, we see a clear and predictable decline. At 1080p, the average FPS is 90. At 1440p, it drops to 62, a 31% reduction. At 4K, it falls to 36, a 60% reduction from the 1080p baseline.

This pattern is consistent across all settings. For the High preset, the FPS drops from 95 at 1080p to 65 at 1440p and then to 38 at 4K. This is a clear indication that the RTX 5050’s rendering throughput is being consumed by the increasing pixel count. The GPU's 8 GB of VRAM is also a likely factor, as 4K textures and buffers can exceed this capacity, leading to performance drops.

The CPU's role in this scaling is minimal. If the CPU were the bottleneck, the FPS would remain relatively flat across resolutions, as the CPU workload does not change significantly with resolution. Instead, the FPS drops sharply, confirming that the GPU is the limiting component. The data suggests that the Core Ultra 9 290K Plus has significant headroom, as it is able to push the RTX 5050 to its limits at 1080p, but the GPU cannot maintain that performance as the resolution increases.

The scaling from 1440p to 4K is particularly steep. For example, on Medium settings, the FPS drops from 85 to 50, a 41% decrease. This is a larger relative drop than from 1080p to 1440p, where the FPS drops from 124 to 85, a 31% decrease. This indicates that the GPU's memory bandwidth and fill rate are being overwhelmed at 4K, making it an unsuitable resolution for high-refresh-rate gameplay with this card.

Settings Recommendations

Based on the measured data, the optimal experience for this hardware combo in Thief is found at 2560x1440 with Medium settings. This configuration produces an average FPS of 85, which is well above the 60 FPS threshold for smooth gameplay. The minimum FPS of 72 ensures that there are no major stutters, and the maximum of 98 FPS provides a good margin. This preset offers a significant visual improvement over Low settings while maintaining a high frame rate.

For users with 1080p displays, the Ultra preset is the best choice. The average FPS of 90 is excellent, and this allows players to experience the game with all visual features enabled. The high settings preset at 1080p also delivers a very smooth 95 FPS average, making it a viable alternative if the small performance cost of Ultra is a concern.

At 4K, the options are limited. The Low preset is the only one that provides a playable average FPS of 66. However, this requires a significant visual compromise. The Medium preset at 4K, with an average of 50 FPS, may be acceptable for players who prioritize visual quality over a locked 60 FPS, but it is not ideal.

The data suggests that players should avoid the Ultra preset at 1440p and 4K. At 1440p, Ultra yields 62 FPS, which is playable but not optimal. At 4K, Ultra yields 36 FPS, which is below the playable threshold for most gamers. The High preset at 1440p offers an average of 65 FPS and may be a better choice than Ultra if the player wants a higher frame rate with a minor reduction in visual detail. In summary, the best experience is 1440p Medium, with 1080p Ultra as the best high-fidelity option.