Thief

Resolution Scaling — how FPS drops from 1080p to 4K and what it says about the limiting component

The measured frame rates for Thief reveal a classic scaling pattern that shifts from CPU-bound at 1080p to increasingly GPU-bound as resolution climbs. At 1080p with Low settings, the AMD Ryzen 5 2600 and Radeon RX 6750 XT produce 111.3 FPS, but this drops to 96.8 FPS at 1440p and 79.6 FPS at 4K. That represents a 13% reduction from 1080p to 1440p, and a further 18% reduction from 1440p to 4K under the same Low preset.

The scaling behavior changes dramatically when moving to higher quality settings. With Ultra settings, the frame rate at 1080p is 88.3 FPS, which falls to 76.4 FPS at 1440p (a 13% drop) and then to 44.2 FPS at 4K (a 42% drop from 1440p). This steep cliff between 1440p and 4K at Ultra strongly indicates that the GPU is becoming the primary bottleneck at 4K, whereas at 1080p the CPU is likely constraining performance — the relatively modest 23 FPS gap between Low and Ultra at 1080p (111.3 vs 88.3) suggests the processor cannot fully feed the graphics card even at lower quality settings.

Looking at the Middle ground, Medium settings show 97.5 FPS at 1080p, 88.4 FPS at 1440p, and 61.2 FPS at 4K. The scaling from 1080p to 1440p is again around 9%, but the leap to 4K costs 31% performance. High settings follow a similar trajectory: 91.9 FPS at 1080p, 80.7 FPS at 1440p (12% drop), and 46 FPS at 4K (43% drop). The consistency of these 4K penalties across Medium, High, and Ultra — all in the 31-43% range — confirms that the 12 GB VRAM buffer and 192-bit memory interface are not the limiting factors; rather, the GPU's raw compute throughput at 4K is what caps performance.

Interestingly, the gap between Low and Ultra narrows as resolution increases. At 1080p, Ultra is 21% slower than Low (88.3 vs 111.3 FPS). At 1440p, that gap shrinks to 21% as well (76.4 vs 96.8 FPS). At 4K, Ultra is 44% slower than Low (44.2 vs 79.6 FPS). This widening delta at higher resolutions is the signature of a system transitioning from CPU-limited to GPU-limited — at 4K, the GPU is working far harder per frame, making quality settings matter more.

The data implies that at 1080p and 1440p, the Ryzen 5 2600's six cores and twelve threads are sufficient to keep pace with most settings, but the 4K results are entirely dictated by the RX 6750 XT's capabilities. For a game released with a 2014 date, the absolute frame rates are high across the board, but the scaling curve suggests this pairing is best suited for 1080p or 1440p gaming if high quality presets are desired.

GPU Role — VRAM, clocks, and how they relate to this game's results

The AMD Radeon RX 6750 XT brings 12 GB of GDDR6 memory on a 192-bit bus, yielding 432.0 GB/s of bandwidth. Its base clock of 2150 MHz boosts to 2600 MHz, with a game clock of 2495 MHz. These specifications place the GPU in the 79th percentile among all GPUs in the database, with an average benchmark score of 35327. The nearest rival data shows it trading blows with the NVIDIA GeForce RTX 3080 (deltaPct of -1.3%), meaning the RX 6750 XT is roughly 1.3% slower in average score — a negligible difference in real-world usage.

For Thief specifically, the GPU's memory subsystem appears sufficient. The 12 GB capacity is far beyond what a 2014 game would require at any resolution, and the 432.0 GB/s bandwidth is ample for the textures and geometry in this title. The measured frame rates at 4K Ultra (44.2 FPS) suggest the GPU is compute-bound rather than memory-bound, as the 2560 shading units and 64 ROPs are being pushed to their limits at that resolution.

The GPU's 13.31 TFLOPS of FP32 performance and 416.0 GTexel/s texture rate are the likely determinants of the 4K scaling cliffs observed. At 1080p and 1440p, these figures are more than adequate, but 4K's four times pixel count (relative to 1080p) overwhelms the compute resources. The 79th percentile ranking indicates this is a high-performing GPU in general, yet it still cannot maintain 60 FPS at 4K Ultra in Thief — the measured 44.2 FPS falls short.

Comparing the RX 6750 XT to its nearest rival, the AMD Radeon RX 5600M (deltaPct of 0.2%), the desktop card's advantage is essentially statistical noise. The RTX 5070 Ti Mobile and AMD Radeon 880M are similarly close, within 0.9% either way. This tells a curious story: in synthetic benchmarks, these diverse GPUs cluster tightly, but Thief's actual frame rates would likely diverge more significantly due to driver optimizations and architectural differences.

The GPU's end-of-life production status and 2022 release date suggest it is a mature product with mature drivers, which likely benefits consistency in older titles like Thief. The 250 W TDP and 600 W suggested PSU indicate a power-hungry card, but the benchmark data shows it delivers strong performance in this game's engine.

How This Combo Ranks — use comboRankInGame vs other tested combos

The combination of AMD Ryzen 5 2600 and AMD Radeon RX 6750 XT ranks 1085 out of 1642 tested combos for Thief. This places it in the 34th percentile of all system pairings — meaning roughly two-thirds of tested combinations perform better in this specific game. Given that the CPU sits in the 77th percentile and the GPU in the 79th percentile individually, the combo ranking is notably lower than either component's standalone percentile.

This discrepancy suggests a potential bottleneck or suboptimal pairing for this particular title. The Ryzen 5 2600's six cores and twelve threads, with a boost clock of 3.90 GHz, may be holding back the RX 6750 XT in Thief's engine. The game's 2014 release date means it was designed for CPUs of that era, and its threading model may not scale well beyond a few cores, leaving the Ryzen's additional threads underutilized.

The rank of 1085 out of 1642 implies that many combinations with weaker GPUs but stronger single-threaded CPUs outperform this pairing. The CPU's nearest rivals — the Intel Core i5-1345U (deltaPct -0.4%), AMD Ryzen 5 7530U (0.6%), Intel Core i5-11600KF (-0.6%), and Intel Core i7-10700F (0.7%) — all have similar average benchmark scores, but their newer architectures likely deliver better single-thread performance that matters for this game.

Conversely, the GPU's nearest rivals include the NVIDIA GeForce RTX 3080, which is only 1.3% faster on average. Yet in Thief, a system with an RTX 3080 and a stronger CPU would likely rank far higher than 1085th, reinforcing the idea that the CPU is the weak link in this combo for this specific title. The data suggests that for Thief, prioritizing CPU single-thread performance over GPU raw power is the more effective strategy.

FAQ

Q: What is the best resolution for this combo in Thief?

A: The data shows 1080p provides the highest frame rates across all settings, with Low achieving 111.3 FPS and Ultra reaching 88.3 FPS. At 1440p, Ultra still maintains 76.4 FPS, which is playable, but 4K Ultra drops to 44.2 FPS.

Q: Does the RX 6750 XT bottleneck at 4K in Thief?

A: Yes, the scaling data indicates a severe GPU bottleneck at 4K. Ultra settings at 4K are 44% slower than Low at 4K (44.2 vs 79.6 FPS), and the 42% drop from 1440p to 4K at Ultra confirms the GPU is the limiting component at this resolution.

Q: How does the Ryzen 5 2600 compare to its nearest rivals?

A: The Ryzen 5 2600 has an average benchmark score of 19616, which is nearly identical to the Intel Core i5-1345U (19691, -0.4% delta), AMD Ryzen 5 7530U (19508, +0.6%), Intel Core i5-11600KF (19740, -0.6%), and Intel Core i7-10700F (19485, +0.7%). All are within 0.7% of each other.

Q: Is the 12 GB VRAM sufficient for Thief at 4K?

A: Yes, the 12 GB GDDR6 memory is more than adequate for a 2014 game. The measured frame rates at 4K (44.2-79.6 FPS depending on settings) are limited by compute performance, not memory capacity or bandwidth.

Q: What settings provide the best experience at 1440p?

A: At 1440p, High settings deliver 80.7 FPS, while Ultra only drops to 76.4 FPS. The difference between High and Ultra is small (5.3%), so Ultra is recommended if visual quality is prioritized, as both remain above 60 FPS.

Q: How does this combo rank among all tested configurations?

A: This combo ranks 1085 out of 1642 tested combinations for Thief, placing it in the lower third of all pairings despite both components ranking in the upper 80th percentile individually.

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

The AMD Ryzen 5 2600 is a six-core, twelve-thread processor from the 2000 series, based on the Zen architecture (specifically Zen+ Pinnacle Ridge) on a 12 nm process. It operates at a base clock of 3.40 GHz and boosts to 3.90 GHz, with 16 MB of shared L3 cache. Its benchmark data shows a Cinebench R23 multicore score of 11186 and single-core score of 1579, placing it in the 77th percentile among all CPUs.

In Thief, the CPU's performance appears to be a limiting factor at lower resolutions. At 1080p, the difference between Low and Ultra settings is only 23 FPS (111.3 vs 88.3), which is a smaller gap than would be expected if the GPU were fully utilized. This compression of frame rates across quality settings at 1080p is a classic sign of CPU bottlenecking — the processor cannot feed the GPU enough frames to differentiate between settings.

The CPU's nearest rivals show comparable average scores (within 0.7%), but the Ryzen's 3.90 GHz boost clock is modest by modern standards. Thief, being a 2014 title, likely relies heavily on single-thread performance, and the Ryzen 5 2600's single-core Cinebench R23 score of 1579 is adequate but not exceptional. The 12 threads provide good multi-threaded throughput for other workloads, but this game may not leverage them effectively.

The 65 W TDP and 46.9 GB/s memory bandwidth are reasonable for the era, but the CPU's memory support is dual-channel DDR4, which matches the game's requirements. The data suggests that upgrading to a CPU with higher single-thread performance would yield noticeable gains at 1080p, while the GPU would remain the bottleneck at 4K regardless of CPU choice.

The CPU's percentiles and rival comparisons indicate it is a mid-range performer in 2024, but for a 2018-released processor, it holds its own. The 77th percentile ranking means it outperforms 77% of all CPUs in the database, yet in Thief, the combo rank drops to 1085th out of 1642, confirming that this game does not favor the Ryzen 5 2600's architecture.

Settings Recommendations — which preset gives the best experience per the measured rows

Based on the measured FPS data, the optimal settings depend heavily on the target resolution and desired frame rate. At 1080p, all settings deliver playable frame rates, with Low at 111.3 FPS being the fastest and Ultra at 88.3 FPS being the slowest but still well above 60 FPS. For a smooth 60+ FPS experience with maximum visual fidelity, Ultra at 1080p is the clear choice, offering 88.3 FPS.

At 1440p, the situation is slightly more nuanced. High settings provide 80.7 FPS, while Ultra only drops to 76.4 FPS — a mere 5.3 FPS difference. Since both exceed 60 FPS, Ultra is recommended for 1440p users who want the best visual quality without sacrificing playability. Medium settings at 88.4 FPS are also viable, but the visual trade-off is harder to justify when Ultra is only 12 FPS slower.

At 4K, the recommendations change dramatically. Ultra settings produce 44.2 FPS, which is below the 60 FPS threshold for smooth gameplay. High settings at 46 FPS are only marginally better, while Medium at 61.2 FPS crosses the 60 FPS line. Low settings at 79.6 FPS are the only preset that provides comfortable headroom above 60 FPS. For 4K gaming, Medium settings at 61.2 FPS offer the best balance of visual quality and performance, while Low at 79.6 FPS is recommended for those prioritizing frame rate stability.

The data shows that Medium settings at 4K (61.2 FPS) provide a 38% improvement over Ultra at 4K (44.2 FPS), while the visual difference between Medium and Ultra in a 2014 game is likely modest. Conversely, at 1080p, the gap between Low and Ultra is only 21%, so users can afford to max out settings without significant performance penalties. The overall recommendation: use Ultra at 1080p or 1440p, and Medium at 4K.

Measured FPS Breakdown — resolution by resolution, settings by settings, exact numbers

The complete measured dataset for Thief with the AMD Ryzen 5 2600 and AMD Radeon RX 6750 XT is as follows:

1920x1080 (1080p):

• Low: 111.3 FPS
• Medium: 97.5 FPS
• High: 91.9 FPS
• Ultra: 88.3 FPS

The spread from Low to Ultra is 23 FPS, with High and Ultra being only 3.6 FPS apart. This clustering at the top end suggests diminishing returns beyond High settings at this resolution.

2560x1440 (1440p):

• Low: 96.8 FPS
• Medium: 88.4 FPS
• High: 80.7 FPS
• Ultra: 76.4 FPS

At 1440p, the gap between Low and Ultra widens to 20.4 FPS, with Medium providing a solid middle ground. High and Ultra remain within 4.3 FPS of each other, again showing minimal benefit from the highest preset.

3840x2160 (4K):

• Low: 79.6 FPS
• Medium: 61.2 FPS
• High: 46 FPS
• Ultra: 44.2 FPS

The 4K results show the most dramatic spread, with Low to Ultra differing by 35.4 FPS. The cliff between Medium (61.2) and High (46) is particularly steep at 15.2 FPS, indicating that High and Ultra settings push the GPU beyond its comfortable operating range at this resolution.

Comparing across resolutions for each setting: Low scales from 111.3 to 96.8 to 79.6 FPS (1080p to 1440p to 4K), representing drops of 13% and 18% respectively. Medium scales from 97.5 to 88.4 to 61.2 FPS, with drops of 9% and 31%. High scales from 91.9 to 80.7 to 46 FPS, with drops of 12% and 43%. Ultra scales from 88.3 to 76.4 to 44.2 FPS, with drops of 13% and 42%. The accelerating penalties at 4K for Medium, High, and Ultra reinforce the GPU-bound nature of higher resolutions, while Low's more gradual decline suggests the CPU remains a factor even at 4K with minimal graphical load.