Thief

# Thief with AMD Ryzen 5 5600GT + NVIDIA GeForce RTX 5050

The pairing of AMD's Ryzen 5 5600GT and NVIDIA's GeForce RTX 5050 produces a distinctly GPU-bound experience in Thief, a 2014 stealth title that predates the modern GPU era. With the CPU's 6 cores and 12 threads running at a 3.60 GHz base and 4.60 GHz boost clock, the processor is rarely the bottleneck — benchmark results across resolutions show FPS scaling that tracks almost perfectly with pixel count, which is the classic signature of a graphics-limited workload. The RTX 5050, built on Blackwell 2.0 architecture with 8 GB of GDDR6 memory on a 128-bit bus, delivers playable frame rates at 1080p and 1440p, but the data reveals a sharp drop-off at 4K that defines the practical ceiling for this combination.

Resolution Scaling

The measured FPS data shows a consistent pattern: as resolution increases from 1920x1080 to 3840x2160, average frame rates fall by roughly 55-60% across all settings presets, which is a textbook indication that the GeForce RTX 5050 is the limiting component. At High settings, the average FPS drops from 95 at 1080p to 65 at 1440p and finally to 38 at 4K — a reduction of 31.6% going from 1080p to 1440p, then another 41.5% going from 1440p to 4K. This progressive decline is steeper than what a CPU-bound scenario would produce, because the Ryzen 5 5600GT's single-thread performance (3dmark_single_thread score of 913) and multi-thread capability (cinebench_r23_multicore score of 17272) are more than sufficient to feed the GPU at lower resolutions.

The Low settings row provides the clearest view of the scaling behavior. At 1080p Low, the combo achieves 164 average FPS, which drops to 112 at 1440p and 66 at 4K. The fact that 4K Low still delivers 66 FPS while 4K Ultra only manages 36 FPS indicates that the RTX 5050's raw throughput — 13.17 TFLOPS FP32 and a pixel rate of 82.30 GPixel/s — is being strained by the higher pixel counts, not by any CPU limitation. If the processor were the bottleneck, we would expect to see the 1080p and 1440p numbers cluster closer together, since the CPU's work per frame doesn't scale with resolution.

The Medium settings row includes min and max FPS values, which add another layer to the analysis. At 4K Medium, the average is 50 FPS with a min of 43 and max of 58 — a relatively tight 15 FPS spread that suggests consistent frame pacing under GPU load. At 1440p Medium, the average rises to 85 FPS with a min of 72 and max of 98, while 1080p Medium shows 124 average with 106 min and 143 max. The widening range at lower resolutions hints that the CPU's frame submission times are becoming more variable as the GPU finishes frames faster, but the overall scaling still points to the GPU as the primary constraint.

What this scaling implies for gameplay is straightforward: the RTX 5050 is best utilized at 1080p or 1440p, where it can maintain high frame rates even at Ultra settings. At 4K, the user must accept either a significant settings reduction or a sub-60 FPS experience, as only the Low preset crosses the 60 FPS threshold at that resolution.

FAQ

Q: What is the highest average FPS this combo achieves in Thief?

A: The maximum measured average FPS is 164, achieved at 1920x1080 with Low settings. This is the only configuration in the data that exceeds 160 FPS.

Q: Can this combo run Thief at 4K with playable frame rates?

A: Yes, but only at Low settings, where it averages 66 FPS. At Medium settings, the average drops to 50 FPS (min 43, max 58), while High and Ultra average 38 and 36 FPS respectively, which falls below the commonly accepted 60 FPS playability threshold.

Q: How does the RTX 5050 compare to its nearest GPU rivals in average benchmark score?

A: The RTX 5050 has an average benchmark score of 21035, placing it 1.6% above the AMD Radeon RX 5600 XT (score 20713) and 0.6% below the AMD Radeon RX Vega M GL (score 21153). It also sits 1% behind the AMD Radeon HD 8970M (score 21237) and 1.6% behind the NVIDIA RTX A4000 Mobile (score 21379).

Q: What is the CPU's multi-threaded performance relative to its nearest rivals?

A: The Ryzen 5 5600GT's average benchmark score is 20733, which is 0.3% below the Intel Core i5-12490F (score 20802), 0.4% below the Intel Xeon 6325P (score 20821), 0.4% below the Intel Core Ultra 5 125U (score 20826), and 0.5% below the AMD EPYC 7J13 (score 20845). This places the 5600GT in the 74th percentile of all CPUs.

Q: Does the game favor the CPU or GPU in this pairing?

A: The resolution scaling data strongly indicates a GPU-bound workload. The FPS drops significantly as resolution increases across all settings, while the CPU's substantial multi-thread capability (cinebench_r23_multicore score of 17272) and 6-core/12-thread configuration are never the primary constraint.

Q: What is the best settings preset for 1440p gameplay?

A: At 2560x1440, the Ultra preset averages 62 FPS, while High averages 65 FPS. The Medium preset provides a more comfortable margin at 85 FPS (min 72, max 98), making it the most balanced choice for sustained performance without sacrificing too much visual quality.

Settings Recommendations

Examining the measured FPS rows, the data suggests that the optimal balance between visual fidelity and performance depends heavily on the target resolution. At 1920x1080, all four settings presets deliver playable frame rates — Low averages 164 FPS, Medium 124 FPS, High 95 FPS, and Ultra 90 FPS. The difference between High and Ultra is only 5 FPS (95 vs 90), which makes Ultra the more attractive choice for 1080p users who want maximum quality without a meaningful performance penalty. The 90 FPS average at Ultra is still well above the 60 FPS threshold, and the GPU's 8 GB VRAM is unlikely to be a limitation at this resolution.

For 2560x1440, the picture changes. Ultra averages 62 FPS, which is technically playable but leaves little headroom for demanding scenes. High averages 65 FPS, a marginal improvement, while Medium jumps to 85 FPS with a min of 72 FPS. The data indicates that Medium is the sweet spot at 1440p — it provides a 27% performance improvement over High (85 vs 65 FPS) while presumably retaining a reasonable level of visual quality. The min FPS of 72 at Medium confirms that even in the most demanding moments, the frame rate stays comfortably above 60 FPS.

At 3840x2160, the recommendations become more constrained. Low is the only preset that averages above 60 FPS, at 66 FPS. Medium averages 50 FPS with a min of 43 FPS, which may be acceptable for slower-paced stealth gameplay but risks noticeable stuttering in action sequences. High and Ultra average 38 and 36 FPS respectively, which falls into the range where the game may feel less responsive. For 4K users, the data suggests accepting Low settings or dropping to 1440p for a better experience.

The overall pattern from the measured rows is that the RTX 5050 has sufficient headroom at 1080p to run Ultra settings, adequate performance at 1440p with Medium or High, and only Low settings at 4K. Users prioritizing frame rate over visual quality should consider the Medium preset at 1440p as the best overall compromise, as it delivers the highest average FPS above 60 with a tight min-max range.

How This Combo Ranks

In the database of tested CPU-GPU combinations for Thief, this pairing of the AMD Ryzen 5 5600GT and NVIDIA GeForce RTX 5050 holds a combo rank of 2258 out of 3720 total tested combos. This places the configuration in the lower half of all tested systems, which is somewhat surprising given the individual component specifications. The rank of 2258 out of 3720 means that approximately 60.7% of tested combos perform better in this specific game, while 39.3% perform worse.

This ranking is likely influenced by the game's age — Thief was released in 2014, and its engine may not scale efficiently with newer hardware architectures. The RTX 5050's Blackwell 2.0 architecture, with its 2560 shading units and 80 tensor cores, is designed for modern DirectX 12 Ultimate workloads, but Thief's rendering pipeline may not fully utilize these features. The GPU's percentileVsAllGpus of 66 indicates it outperforms 66% of all GPUs in general benchmark suites, yet the combo's rank in this specific game suggests that raw compute power doesn't directly translate to in-game performance for this title.

The CPU's percentileVsAllCpus of 74 places it above 74% of all CPUs, and the 5600GT's nearest rivals in average benchmark score — the Intel Core i5-12490F, Intel Xeon 6325P, Intel Core Ultra 5 125U, and AMD EPYC 7J13 — all have virtually identical scores within a 0.5% range. This indicates that the CPU is not a weak link in the pairing, yet the combo rank of 2258 suggests that the GPU's performance in Thief is the limiting factor. The gap between the GPU's 66th percentile ranking and the combo's 60.7th percentile ranking (worse than 60.7% of combos) highlights how game-specific optimization can vary significantly from synthetic benchmarks.

Notably, this rank is out of 3720 combos, which is a substantial sample size. The fact that the RTX 5050's nearest GPU rivals include the AMD Radeon RX 5600 XT (a previous-generation mid-range card) and the AMD Radeon RX Vega M GL (an integrated solution) suggests that the 5050's performance class is competitive with older mid-range GPUs, which may explain why the combo doesn't rank higher in a 2014 title that predates its architecture.

CPU Role

The AMD Ryzen 5 5600GT is a 6-core, 12-thread processor based on the Zen 3 architecture (Cezanne codename) built on a 7 nm process at TSMC, with a die size of 180 mm² containing 10,700 million transistors. Its base clock of 3.60 GHz and boost clock of 4.60 GHz provide strong single-thread performance, as evidenced by a 3dmark_single_thread score of 913 and a cinebench_r23_singlecore score of 2438. The CPU's 16 MB of L3 cache, along with 64 KB L1 and 512 KB L2 per core, ensures that game data is readily accessible without frequent memory fetches.

In Thief, the CPU's role appears to be secondary to the GPU, based on the resolution scaling data. The fact that 1080p Low achieves 164 FPS while 4K Low only achieves 66 FPS — a drop of nearly 60% — indicates that the CPU is not the bottleneck at lower resolutions. If the CPU were limiting, we would expect the 1080p and 1440p numbers to be closer together, since the CPU's work per frame remains constant regardless of resolution. Instead, the FPS scales inversely with pixel count, which is the hallmark of a GPU-bound scenario.

The CPU's multi-threaded capabilities are substantial for a 65 W TDP part — its cinebench_r23_multicore score of 17272 and passmark_multithread score of 20325 place it in the 74th percentile of all CPUs. In a stealth game like Thief, which likely doesn't heavily utilize more than a few threads, the 6-core/12-thread configuration provides ample headroom for background tasks and physics calculations without impacting frame rates. The passmark_physics score of 875 suggests that the CPU can handle the game's physics simulation without issue.

The memory bandwidth of 51.2 GB/s, delivered through dual-channel DDR4 support, is sufficient for the game's data demands, though notably, the CPU's PCIe Gen 3 interface (16 lanes) is older than the GPU's PCIe 5.0 x8 interface. This mismatch could theoretically limit data transfer speeds, but for a 2014 game with modest texture and geometry requirements, the practical impact is likely negligible. The 5600GT's integrated Radeon Vega 7 graphics are not used in this configuration, but their presence doesn't affect dedicated GPU performance.

Measured FPS Breakdown

At 1920x1080, the measured FPS values show a clear hierarchy across settings. The Low preset delivers the highest average at 164 FPS, followed by Medium at 124 FPS (with a min of 106 and max of 143), High at 95 FPS, and Ultra at 90 FPS. The gap between Low and Medium is 40 FPS, while the gap between Medium and High is 29 FPS, and the gap between High and Ultra is only 5 FPS. This suggests that the Ultra preset offers diminishing returns in visual quality relative to High, while the jump from Low to Medium provides a significant performance cost.

At 2560x1440, the same pattern emerges but with compressed values. Low averages 112 FPS, Medium 85 FPS (min 72, max 98), High 65 FPS, and Ultra 62 FPS. The difference between High and Ultra narrows to just 3 FPS at this resolution, confirming that the Ultra preset is essentially equivalent to High in performance terms. The Medium preset's min FPS of 72 is particularly notable — it's higher than the average FPS of both High and Ultra, making Medium the clear performance winner at 1440p.

At 3840x2160, the values drop significantly. Low averages 66 FPS, Medium 50 FPS (min 43, max 58), High 38 FPS, and Ultra 36 FPS. The gap between Low and Medium is 16 FPS, while the gap between High and Ultra narrows to just 2 FPS. This resolution is clearly the most demanding for the RTX 5050, with only the Low preset maintaining an average above 60 FPS. The Medium preset's min FPS of 43 indicates that the game can dip into the 40s during intense moments, which may be noticeable in a stealth game where smooth camera movement is important.

The complete set of measured FPS data reveals that the RTX 5050's performance scales roughly linearly with pixel count. Going from 1080p to 1440p (a 1.78x increase in pixels) reduces average FPS by about 30-40% across settings, while going from 1440p to 4K (a 2.25x increase) reduces FPS by another 40-50%. This scaling behavior is consistent with a GPU that has sufficient memory bandwidth (320.0 GB/s) but limited compute throughput (13.17 TFLOPS) for high-resolution rendering.

GPU Role

The NVIDIA GeForce RTX 5050 is built on the Blackwell 2.0 architecture with the GB207 chip, manufactured on a 5 nm process at TSMC with 16,900 million transistors on a 149 mm² die. Its base clock of 2317 MHz and boost clock of 2572 MHz, combined with 2560 shading units, 80 TMUs, and 32 ROPs, deliver a peak pixel rate of 82.30 GPixel/s and texture rate of 205.8 GTexel/s. The 8 GB of GDDR6 memory on a 128-bit bus provides 320.0 GB/s of bandwidth, which is adequate for the game's texture loads but may be a constraint at 4K with high-resolution textures.

In Thief, the GPU's performance is the primary determinant of frame rates, as evidenced by the resolution scaling analysis. The RTX 5050's passmark_g3d score of 17326 and 3dmark_steel_nomad_dx12 score of 2502 place it in the 66th percentile of all GPUs, yet its in-game performance at 4K Ultra (36 FPS) shows that even a mid-range modern GPU struggles with this game at maximum settings on a 4K display. The GPU's DirectX 12 Ultimate support (12_2) and Vulkan 1.4 compatibility ensure that the game can run with modern API features, but Thief's 2014 release date means it was likely optimized for older DirectX 11 pipelines.

The VRAM capacity of 8 GB is sufficient for 1080p and 1440p gameplay, as the FPS data shows no abnormal drops that would indicate memory swapping. At 4K, the 8 GB capacity may become a limiting factor if the game loads high-resolution textures, but the measured FPS at 4K Low (66 FPS) suggests that the GPU can still perform adequately when texture quality is reduced. The memory clock of 2500 MHz (20 Gbps effective) and 128-bit bus width produce a bandwidth of 320.0 GB/s, which is a mid-range figure that aligns with the GPU's overall performance class.

The RTX 5050's nearest GPU rivals in average benchmark score provide context for its expected performance. The AMD Radeon RX 5600 XT scores 20713 (1.6% lower), while the AMD Radeon RX Vega M GL scores 21153 (0.6% higher), the AMD Radeon HD 8970M scores 21237 (1% higher), and the NVIDIA RTX A4000 Mobile scores 21379 (1.6% higher). These close scores indicate that the RTX 5050 sits in a competitive mid-range tier, and its in-game performance in Thief is consistent with this positioning — capable at 1080p and 1440p, but not a high-end solution for 4K gaming. The GPU's 20 RT cores and 80 tensor cores are not directly relevant to Thief, as the game predates widespread ray tracing adoption, but they contribute to the GPU's overall compute capabilities.