Thief

# FAQ

Q: What is the combo rank for the AMD Ryzen 9 7900X with the NVIDIA GeForce RTX 5050 in Thief?

A: This combo ranks 1710 out of 2851 tested combinations in the database, placing it in the lower-middle portion of all tested hardware pairings for this game.

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. Its closest rival, the AMD Radeon RX Vega M GL, scores 21153, which is 0.6% higher. The AMD Radeon HD 8970M is 1% higher at 21237, while the AMD Radeon RX 5600 XT trails by 1.6% at 20713.

Q: What is the best measured FPS result for this combo at 1080p?

A: At 1920x1080 with Low settings, the combo achieves 164 average FPS, which is the highest measured result across all resolutions and settings in the dataset.

Q: Does the RTX 5050 have enough VRAM for Thief at 4K?

A: The RTX 5050 features 8 GB of GDDR6 memory on a 128-bit bus with 320.0 GB/s bandwidth. At 3840x2160 on Ultra settings, the combo averages 36 FPS, while High settings yield 38 FPS, indicating the GPU is heavily taxed at this resolution regardless of VRAM capacity.

Q: What is the CPU's percentile ranking among all CPUs?

A: The AMD Ryzen 9 7900X sits in the 91st percentile of all CPUs by average benchmark score, with an average score of 53288.

Q: How does the Ryzen 9 7900X compare to its nearest CPU rivals?

A: The AMD EPYC 7313P scores 53206, just 0.2% lower. The Intel Xeon Phi 7290 is 0.3% higher at 53469, and the Intel Core i7-14700F is 0.6% higher at 53620. The Intel Xeon 634 trails by 0.6% at 52974.

---

How This Combo Ranks

The combination of the AMD Ryzen 9 7900X and NVIDIA GeForce RTX 5050 lands at position 1710 out of 2851 tested combos for Thief. This places the pairing in the 60th percentile of all tested configurations — not a top-tier result, but comfortably above the median. The rank reflects a system where the CPU is exceptionally strong relative to the GPU, creating a bottleneck that limits overall performance in this game.

The Ryzen 9 7900X is a 12-core, 24-thread processor from AMD's 7000 series, built on the Zen 4 architecture (Raphael) using a 5 nm process at TSMC. Its multi-threaded capabilities are substantial: Cinebench R23 multi-core score of 29300, Geekbench multi-core of 19267, and PassMark multi-thread of 51406. These numbers place the CPU in the 91st percentile of all CPUs, with an average benchmark score of 53288.

The RTX 5050, by contrast, sits in the 66th percentile of all GPUs with an average score of 21035. The GPU's nearest rivals include the AMD Radeon RX Vega M GL (21153, 0.6% higher), AMD Radeon HD 8970M (21237, 1% higher), and AMD Radeon RX 5600 XT (20713, 1.6% lower). The delta between CPU and GPU percentile rankings — 91 versus 66 — is stark and directly explains why the combo rank (1710 of 2851) is lower than the CPU's individual standing would suggest.

In practical terms, this rank means the system will deliver playable frame rates at moderate settings, but the GPU holds back what the CPU could otherwise drive. The data indicates that pairing this CPU with a stronger GPU would yield substantially higher FPS in Thief, as the processor has ample headroom in every measured scenario.

---

GPU Role

The NVIDIA GeForce RTX 5050 is the primary determinant of frame rates in Thief for this combo. Its specifications — 2560 shading units, 80 TMUs, 32 ROPs, 20 RT cores, and 80 tensor cores — combine to produce 13.17 TFLOPS of FP32 performance. The GPU runs at a base clock of 2317 MHz and boosts to 2572 MHz, with memory clocked at 2500 MHz (20 Gbps effective).

Memory configuration matters here: 8 GB of GDDR6 on a 128-bit bus delivers 320.0 GB/s of bandwidth. In Thief, a 2014 stealth title, this bandwidth is sufficient for 1080p and 1440p gaming, but at 4K the GPU's overall compute and memory throughput become limiting factors.

The GPU's benchmark results corroborate its mid-range positioning. PassMark G3D score is 17326, while Geekbench OpenCL and Vulkan scores are 90334 and 89381 respectively. The 3DMark Steel Nomad DX12 score of 2502 further confirms its class. These numbers place the GPU at the 66th percentile, with an average benchmark score of 21035.

At 3840x2160, the RTX 5050 delivers 38 FPS on High settings, 50 FPS on Medium, and 66 FPS on Low. The gap between Low and High at 4K is 28 FPS, showing how sensitive this GPU is to settings changes at high resolution. The Ultra preset drops to 36 FPS — barely above the 30 FPS threshold for playability.

The GPU's role is clear: it is the limiting component in nearly every measured scenario. The CPU has far more compute headroom than the GPU can utilize, as evidenced by the CPU's 91st percentile ranking versus the GPU's 66th. Thief's engine, while not modern, still demands GPU throughput that the RTX 5050 can only provide at lower resolutions and settings.

---

Settings Recommendations

Based on the measured FPS data, the optimal settings depend heavily on the target resolution and desired frame rate.

At 1920x1080, all settings presets deliver playable results. Low settings yield 164 FPS, Medium produces 124 FPS with a minimum of 106 FPS, High achieves 95 FPS, and Ultra drops to 90 FPS. For players with 60 Hz or 120 Hz displays, High or Ultra settings provide the best visual quality while maintaining well above 60 FPS. Even Ultra, at 90 FPS, leaves headroom for fluctuations.

At 2560x1440, the picture changes. Low settings produce 112 FPS, Medium 85 FPS (minimum 72 FPS), High 65 FPS, and Ultra 62 FPS. The recommendation here is Medium or High — both maintain averages above 60 FPS, though High's 65 FPS average leaves less margin. Ultra at 62 FPS is borderline, and any demanding scene could push it below 60.

At 3840x2160, the options narrow considerably. Low settings deliver 66 FPS, Medium 50 FPS (minimum 43 FPS), High 38 FPS, and Ultra 36 FPS. For 4K gaming, Low is the only preset that reliably exceeds 60 FPS. Medium is playable at 50 FPS for those willing to accept lower frame rates, but High and Ultra fall below 40 FPS, which most players would find unsatisfactory for a stealth game requiring precise timing.

The best overall experience for this combo is 1920x1080 on High settings: 95 FPS average with strong visual quality. For 1440p, Medium at 85 FPS offers the best balance. At 4K, Low at 66 FPS is the only sensible choice for smooth gameplay. The data shows diminishing returns when pushing higher settings at higher resolutions — the FPS drops are steep and consistent.

---

Measured FPS Breakdown

1920x1080:

• Low: 164 average FPS
• Medium: 124 average FPS, 106 minimum, 143 maximum
• High: 95 average FPS
• Ultra: 90 average FPS

At 1080p, the RTX 5050 demonstrates its strongest performance. The step from Low to Medium costs 40 FPS (164 to 124), a 24% reduction. Moving from Medium to High costs another 29 FPS (124 to 95), a 23% drop. High to Ultra is a smaller penalty of just 5 FPS (95 to 90), suggesting diminishing returns at the top end.

2560x1440:

• Low: 112 average FPS
• Medium: 85 average FPS, 72 minimum, 98 maximum
• High: 65 average FPS
• Ultra: 62 average FPS

At 1440p, the pattern shifts. Low to Medium costs 27 FPS (112 to 85), a 24% reduction. Medium to High costs 20 FPS (85 to 65), a 24% drop. High to Ultra is minimal at 3 FPS (65 to 62). The consistent percentage drops suggest the GPU is scaling predictably with settings.

3840x2160:

• Low: 66 average FPS
• Medium: 50 average FPS, 43 minimum, 58 maximum
• High: 38 average FPS
• Ultra: 36 average FPS

At 4K, the GPU is clearly overextended. Low to Medium costs 16 FPS (66 to 50), a 24% reduction. Medium to High costs 12 FPS (50 to 38), a 24% drop. High to Ultra costs only 2 FPS (38 to 36). The minimum FPS at 4K Medium (43) is close to the average, indicating frame time consistency but overall low performance.

The data reveals a consistent ~24% FPS penalty when moving from Low to Medium and Medium to High across all resolutions. The High-to-Ultra penalty shrinks at higher resolutions, suggesting that Ultra adds little beyond High in terms of GPU load in Thief.

---

CPU Role

The AMD Ryzen 9 7900X is a 12-core, 24-thread processor with a base clock of 4.70 GHz and boost clock of 5.60 GHz. It features 64 KB L1 cache per core, 1 MB L2 per core, and 64 MB shared L3 cache. Memory support is DDR5 with dual-channel configuration and 83.2 GB/s bandwidth. The CPU supports PCIe Gen 5 with 24 lanes and has integrated Radeon Graphics.

Benchmark results show the CPU's strength: Cinebench R23 multi-core 29300, single-core 2016.5; Geekbench multi-core 19267, single-core 2617; PassMark multi-thread 51406, single-thread 4238. The CPU scores 10972 in 3DMark 16-thread, 2137 in 2-thread, 4151 in 4-thread, 7798 in 8-thread, and 12536 in max-thread tests. Single-thread 3DMark is 1093.

These numbers place the CPU in the 91st percentile of all CPUs, with an average benchmark score of 53288. Its nearest rivals are all within 0.6%: AMD EPYC 7313P (53206, 0.2% lower), Intel Xeon Phi 7290 (53469, 0.3% higher), Intel Xeon 634 (52974, 0.6% lower), and Intel Core i7-14700F (53620, 0.6% higher).

In Thief, the CPU's role is to feed the GPU with draw calls and game logic. Given the game's 2014 release, it does not heavily utilize multi-threading. The 7900X's single-thread performance — Cinebench R15 single-core 323, Cinebench R23 single-core 2016.5, Geekbench single-core 2617 — is more than sufficient. The measured FPS data shows no scenario where the CPU appears to be the limiting factor: FPS scales consistently with GPU settings and resolution changes.

The CPU's massive multi-threading headroom (12 cores, 24 threads) remains largely unused in Thief. The GPU, with its 66th percentile ranking, is the bottleneck in every measured configuration. This is confirmed by the FPS patterns: at any given resolution, reducing settings increases FPS proportionally, which is characteristic of GPU-bound scenarios.

---

Resolution Scaling

The FPS data shows how performance degrades as resolution increases from 1920x1080 to 3840x2160. This scaling reveals which component limits performance and how the system responds to increased pixel counts.

At Low settings, the FPS progression is: 164 at 1080p, 112 at 1440p, 66 at 4K. The drop from 1080p to 1440p is 52 FPS (32%), and from 1440p to 4K is 46 FPS (41%). From 1080p to 4K, the total drop is 98 FPS (60%). This steep decline indicates the GPU is heavily pixel-bound at 4K.

At Medium settings: 124 FPS at 1080p, 85 at 1440p, 50 at 4K. The 1080p-to-1440p drop is 39 FPS (31%), 1440p-to-4K is 35 FPS (41%), and the total 1080p-to-4K drop is 74 FPS (60%). The consistent ~60% reduction from 1080p to 4K across settings suggests the GPU's fill rate and memory bandwidth are the limiting factors.

At High settings: 95 FPS at 1080p, 65 at 1440p, 38 at 4K. The drops are 30 FPS (32%) and 27 FPS (42%), totaling 57 FPS (60%). At Ultra settings: 90 FPS at 1080p, 62 at 1440p, 36 at 4K. Drops of 28 FPS (31%) and 26 FPS (42%), totaling 54 FPS (60%).

The pattern is remarkably consistent: regardless of settings preset, the 1080p-to-4K FPS reduction is approximately 60%. This uniformity indicates a GPU-bound scenario where the limiting factor is the RTX 5050's raw throughput, not memory capacity or CPU capability.

The 1080p-to-1440p drop of roughly 31-32% is slightly less than the 1440p-to-4K drop of 41-42%, reflecting the quadratic increase in pixel count: 1440p has 1.78 times the pixels of 1080p, while 4K has 2.25 times the pixels of 1440p. The RTX 5050's 8 GB VRAM and 320.0 GB/s bandwidth cannot keep pace with the demands of 4K rendering.

This scaling analysis confirms the CPU is not the bottleneck. If the Ryzen 9 7900X were limiting, we would expect FPS to remain relatively flat across resolutions at lower settings, where CPU overhead dominates. Instead, FPS drops consistently with pixel count, which is the signature of a GPU-limited system. The 12-core, 24-thread processor has ample headroom; the RTX 5050 is the constraining component in every measured configuration.