Thief

# CPU Role — cores, clocks, and how they relate to this game's results

The AMD Ryzen 5 2600 is a 6-core, 12-thread processor built on the Zen architecture, with a base clock of 3.40 GHz and a boost clock of 3.90 GHz. This CPU sits in the 77th percentile of all processors tested, with an average benchmark score of 19,616. Its nearest rivals in aggregate CPU performance are the Intel Core i5-1345U (19,691, -0.4% delta), AMD Ryzen 5 7530U (19,508, +0.6% delta), Intel Core i5-11600KF (19,740, -0.6% delta), and Intel Core i7-10700F (19,485, +0.7% delta). The tight clustering of these deltas—all within roughly one percent—indicates that the Ryzen 5 2600 delivers compute throughput squarely in the mainstream desktop range.

For Thief, a 2014 stealth title, the CPU's role is primarily to feed the GPU with draw calls and maintain a stable frame pacing floor. In single-threaded workloads, the Ryzen 5 2600 scores 1,579 in Cinebench R23 single-core and 2,239 in Passmark single-thread, which are modest figures by modern standards. The game's engine, released in early 2014, was designed for the multi-core CPUs of that era, so the 6-core/12-thread configuration here is more than sufficient to avoid becoming the primary bottleneck at most settings. The L3 cache is 16 MB shared, and memory bandwidth is rated at 46.9 GB/s over dual-channel DDR4, which provides adequate data throughput for a game of this vintage.

The measured FPS data confirms that this CPU does not meaningfully cap performance. At 1080p Low, the combo reaches 107.5 FPS average, while at 4K Ultra it drops to 60.2 FPS. The scaling pattern from Low to Ultra at each resolution shows a relatively modest performance penalty—typically 15-20%—which suggests that the GPU is the dominant limiting factor once settings increase. At 1080p Low, where CPU load is proportionally higher relative to GPU work, the system still delivers triple-digit frame rates, indicating the Ryzen 5 2600 has ample headroom for this title.

# FAQ

Q: How does the Ryzen 5 2600 compare to its closest CPU rivals in aggregate performance?

A: The Ryzen 5 2600 scores 19,616 on average. The Intel Core i5-1345U is 0.4% faster (19,691), the AMD Ryzen 5 7530U is 0.6% slower (19,508), the Intel Core i5-11600KF is 0.6% faster (19,740), and the Intel Core i7-10700F is 0.7% slower (19,485). All four rivals fall within one percentage point, making them statistically indistinguishable in overall throughput.

Q: What frame rates does this combo achieve at 1080p across different settings?

A: At 1920x1080, the average FPS is 107.5 on Low, 101.7 on Medium, 89.2 on High, and 87.5 on Ultra. The spread from Low to Ultra is 20 FPS, which represents an 18.6% drop from the lowest to highest settings preset.

Q: Is the RTX 5070 the bottleneck at any resolution?

A: The data indicates GPU-bound behavior at higher resolutions. From 1080p to 1440p at High settings, average FPS drops from 89.2 to 82.0 (an 8.1% reduction), and from 1440p to 4K it falls further to 64.8 (a 21.0% reduction). This escalating penalty with resolution is characteristic of GPU limitations, not CPU constraints.

Q: How does this combo rank among all tested combinations for Thief?

A: The combo ranks 995th out of 1,642 tested combinations, placing it in the upper-middle portion of the distribution. This rank reflects the pairing of a mid-range 2018 CPU with a high-end 2025 GPU, where the GPU contributes most of the gaming performance.

Q: What is the best performing settings preset at each resolution?

A: Low settings yields the highest average FPS at every resolution: 107.5 at 1080p, 99.6 at 1440p, and 83.1 at 4K. Ultra settings produces the lowest frame rates: 87.5 at 1080p, 78.9 at 1440p, and 60.2 at 4K.

Q: Does the CPU's single-thread performance limit the game at 1080p?

A: No. Even at 1080p Low, where CPU overhead is most pronounced relative to GPU rendering time, the combo still produces 107.5 FPS. The Ryzen 5 2600's single-core score of 1,579 in Cinebench R23 is adequate for Thief's engine, which does not require extreme per-thread performance to maintain high frame rates.

# How This Combo Ranks

The combination of the AMD Ryzen 5 2600 with the NVIDIA GeForce RTX 5070 ranks 995th out of 1,642 tested combos in Thief. This places the pairing in the 60th percentile of all tested systems—a solidly mid-pack result that might surprise given the GPU's capabilities. The RTX 5070 itself sits in the 83rd percentile of all GPUs, with an average benchmark score of 41,687, and its nearest rivals include the AMD Radeon Pro 5300 (41,610, +0.2% delta), NVIDIA Tesla M40 (41,897, -0.5% delta), NVIDIA GeForce RTX 3090 (41,441, +0.6% delta), and AMD Radeon Pro 580X (41,991, -0.7% delta). All of these rival GPUs are within one percent of the RTX 5070's aggregate score, underscoring that this GPU's raw compute is not dramatically ahead of its nearest competitors.

The relatively modest combo rank reflects the CPU's position. The Ryzen 5 2600, with its 77th percentile standing, is a capable but not exceptional processor for gaming. In a game like Thief, which is not particularly demanding on CPU resources, the 2600 does not hold back the RTX 5070 significantly—but it also does not boost the pairing. The data suggests that most of the variance in this combo's rank comes from the GPU, which delivers frame rates that are respectable but not top-tier due to the CPU's older architecture and lower IPC compared to newer processors.

In the context of the full combo distribution, this pairing sits near the median. Systems with newer CPUs and the same GPU would likely rank higher, while those with weaker CPUs would rank lower. The 995th position out of 1,642 indicates that this is a balanced-but-not-optimized pairing, where the GPU's high-end capabilities are partially tempered by the CPU's mid-range standing.

# Measured FPS Breakdown

The measured FPS data covers three resolutions (1920x1080, 2560x1440, 3840x2160) and four settings presets (Low, Medium, High, Ultra). All data points are measured averages with no minimum or maximum frame times recorded.

At 1920x1080, the average FPS values are: Low 107.5, Medium 101.7, High 89.2, Ultra 87.5. The progression from Low to Ultra shows a steady decline, with the largest single-step drop occurring between Medium and High (12.5 FPS, or 12.3%). The difference between High and Ultra is just 1.7 FPS (a 1.9% reduction), indicating that the Ultra preset adds little visual cost beyond High at this resolution.

At 2560x1440, the averages are: Low 99.6, Medium 85.6, High 82.0, Ultra 78.9. The drop from Low to Medium is substantial (14.0 FPS, or 14.1%), while the Medium-to-High and High-to-Ultra steps are smaller (3.6 FPS and 3.1 FPS, respectively). This pattern suggests that Medium introduces a significant rendering load increase, but subsequent preset tiers add progressively less strain.

At 3840x2160, the averages are: Low 83.1, Medium 73.0, High 64.8, Ultra 60.2. The Low-to-Medium drop is 10.1 FPS (12.2%), Medium-to-High is 8.2 FPS (11.2%), and High-to-Ultra is 4.6 FPS (7.1%). The diminishing penalties at higher presets indicate that the GPU is becoming increasingly saturated as resolution and settings scale, with the absolute frame rates approaching the 60 FPS threshold at Ultra.

Across all resolutions, Low settings consistently delivers the highest frame rates, while Ultra delivers the lowest. The maximum frame rate observed in the entire dataset is 107.5 FPS (1080p Low), and the minimum is 60.2 FPS (4K Ultra).

# Resolution Scaling

The resolution scaling behavior reveals where the bottleneck lies. From 1080p to 1440p at each settings preset, the FPS drops are: Low from 107.5 to 99.6 (7.3% reduction), Medium from 101.7 to 85.6 (15.8% reduction), High from 89.2 to 82.0 (8.1% reduction), and Ultra from 87.5 to 78.9 (9.8% reduction). The Medium preset shows the largest relative drop, which is anomalous compared to the other presets.

From 1440p to 4K, the reductions are more pronounced: Low from 99.6 to 83.1 (16.6% reduction), Medium from 85.6 to 73.0 (14.7% reduction), High from 82.0 to 64.8 (21.0% reduction), and Ultra from 78.9 to 60.2 (23.7% reduction). The increasing percentage drops at higher settings and resolutions strongly indicate that the GPU is the limiting component. A CPU-bound scenario would show minimal FPS changes as resolution increases, since the CPU's workload remains roughly constant regardless of pixel count. Instead, the data shows a clear pattern of escalating penalties with more pixels, which is the hallmark of GPU saturation.

The 4K Ultra result of 60.2 FPS is particularly telling. It sits just above the 60 FPS threshold, suggesting that the RTX 5070 is working near its limit for this title at maximum settings. The 23.7% drop from 1440p Ultra to 4K Ultra is the largest proportional decrease in the dataset, confirming that pixel throughput is the primary constraint. Meanwhile, the 1080p Low result of 107.5 FPS demonstrates that when the GPU load is light, the CPU can still drive high frame rates without becoming a bottleneck.

# Settings Recommendations

Based on the measured data, the optimal settings preset depends on the target resolution and desired frame rate. At 1080p, all four presets deliver above 87 FPS average, so the choice comes down to visual fidelity versus headroom. Ultra at 87.5 FPS is only 2% slower than High at 89.2 FPS, making Ultra the best choice for 1080p users who want maximum quality without sacrificing meaningful performance. The 20 FPS gap between Low and Ultra (107.5 vs. 87.5) is notable, but both are playable, so Ultra is recommended for the best experience.

At 1440p, the Medium preset (85.6 FPS) provides a balanced point, sitting 4.4% above High (82.0 FPS) and 8.5% above Ultra (78.9 FPS). The step from Medium to High costs 3.6 FPS, which may be worth it for improved visuals, but Ultra adds only marginal quality over High for a further 3.1 FPS loss. For users prioritizing smoothness, Medium is the sweet spot; for those wanting higher fidelity, High offers a reasonable compromise at 82.0 FPS.

At 4K, the situation changes. Low delivers 83.1 FPS, but Medium drops to 73.0 FPS, and High falls to 64.8 FPS. Ultra at 60.2 FPS is playable but leaves no headroom for demanding scenes. The data suggests that High at 4K (64.8 FPS) is the highest preset that maintains a comfortable margin above 60 FPS, while Ultra is viable for those who accept occasional dips below the 60 FPS threshold. Given the diminishing returns between High and Ultra (only 4.6 FPS difference), High is the recommended preset at 4K for the best balance of visual quality and frame rate stability.

Overall, the best experience across all resolutions is achieved by selecting Ultra at 1080p, High at 1440p, and High at 4K. These choices maximize visual quality while keeping average FPS above 60 at every resolution, with the 1080p Ultra result of 87.5 FPS providing ample headroom for smooth gameplay.