Thief
This combination provides smooth gameplay with an average of 104 FPS, suitable for most gaming scenarios.
Average FPS: resolution vs quality settings
| Ultra | High | Medium | Low | |
|---|---|---|---|---|
| 4K Ultra HD | 68 | 71 | 92 | 102 |
| 1440p QHD | 98 | 99 | 110 | 121 |
| 1080p Full HD | 111 | 113 | 124 | 138 |
Cell color: green is 120+ FPS, teal is 60+, amber is 30+, red is below 30.
Thief with AMD Ryzen 5 5600F + NVIDIA GeForce RTX 5070 Ti, In-Depth Analysis
The AMD Ryzen 5 5600F and NVIDIA GeForce RTX 5070 Ti combination delivers smooth, high-refresh gameplay in Thief across all tested resolutions, with the GPU stepping down as the resolution rises. At 1080p, the system never drops below 111 FPS on any preset, and even the most demanding 4K Ultra setting maintains a 68 FPS average. This pairing scores in the 86th percentile for the GPU and 85th percentile for the CPU on their respective aggregate benchmarks, and it ranks 1360th out of 3720 tested configurations in this specific game. The data indicates a well-balanced system where the GPU handles high-fidelity visuals while the Ryzen 5 5600F provides consistent frame delivery, making this a strong candidate for both 1440p UHD and 1080p high-refresh setups.
FAQ
Q: What is the measured average FPS at 1080p Ultra settings?
A: At 1920x1080, the Ultra preset averages 111 FPS. Since the measured rows only include the average, this is the highest resolution and setting combination tested for this game.
Q: Does the system maintain playable FPS at 4K Ultra?
A: Yes. The 3840x2160 Ultra preset reports a 68 FPS average. While this is below the 60–70 FPS comfort zone for many competitive players, it remains playable and is consistent with the GPU’s core position relative to its rivals.
Q: What is the performance difference between Low and Ultra at 1080p?
A: The Low preset reaches 138 FPS, while Ultra yields 113 FPS, a difference of about 10% (25 FPS). This scaling shows that the CPU still matters even at 1080p, but the GPU does not create an overall bottleneck.
Q: How does the RTX 5070 Ti compare to its nearest rival in aggregate benchmarks?
A: In PassMark G3D, the RTX 5070 Ti has an average score of 32974 and is 0.1% behind the AMD Radeon RX Vega 64 (50001), 0.4% ahead of the Intel Arc A550M (49737), and 2% behind the RX 6900 XT (50951) and 3.1% ahead of the RX 6800 XT (48477). The GPU is above the 86th percentile of all GPUs.
Q: What are the CPU’s boost clock and memory bandwidth?
A: The Ryzen 5 5600F boosts to 4.00 GHz from a 3.00 GHz base, uses a dual-channel DDR4 memory interface with 51.2 GB/s bandwidth, and carries a 32 MB shared L3 cache. In PassMark single-thread, it scores 2872 and multithread scores 19236, placing it in the 85th percentile.
GPU Role
The NVIDIA GeForce RTX 5070 Ti is built on the Blackwell 2.0 architecture using the GB203 die, fabricated on a 5 nm TSMC node. The GPU card’s 16 GB GDDR7 memory on a 256-bit bus provides a memory bandwidth of 896.0 GB/s, which is more than enough to feed the high shading-unit count of 8960. Its base clock is 2295 MHz, boost to 2442 MHz, and it offers a texture rate of 686.6 GTexel/s and a pixel rate of 235.4 GPixel/s. The 70 RT cores and 280 tensor cores support DirectX 12 Ultimate and Vulkan 1.4, ensuring forward-looking API support.
In terms of fixed performance, the card scores 32974 in PassMark G3D, 20203 in Compute, and 6604 in the 3DMark Steel Nomad DX12 test. The GPU’s percentile standing (86th) is only one point above the CPU’s 85th percentile, indicating both components are in the same performance tier. However, the GPU’s raw speed is what carries the system in rasterized games like Thief. At 4K, the graph shows a clear dependence on the GPU: the High preset at 4K averages 71 FPS, while Ultra drops to 68 FPS, with the memory bandwidth and 256-bit interface proving crucial for high-texture workloads.
The GPU’s core clock (2295 MHz base, 2452 MHz boost) provides the raw compute needed to handle the game’s geometry and physics, but the relatively low GPU power target (300 W) and single 700W recommended PSU suggest it is not a runaway power hog. The 704 mm length and dual-slot width make it compatible with most mid-tower cases, and the 3x DisplayPort 2.1b and 1x HDMI 2.1b outputs support up to 4K 240Hz displays. In Thief, the GPU alone is not the limiting factor at 1080p—where the CPU still holds the system above 130 FPS—but it becomes the bottleneck as resolution increases.
CPU Role
The AMD Ryzen 5 5600F is a 6-core, 12-thread processor built on the Zen 3 architecture (code named Vernmere) using a 7 nm TSMC process. It operates at a base 3.00 GHz and boosts to 4.00 GHz, has a 65W TDP, and is designed for socket AM4 with dual-channel DDR4-2133 memory support (1.2 GB/s bandwidth). The processor integrates 64 scalability per core and 32 MB of shared L3 cache, providing sufficient cache hit rates for older, single- or dual-thread titles like Thief.
In PassMark benchmarks, the CPU scores 2872 in single-thread (same as the single-thread test) and 19236 in multithread, with an average benchmark score of 36945. It takes a full 12-thread advantage; however, Thief is a directX 11-era game (release date 2014-02-24) that is not heavily multi-threaded. The data confirms this pattern: at 1080p Low, the CPU is feeding 138 FPS, which is well above the monitor refresh rates of the era. But as resolution climbs, the GPU becomes the limiting factor. For example, at 4K Ultra, the frame rate drops to 68 FPS, a 34% drop from 1080p Ultra, showing the CPU is not the primary bottleneck at higher resolutions.
The CPU’s 50.2GB/s memory bandwidth is sufficient for the game’s assets. The BIOS has a huge 32 MB L3 cache; this can reduce memory latency and helps maintain stable frame pacing. The 5600F’s aggregate performance places it 87th among all CPUs (85th percentile), with its closest competitor being the Intel Core Ultra Ultra 9 285K (by 0 – 0.2% in delta). In practice, the 5600F delivers exactly the level of performance needed to prevent frame stuttering typical of a 2014 stealth game when paired with a top-tier GPU.
Measured FPS Breakdown
Measured performance is broken down by resolution and preset for this specific configuration. All numbers are measured in-engine and are available for Low, Medium, High, and Ultra presets (except for Medium which has min/max also reported).
1920x1080 (Full HD):
- Low: 138 FPS average
- Medium: 124 FPS average (min 106, max 143)
- High: 113 FPS average
- Ultra: 111 FPS average
2560x1440 (WQHD):
- Low: 121 FPS average
- Medium: 110 FPS average (min 93, max 126)
- High: 99 FPS average
- Ultra: 98 FPS average
3840x2160 (4K UHD):
- Low: 102 FPS average
- Medium: 92 FPS average (min 78, max 106)
- High: 71 FPS average
- Ultra: 68 FPS average
The data shows a smooth escalation of FPS as settings are lowered at 1080p, where the difference between Low and Ultra is only 7 FPS (138 vs 111). That margin widens at 144p (from 121 to 98, a 19 FPS spread) and grows drastically at 4K (102 to 68, a 34% reduction). The only preset with min/max data is Medium; at 1080p the minimum is 106, meaning the worst-case frame is still above 60, and at 4K the minimum is 78, which is still comfortable for a single-player stealth game.
The 144p differences are ‒ Low’s 121, Medium 110, High 99, Ultra 98—showing only 3 FPS between High and Ultra, indicating that Ultra may not be necessary for smooth play at 1440p. At 4K the gap widens: 102, 92, 71, 68.
Resolution Scaling
To isolate the impact of resolution, we compare the average FPS across the three resolutions for each preset.
- Low preset: 138 (1080) → 121 (144) → 102 (4K) – a drop of 17% from 1080p to 144p, and another 16% from 144 to 4K.
- Medium preset: 124 → 110 → 92 – a 11% drop from 1080 to 144, then 16% lower to 4K.
- High preset: 113 → 99 → 71 – a 12% drop to 144p, then 28% drop to 4K.
- Ultra preset: 111 → 98 → 68 – an 11% drop to 144, then 30% drop to 4K.
These numbers indicate a clear resolution transition. At 1080p, the CPU limits the system: the difference between Low and Ultra is only 8% (38 FPS), but when moving to 4K, the same preset change results in a 34% performance reduction (from 133 to 116). This pattern shows that the RTX 5070 Ti is the dominant factor at 4K, while the Ryzen 5 gives the system at 1080p a performance cap. The per-percent scaling from 1080p to 144p is mild (11–17%), but the jump from 144p to 4K is steep (16–30%), making 1440p the sweet spot for any preset.
Because the GPU’s memory bandwidth of 896 GB/s is unchanged, the huge pixel count at 4K (approxined 8.3 million pixels) necessitates higher shading and texture bandwidth per frame, which it delivers with only a 50 FPS penalty on 4K low vs 1080p low. The raw compute (43.94 TFloFi) is the same at all resolutions, so the drop is fully attributed to the pixel pars and workload increase.
Settings Recommendations
Based on the measured FPS for the Ryzen 5 + RTX 5070 Ti, the optimal balance of visual quality and frame rate depends on the target display.
For 1080p gamers with high-refresh monitors (144Hz+), the Ultra preset delivers an average of 111 FPS, which is above the 60 FPS refresh rate but could stay under 144 in some scenes. If eye-candy is a priority, Ultra is perfect because the CPU can still inject enough frames; the difference between Medium and Ultra is only 13 FPS (124 vs 111). However, if the user craves the absolute max rate, Low (138 FPS) or Medium (124) become the go‑to settings.
For 1440p, the Ultra setting averages 98 FPS, which is excellent for a 144Hz monitor, still holding above 100. High is 3 FPS less (99), so the difference between them is negligible – both produce silky AA. Medium is still 110 FPS, so lowering from Ultra to Medium gives a 12 FPS boost, at the cost of shadow quality only. Thus, 1440p Ultra is recommended, as it maintains high FPS without needing a settings reduction.
For 4K, the narrative changes. The Ultra preset’s 68 FPS average is below the 60 FPS disirable threshold, and the absence of min/max for that preset prevents assurance of frame stability. The High preset offers 71 FPS, eight frames above Ultra, and the Medium preset averages 92 FPS with a min of 78 FPS. Given that a 92 FPS average with a floor of 78 FPS falls well into the 60–120 Hz range in most games, the best setting for 4K is Medium. High and Ultra provide only a 2–3 FPS difference over Medium (98 to 99), but they are around 21–24 FPS slower, and every frame counts at 4K to keep the GPU from dropping below 60 in intense scenes. Therefore, for 4K the recommend setting is Medium, as it guarantees smooth play; if the user prioritizes visuals and is willing to accept below-60 dips, Ultra is playable with a 68 FPS average.
In every case, the data shows that the chipset delivers high frame rates even at max settings until 4K, where a modest reduction to Medium still preserves an excellent 92 FPS.
Hardware Specifications
AMD Ryzen 5 5600F
NVIDIA GeForce RTX 5070 Ti
FPS Benchmarks by Resolution & Settings
Performance data for AMD Ryzen 5 5600F + NVIDIA GeForce RTX 5070 Ti in Thief
| Resolution | Settings | Avg FPS |
|---|---|---|
| 3840x2160 | Low | 102.0 |
| 3840x2160 | Medium | 92.0 |
| 3840x2160 | High | 71.0 |
| 3840x2160 | Ultra | 68.0 |
| 2560x1440 | Low | 121.0 |
| 2560x1440 | Medium | 110.0 |
| 2560x1440 | High | 99.0 |
| 2560x1440 | Ultra | 98.0 |
| 1920x1080 | Low | 138.0 |
| 1920x1080 | Medium | 124.0 |
| 1920x1080 | High | 113.0 |
| 1920x1080 | Ultra | 111.0 |
Thief with Ryzen 5 5600F + 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 5600F + RTX 5070 Ti
Explore FPS benchmarks for other games using this same hardware combination.