Thief

The Intel Core Ultra 5 235 is a 14-core, 14-thread desktop processor based on the Arrow Lake architecture, running at a base clock of 3.40 GHz and a boost clock of 5.00 GHz. This CPU sits in the 89th percentile of all processors, with an average benchmark score of 46062, placing it just 0.1% above the AMD Ryzen AI 9 HX 375 and 0.1% below the Intel Core i9-13900HX. The chip’s multi-threaded performance is substantial, as indicated by a Cinebench R23 multi-core score of 14769 and a single-core score of 2085. For a game like Thief, which launched in 2014, this modern processor provides far more computational headroom than the title requires, meaning the CPU is unlikely to be the primary constraint on frame rates in most scenarios. The core count and high boost clock are overkill for a game of this era, but the data shows that at lower resolutions and settings, the processor’s speed still matters because the frame rates are high enough to expose any per-thread overhead. The 24 MB of shared L3 cache and dual-channel DDR5 memory support with 102.4 GB/s of bandwidth are more than adequate for Thief’s data demands, and the 65 W TDP indicates a power-efficient part that runs cool under load. Benchmark results from PassMark show a multi-thread score of 37816 and a single-thread score of 4516, reinforcing that this is a top-tier CPU for both threaded and lightly-threaded workloads.

The AMD Radeon RX 580 is an older graphics card, based on the Polaris 20 chip with GCN 4.0 architecture, manufactured on a 14 nm process. It features 8 GB of GDDR5 memory on a 256-bit bus, providing 256.0 GB/s of bandwidth, with a base clock of 1257 MHz and a boost clock of 1340 MHz. The GPU has 2304 shading units, 144 texture mapping units, and 32 render output units, delivering a pixel rate of 42.88 GPixel/s and a texture rate of 193.0 GTexel/s. Its FP32 performance is 6.175 TFLOPS, and it holds a 53rd percentile ranking among all GPUs, with an average benchmark score of 12928, which is 0.1% behind the NVIDIA GeForce RTX 3050 Ti Mobile and 0.4% behind the NVIDIA GeForce GTX 1660 SUPER. For Thief, the RX 580’s 8 GB VRAM is sufficient even at 4K, but the card’s raw compute power is the limiting factor. The data shows that the GPU’s performance drops off steeply as resolution increases, which is expected for a card of this vintage. The DirectX 11 benchmark score of 60 and DirectX 12 score of 43 from PassMark indicate that the card handles older APIs reasonably well, which aligns with Thief’s engine requirements. The 185 W TDP and the 1x 8-pin power connector are typical for this class of card, and the PCIe 3.0 x16 interface is not a bottleneck for this game’s moderate data transfer needs.

FAQ

Q: How does the Intel Core Ultra 5 235 compare to its closest rival, the AMD Ryzen AI 9 HX 375?

A: The Intel Core Ultra 5 235 has an average benchmark score of 46062, which is 0.1% higher than the AMD Ryzen AI 9 HX 375’s score of 46030. This is a negligible difference, meaning the two processors perform nearly identically in synthetic benchmarks, and any real-world gaming difference would be imperceptible.

Q: Is the AMD Radeon RX 580’s 8 GB VRAM enough for Thief at 4K resolution?

A: Yes, the 8 GB VRAM is not the limiting factor in this configuration. The measured FPS at 3840x2160 with High settings is 19 FPS, and at Low settings it is 33 FPS. The frame rate drop is due to the GPU’s compute limitations, not memory capacity, as the card’s memory bandwidth of 256.0 GB/s is adequate for this game’s textures.

Q: What is the performance gap between the RX 580 and its nearest rival, the NVIDIA GeForce GTX 1660 SUPER?

A: The RX 580 has an average benchmark score of 12928, which is 0.4% lower than the GeForce GTX 1660 SUPER’s score of 12986. This indicates that the GTX 1660 SUPER is slightly faster on average, but the difference is small enough that in-game performance for Thief would be very similar.

Q: How does the CPU’s single-thread performance affect Thief’s frame rates?

A: The CPU’s single-thread score of 4516 in PassMark and 2085 in Cinebench R23 is exceptionally high. Since Thief is a 2014 title that may not fully utilize multiple cores, this strong single-thread performance ensures that the CPU does not bottleneck the GPU, especially at lower resolutions where frame rates are higher.

Q: At which resolution does the AMD Radeon RX 580 become the clear bottleneck in this system?

A: The data shows a clear trend: at 1920x1080 with Low settings, the system achieves 83 FPS, indicating the CPU is still feeding the GPU well. However, at 3840x2160 with Ultra settings, the frame rate drops to 18 FPS. The GPU is the primary bottleneck at 4K, as the CPU’s performance is far beyond what is needed for this game.

Q: What is the overall ranking of this CPU-GPU combo in Thief?

A: This combination of the Intel Core Ultra 5 235 and AMD Radeon RX 580 ranks 2617 out of 2851 tested combos in Thief. This places it in the lower third of all tested systems, highlighting that the GPU’s age holds back the modern CPU’s potential in this specific game.

How This Combo Ranks

The combination of the Intel Core Ultra 5 235 and the AMD Radeon RX 580 ranks 2617 out of 2851 tested combos in Thief. This is a low ranking, putting the system in the bottom 8% of all tested configurations. The primary reason for this poor showing is the GPU, which ranks in the 53rd percentile of all graphics cards, while the CPU ranks in the 89th percentile. The CPU is a modern, high-performance part that is held back by the older GPU, which lacks the raw throughput needed for higher resolutions. The data indicates that this pairing is heavily unbalanced for Thief, as the CPU’s capabilities are largely wasted because the GPU cannot keep up. In practical terms, this means that users with this combo will see frame rates dictated almost entirely by the RX 580’s performance, and any upgrade to the CPU would yield no tangible benefit in this game. The ranking reflects the GPU’s position relative to other tested cards, and the fact that the CPU is over five years newer does not compensate for the GPU’s age in this specific workload.

Resolution Scaling

The measured FPS data shows a steep drop as resolution increases, which is a clear indicator of GPU-bound performance. At 1920x1080 with Low settings, the system achieves 83 FPS, which drops to 57 FPS at 2560x1440 and further to 33 FPS at 3840x2160. This represents a 31% drop from 1080p to 1440p and a 60% drop from 1080p to 4K. The pattern is consistent across all settings. For High settings, the frame rates are 48 FPS at 1080p, 33 FPS at 1440p, and 19 FPS at 4K, showing a similar proportional decline. This scaling behavior indicates that the GPU is the limiting component at every resolution, as the CPU’s performance is more than sufficient to handle the game’s logic at even the highest frame rates. The fact that the frame rate drops by roughly half when going from 1080p to 4K on Low settings (83 to 33 FPS) suggests that the GPU is being saturated by the pixel fill rate and shading demands. If the CPU were the bottleneck, we would expect to see frame rates stay relatively flat across resolutions, but the data clearly shows a strong inverse relationship between resolution and performance. This confirms that the RX 580 is the primary constraint, and users should focus on lowering resolution or settings to improve playability.

Measured FPS Breakdown

At 1920x1080, the system delivers playable frame rates across all settings. With Low settings, the average FPS is 83, providing a smooth experience. Medium settings yield 63 FPS on average, with a minimum of 54 FPS and a maximum of 72 FPS, which is still quite fluid. High settings produce 48 FPS, and Ultra settings drop to 46 FPS, both of which are borderline playable for a stealth title where reaction time is less critical. Moving to 2560x1440, the performance degrades noticeably. Low settings average 57 FPS, which is acceptable, but Medium settings drop to 43 FPS (min 37, max 50), entering a range where frame pacing may be noticeable. High settings at 1440p average 33 FPS, and Ultra settings average 31 FPS, which are below the threshold for smooth gameplay. At 3840x2160, the system struggles significantly. Low settings average 33 FPS, Medium settings average 25 FPS (min 22, max 29), High settings average 19 FPS, and Ultra settings average 18 FPS. None of these 4K results are playable for a smooth experience, as even the lowest settings cannot maintain 60 FPS, and the frame rates are consistently below 30 FPS. The data shows that the card’s performance drops by more than half when moving from 1080p Low to 4K Low, illustrating the GPU’s limitations at high pixel counts.

Settings Recommendations

Based on the measured FPS data, the optimal settings for this combo depend on the target resolution. At 1920x1080, the Medium preset is the best balance, delivering 63 FPS on average with a minimum of 54 FPS, ensuring a mostly smooth experience without sacrificing too much visual quality. The High preset at 48 FPS is also viable for players who prefer better visuals, but the lower minimum frame rates may cause occasional stutter. Ultra at 46 FPS is not recommended, as the performance gain from High to Ultra is minimal while the frame rate drops further. At 2560x1440, the Low preset is the only setting that provides a consistently playable experience, with an average of 57 FPS. The Medium preset at 43 FPS, with a minimum of 37 FPS, is a compromise, but it may dip into uncomfortable territory during busy scenes. High and Ultra at 1440p, with averages of 33 and 31 FPS respectively, are not recommended for gameplay. At 3840x2160, no settings provide a playable experience, as even Low averages only 33 FPS, and the frame rates are too low for a smooth game. For users who must play at 4K, the Low preset is the only option, but it will require a significant tolerance for low frame rates. The data strongly suggests that the best experience is achieved at 1080p with Medium settings, which offers the highest frame rate while maintaining a reasonable level of graphical fidelity.