Thief
This combination delivers exceptional performance with an average of 343 FPS, perfect for high refresh rate gaming and competitive play.
Average FPS: resolution vs quality settings
| Ultra | High | Medium | Low | |
|---|---|---|---|---|
| 4K Ultra HD | 131 | 175 | 206 | 242 |
| 1440p QHD | 238 | 317 | 374 | 439 |
| 1080p Full HD | 364 | 484 | 554 | 586 |
Cell color: green is 120+ FPS, teal is 60+, amber is 30+, red is below 30.
Thief with AMD Ryzen 5 5600G + NVIDIA GeForce RTX 5070 Ti, In-Depth Analysis
The AMD Ryzen 5 5600G paired with the NVIDIA GeForce RTX 5070 Ti achieves the highest measured combination among 3,720 tested for the stealth title Thief (released 2014). The data covers three resolutions and four quality presets, yielding an average FPS of 175 at 4K Ultra, 131 at 4K Ultra with a lower preset? Wait, the numbers show Ultra 131, High 175, Medium 206, Low 242 at 4K. The combo ranks 1st overall, and every preset across all resolutions delivers average FPS above 131—the lowest figure measured. This suggests a very high-performing platform, and the analysis below examines where the bottleneck lies (if any), how the components contribute, and how scaling behaves.
Settings Recommendations
The measured FPS matrix provides a clear ranking of presets by frame rate. At 4K, the extremes are 242 FPS on Low and 131 FPS on Ultra. Medium produces 206 FPS, High produces 175 FPS. At 1440p, the same ordering is preserved: Low 439, High 317, Medium 374, Ultra 238. At 1080p, the spread is 586 Low, 554 Medium, 484 High, and 364 Ultra.
Since every preset at every resolution exceeds 100 FPS, the limiting factor is not raw playability but the trade-off between visual quality and frame overhead. The data suggests that High is the most balanced preset. At 4K, High (175 FPS) is 21% slower than Medium (206) but only 11% slower than Low (242). More importantly, High at 4K still produces over 173 FPS, which is well above common display refresh rates. At 1440p, High yields 317 FPS, the highest of the quality. At 1080p, High produces 484 FPS, which is still more than enough to saturate any typical monitor. The only setting that falls below 200 FPS at 4K is Ultra (131 FPS), which may be considered for users with a 120Hz or 144Hz display who prioritize eye-candy over frame headroom.
The Medium preset at each resolution shows a minimum FPS digit in the data: at 4K Medium has a minimum of 175 FPS, at 1440p Medium has a minimum of 318 FPS, and at 1080p Medium has a minimum of 471 FPS. These minimums are close to the average High values, which suggests that Medium offers some frame stability, while High and Low presets do not have min/max recorded. Overall, the measured results recommend High as the default preset for the most robust balance between framerate and scene complexity.
CPU Role
The AMD Ryzen 5 5600G is a 6-core, 12-thread processor based on the Zen 3 (Cezanne) architecture, with a base clock of 3.90 GHz and a boost clock of 4.40 GHz. It integrates 16 MB of L3 cache, 64KB L1 per core, and 512 KB L2 per core, and has a TDP of 65 W. In synthetic benchmarks, the 5600G scores 5455 in 3DMark max threads, 10551 in Cinebench R23 multi-core, and 872 in single-thread 3dmark. It ranks in the 74th percentile of all CPUs and has an average benchmark score of 21088, placing it within 0.2% of the Intel Core i3-1220P (also 0.2% above the Intel Core Ultra 7 256V, and 0.4% above the Ultra 7 155U). These numbers indicate a competent mid-range processor that is almost identically matched by its nearest rivals.
In Thief, the CPU role is not directly quantified, but the high frame rates at 1080p (up to 586 FPS on Low) suggest that the processor is not the bottleneck at lower resolutions. The 5600G’s single-thread performance is strong, and the massive GPU (RTX 5070 Ti) is well feed at lower resolutions, which is consistent with the 1080p row: Low in 1080p records 586 FPS, which is ~157 FPS higher than at 1440p (the data shows a drop of 147 from 586 to 439), while the 4K row continues to drop. The CPU’s three-level cache and 12 threads are fully sufficient for a 2014 stealth title, which is typically not a multi-threaded load but still benefits from clock speed. The data does not include a direct CPU-limited measurement, but given the rank of 1 among 3720, the combination is clearly not constrained by the processor.
GPU Role
The NVIDIA GeForce RTX 5070 Ti is a Blackwell 2.0 architecture GPU built on a 5 nm process, with 56.6 million transistors and a die size of 378 mm². It operates with a base clock of 2295 MHz and a boost clock of 2452 MHz, and is equipped with 16 GB of GDDR7 memory on a 256-bit bus, giving 896 GB/s memory bandwidth. The GPU has 8960 shader units, 280 TMUs, and 96 ROPs, plus 70 RT cores and 280 tensor cores. In raw compute, it delivers 43.94 TFLOPS FP32 and features 235.4 GPixel/s pixel rate and 686.6 GTexel/s texture rate. It ranks in the 86th percentile among all GPUs, with an average benchmark score of 49957, which is within 0.1% of the AMD Radeon RX Vega 64, 0.4% above the Intel Arc A550M, 2% below the AMD Radeon RX 6900 XT, and 3.1% above the AMD Radeon RX 6800 XT.
The game’s measured FPS results show that the GPU is the main scaling component. At 1080p, even the Ultra preset delivers 364 FPS, while at 4K Ultra drops to 131 FPS—a reduction of 233 FPS. This scaling matches the expected impact of shifting from 2 million pixels to 8.3million; a 3-to-7 resolution factor. The GPU’s high memory bandwidth and 16 GB VRAM place no memory bottleneck at any tested resolution. The observed FPS pattern (1080p → 1440p → 4K) is typical of a GPU-limited scenario after accounting for a moderate CPU overhead. The fact that the lowest FPS at 4K still reaches 131 confirms the RTX 5070 Ti’s authority in driving this title.
FAQ
Q: What is the average frame rate at 4K with the Ultra preset?
A: At 3840×2160 with the Ultra preset, the measured average FPS is 131.
Q: Which settings for 1440p produces the highest minimum FPS?
A: The Medium preset at 2560×1440 has a minimum of 318 FPS, the highest recorded minimum across all tested presets at that resolution.
Q: How does the RTX 5070 Ti’s average benchmark score compare to its nearest rival?
A: The GPU has an average benchmark score of 49957. It is 3.1% above the AMD Radeon RX 6800 XT, 2% below the AMD Radeon RX 6900 XT, and nearly identical to the AMD Radeon RX Vega 64 (with a −0.1% difference) and +0.4% over the Intel Arc A550M.
Q: What is the CPU’s 3DMark max-thread score?
A: The Ryzen 5 5600G scores 5455 in 3dmark max threads, and its single-thread 3dmark score is 872.
Q: Which resolution shows the largest difference between Low and Ultra settings?
A: At 4K, the difference is measured as 111 FPS (242 on Low vs 131 on Ultra). At 1440p, that difference is 201 FPS (439 vs 238), and at 1080p it is 222 FPS (586 vs 364). The largest raw gap occurs at 1080p.
Resolution Scaling
The frame rate data reveals a consistent, expected pattern as resolution increases. For the High preset: at 1920×1080 the average is 484 FPS, at 2560×1440 it is 317, and at 3840×2160 it is 175. The Ultra preset shifts from 364 (1080p) to 238 (1440p) to 131 (4K). Low preset goes from 586 to 439 to 242. Medium preset shows 554 to 374 to 206. The relative reduction from 1080p to 1440p is roughly: for High, 484 → 317 (a drop of 167); for Medium, 554 → 374 (a drop of 180); for Low, 586 → 439 (a drop of 147); for Ultra, 364 → 238 (a drop of 126). Scaling to 4K further reduces these numbers to about 64% of the 1440p value: High 317 → 175, Medium 374 → 206, Low 439 → 242, Ultra 238 → 131.
This scaling shape is classic GPU-bound behavior: each step up in resolution multiplies pixel count, and the FPS drops roughly in proportion to pixel count, though the exact values are influenced by the GPU’s load and cache locality. The fact that even at 4K the lowest settings produce 242 FPS suggests the GPU is rarely the only limiting factor; the CPU can still feed it a healthy framerate at high resolutions. The difference between Low and Ultra at each resolution is always more than 18%, indicating that the GPU’s shading capacity and bandwidth are the dominant variables.
Measured FPS Breakdown
The complete measured data set is as follows:
| Resolution | Settings | Avg FPS | Min FPS | Max FPS |
|------------|----------|-------|---------|---------|
| 3840x2160 | High | 175 | – | – |
| 3840x2160 | Low | 242 | – | – |
| 3840x2160 | Medium| 206 | 175 | 237 |
| 3840x2160 | Ultra | 131 | – | – |
| 2560x1440 | High | 317 | – | – |
| 2560x1440 | Low | 439 | – | – |
| 2560x1440 | Medium| 374 | 318 | 430 |
| 2560x1440 | Ultra | 238 | – | – |
| 1920x1080 | High | 484 | – | – |
| 1920x1080 | Low | 586 | – | – |
| 1920x1080 | Medium| 554 | 471 | 637 |
| 1920x1080 | Ultra | 364 | – | – |
The table shows that for Medium presets, the minimum and maximum FPS are captured, while for Low, High, and Ultra, only the average is recorded. In each resolution, the difference between the highest (Low) and lowest (Ultra) is substantial, with the Ultra setting often halving the frame rate compared to Low. The Mega-sample of 4K shows the most consistent scaling, and the gap between High and Ultra (from 175 to 131) is moderate but still demonstrates the visual cost of ultra-level detail.
How This Combo Ranks
This specific CPU–GPU combination achieves a perfect rank of 1 out of 3,270 in the Thief benchmark ranking, meaning no other tested combination of processor and graphics card produces higher measured FPS across the available preset/resolution matrix. The data confirms that the Ryzen 5 5600G + RTX 5070 Ti leads the entire list, beating every other recorded combo in this title.
The CPU alone sits at the 74th percentile of all processors, with an average benchmark score of 21,088. The GPU ranks 86th percentile of all graphics cards, with an average benchmark score of 49,957. The combined system benefits from a high single-thread and multi-thread processor matching a high-end Blackwell GPU. The fact that it ranks 1st overall indicates that neither component introduces a meaningful coupling loss in Thief; the CPU delivers enough threads and speed to keep the GPU fed at every preset, and the GPU’s raw throughput is the most dominant factor in final FPS.
Such a rank is measured data, not a prediction. It signifies that the measured pairs among all 3,720 combinations, this one delivers the highest average FPS at any setting/resolution tested. The data also recognizes the GPU’s dominant role: the RX 6800 XT, which trails the 5070 Ti by 3.1% in general benchmarks, would likely still produce an inferior FPS due to the large delta. The combination’s top position underlines the synergy of the 5600G’s key features (6-core/12-thread, Gen 3 pcie, high boost clock) with the massive memory bandwidth and shader counts of the 5070 Ti.
Hardware Specifications
AMD Ryzen 5 5600G
NVIDIA GeForce RTX 5070 Ti
FPS Benchmarks by Resolution & Settings
Performance data for AMD Ryzen 5 5600G + NVIDIA GeForce RTX 5070 Ti in Thief
| Resolution | Settings | Avg FPS |
|---|---|---|
| 3840x2160 | Low | 242.0 |
| 3840x2160 | Medium | 206.0 |
| 3840x2160 | High | 175.0 |
| 3840x2160 | Ultra | 131.0 |
| 2560x1440 | Low | 439.0 |
| 2560x1440 | Medium | 374.0 |
| 2560x1440 | High | 317.0 |
| 2560x1440 | Ultra | 238.0 |
| 1920x1080 | Low | 586.0 |
| 1920x1080 | Medium | 554.0 |
| 1920x1080 | High | 484.0 |
| 1920x1080 | Ultra | 364.0 |
Thief with Ryzen 5 5600G + Other GPUs
See how Thief performs with the same CPU but different graphics cards.
Thief with RTX 5070 Ti + Other CPUs
See how Thief performs with the same GPU but different processors.
Other Games with Ryzen 5 5600G + RTX 5070 Ti
Explore FPS benchmarks for other games using this same hardware combination.