Thief

Thief

AVERAGE FPS
103
good

This combination provides smooth gameplay with an average of 103 FPS, suitable for most gaming scenarios.

Average FPS: resolution vs quality settings

UltraHighMediumLow
4K Ultra HD 54 57 75 98
1440p QHD 92 97 117 129
1080p Full HD 118 120 132 146

Cell color: green is 120+ FPS, teal is 60+, amber is 30+, red is below 30.

Every measured configuration

3840x2160 Low 98
3840x2160 Medium 75
3840x2160 High 57
3840x2160 Ultra 54
2560x1440 Low 129
2560x1440 Medium 117
2560x1440 High 97
2560x1440 Ultra 92
1920x1080 Low 146
1920x1080 Medium 132
1920x1080 High 120
1920x1080 Ultra 118

Thief with Intel Core i5-11400F + NVIDIA GeForce RTX 3080, In-Depth Analysis

Measured performance data for Thief running on the Intel Core i5-11400F and NVIDIA GeForce RTX 3080 is analyzed below. The system was tested at 1080p, 1440p, and 4K resolutions across four graphics presets (Low, Medium, High, and Ultra), with the resulting frame rates and system characteristics used to build a reproducibility profile. The combo ranks 1,393rd out of 3,720 tested combos, placing it in the upper half of the database.

Settings Recommendations

At 1080p, the measured frame rates span 118 FPS on Ultra to 146 FPS on Low, with a rather compressed range across the higher presets. Ultra produces 118 FPS, High produces 120 FPS, and Medium produces 132 FPS. The difference between High and Ultra is only 2 FPS, while Medium adds 12 FPS over High. At 1440p the pattern repeats: Ultra delivers 92 FPS, High delivers 97 FPS, Medium posts 117 FPS, and Low peaks at 129 FPS. The same relative ordering holds at 4K, where Low achieves 98 FPS, Medium manages 75 FPS, High drops to 57 FPS, and Ultra settles at 54 FPS.

The sweet spot across all three resolutions is the Medium preset. On every configuration, Medium provides the comfortable margin over Ultra: 14 FPS at 1440p, 17 FPS at 1080p, and 21 FPS at 4K. It also retains a higher minimum frame rate than any other preset; the measured minimum for Medium is 112 FPS at 1080p, 99 FPS at 1440p, and 63 FPS at 4K. These minimums remain well above the average of the next-lower preset (Low) while still leaving headroom for the occasional scene spike. At Medium, the frame rate never falls below 63 FPS in the tested set, so the experience remains consistent on any of these displays.

For gamers who prioritize smoother motion over visual flourishes, 1080p Medium offers 132 FPS, and 1440p Medium still averages 117 FPS. Only at 4K does the recommended wrapper tighten: Medium is the highest preset whose mean stays comfortably above the 75 FPS threshold in the data, while High and Ultra are exactly 57 and 54 FPS. Those who must run Ultra at 4K will accept a 54 FPS average, but this is a clear dip from the 118 FPS at 1080p Ultra. Given the only 2 FPS gain from High to Ultra at 1080p and 1440p, the Ultra preset is essentially a negligible visual boost.

Because the game’s minimum frame data is only recorded for the Medium preset (112 FPS at 1080p, 99 FPS at 1440p, 63 FPS at 4K), we cannot assess the consistency of other presets. The Medium preset provides the best observed minimums, so it is the safest for low frame-times.

Thus, the measured results indicate the following recommendation: Set in-game quality to Medium at 1080p or 1440p, and at 4K choose Medium if smoothness is a priority. High and Ultra remain playable at 1080p and 1440p, but they only yield a marginal visual gain while lossing a significant fraction of the frame rate. At 4K, High and Ultra are best avoided.

GPU Role

The NVIDIA RTX 3080 within this combo carries 8704 shading units, 272 texture mapping units, 96 raster output processors, 68 RT cores and 272 Tensor cores, all running on an 8 nm process by Samsung. The GPU’s memory subsystem is a key factor for Thief’s load: 10 GB of GDDR6X attached on a 320-bit bus reaches 760.3 GB/s of bandwidth. The base clock is 1440 MHz, rising to 1710 MHz under boost. These figures gives strong raw compute capacity, with an FP32 throughput of 29.77 TFLOPS.

The game Thief launched in 2014, and its rendering demands are nowhere near the GPU’s limits. The measured frame deltas across settings (e.g., 146 FPS vs 118 FPS at 1080p) indicate a CPU-light, GPU-bound scenario, but even at the highest preset the average frame remains 118 FPS, implying the GPU is not exhausted. The biggest drop in performance occurs when increasing resolution, from 1080p to 1440p and then to 4K: a 1440p High dips to 97 FPS, and 4K High falls to 57 FPS. This pattern is the signature of a graphics-limited workload, with the GPU’s geometry fill rate and texture rate becoming the bottleneck. The GPU can handle 164.2 GPixel/s of pixel fill and 465.1 GTexel/s, and none of the 12 tested configurations reach that limit; only the 4K Ultra test (54 FPS) suggests the GPU is under heavy stress.

The 760.3 GB/s memory bandwidth is more than sufficient for a game of this era. Thief’s oldest assets and its densely packed environments are unlikely to raise the demand beyond the 10 GB VRAM, especially when the average frame sizes are mostly bound by shader and draw calls rather than residence. The high frame times, however, are not recorded for most presets, so we cannot isolate texture memory pressure. In practice, the GPU never develops memory starvation, and the scaled performance from 1080p to 4K follows the trend of pixel shader workload.

FAQ

Q: What is the boost clock frequency of the CPU in the measured combo?

A: The Intel Core i5-11400F has a base clock of 2.60 GHz and a boost clock of 4.40 GHz. It has 6 cores and 12 threads from the architecture Team Rocket Lake.

Q: Which preset delivers the highest average FPS at 1440p?

A: 1440p Low yields 129 FPS, followed by Medium at 117 FPS, High at 97 FPS, and Ultra at 92 FPS. Therefore, the Low preset is the fastest at that resolution.

Q: At 4K, what is the average FPS for the Ultra preset?

A: The 4K Ultra test shows an average frame rate of 54 FPS. High gives 57 FPS, Medium 75 FPS, and Low 98 FPS.

Q: How much VRAM does the RTX 3080 GPU have and what is its memory type?

A: The GPU has 10 GB of GDDR6X memory on a 320‑bit interface, with a memory bandwidth of 760.3 GB/s.

Q: How does the CPU benchmark compare to its closest rival in the data?

A: The CPU’s average benchmark score is 17177. It is 0.3% higher than the AMD EPYC 7H12 (17120) and 0.4% lower than the Intel Core i5-12450H (17239), with a percentile rank of 71 among all CPUs.

Q: How does the GPU’s average score compare to a specific listed competitor?

A: The RTX 3080 logs an average benchmark score of 23172. It is 0.3% lower than the NVIDIA P106-100 (23249) and 0.4% lower than the AMD Radeon RX 6600M (23273). Its percentile among all GPUs is 68.

How This Combo Ranks

This specific pairing (i5-11400F + RTX 3080) ranks 1,393rd out of 3,720 combos tested in Thief engine evaluation. That places it in the upper half, though not near the top 10%. The CPU’s own stand-alone percentile is 71%, and the GPU is at the 68th percentile of all GPUs. When combined in this game, the frame rates are high enough to put it ahead of nearly two‑thirds of other PC configurations tested, which indicates that the i5-11400F does not bottleneck the RTX 3080 in Thief’s normal load.

The CPU’s nearest benchmark rivals provide context: its average score of 17177 sits 0.3% above the AMD EPYC 7H12 and 0.6% above the Intel Core Ultra 7 164U, while trailing the Core i5-12450H by 0.4% and the Ryzen 3 PRO 8300G by 0.6%. All four rivals are within a fractional percentage, meaning the CPU is highly competitive for this game’s compute requirement. On the GPU side, the 3080’s average 23172 is within 0.4% of the AMD Radeon R9 M290X and 0.3% of the NVIDIA P106‑100, showing that the deltas among top-tier parts are small. Yet, the combination of both halves allows this combo to place well inside the top half of Thief configurations, despite not being near the absolute elite.

A rank of 1,393/3,720 means roughly 2 out of every 5 tested combos finish ahead of it. Given that the CPU and GPU individually sit at mid-high percentiles, the game clas ships in holding them back slightly from the very top places. Still, the measured FPS numbers (consistently above 75 at Medium on 4K) confirm that the ranking is not a sign of weakness; it is more a reflection of the hundreds of stronger combos that leverage more recent architectures and larger memory pools.

Resolution Scaling

The frame perfectly scales with resolution in a way that points to a strong GPU limit. At Medium, 1080p averages 132 FPS, 1440p 117 FPS, and 4K 75 FPS. Dropping from 1080p to 1440p reduces frames by a factor of 1.13 (132 ÷ 117), while the pixel count increases by 1.78 (2560×1440 ÷ 1920×1080 = 3,686,400 ÷ 2,073,600). So the frame scaling does not keep up with pixel volume; the CPU becomes more of a constriction. More striking is the 4K rate, 75 FPS, which is only 0.568 × the 1080p Medium. This matches a typical fill-rate bottleneck.

For the High preset, the scaling is steeper: 1080p High is 120 FPS, 1440p High 97 FPS (0.808 × of 1080p), and 4K High 57 FPS (0.475 × of 1080p). The gain of 4K vs 1080p percent of loss in performance is remarkably constant – 0.475 for High and 0.505 for Medium, but the actual frame difference from 1080p to 4K is 63 FPS (120 to 57) and 57 FPS (132 to 75). In every preset, the decrease is dramatic, and the 1080p-to-4K scaling is a 60% – 55% reduction.

This pattern indicates that during rendering, the GPU’s fill rate and ROP output are the limiting factor, not the CPU’s six core, because the compute load remains relatively constant while the number of pixels quadruples. The 1440p result of High still 97 FPS, meaning the GPU can keep up with that resolution, but by 2160p the GPU output drops to roughly half of what i saw at 1080p. If the CPU were the limit, resolution scaling would not cause such a steep fall; instead the absolute FPS at 1080p would be capped and then only slightly reduce. The data thus confirms the RTX 3080 is the primary performance selector in Thief, with the i5-11400F providing enough horsepower to not bottleneck at the lower resolutions.

Measured FPS Breakdown

| Resolution | Settings | Average | Min | Max |

|------------|----------|---------|------|-----|

| 1920x1080 | Low | 146 | – | – |

| 1920x1080 | Medium | 132 | 112 | 152 |

| 1920x1080 | High | 120 | – | – |

| 1920x1080 | Ultra | 118 | – | – |

| 2560x1440 | Low | 129 | – | – |

| 2560x1440 | Medium | 117 | 99 | 134 |

| 2560x1440 | High | 97 | – | – |

| 2560x1440 | Ultra | 92 | – | – |

| 3840x2160 | Low | 98 | – | – |

| 3840x2160 | Medium | 75 | 63 | 86 |

| 3840x2160 | High | 57 | – | – |

| 3840x2160 | Ultra | 54 | – | – |

The table lists the observed average FPS, plus the minimum and maximum where recorded (currently only Medium has both). For the other presets, only the average is available. Notably, the Medium row consistently shows the lowest frame floor at each resolution, which suggests it is the stablest preset in the data. At 1080p, the average 132 FPS contains absolute lows above 112 FPS; at 1440p, the minimum stays above 99 FPS; and at 4K, the minimum is still 63 FPS, which is higher than the average of High at 4K (57). This implies that the Medium preset avoids frame dips that would otherwise place windows with the high or ultra, which are not measured for external events. Fine-grained variance data for other presets is absent, but the Medium pattern—combined with the high maximum—makes Medium the stable choice for the biggest majority of monitor refresh rates.

From the table, the exact average values demonstrate a monotonic descent with rising quality settings at every resolution: Low > Medium > High > Ultra. The gaps between Low and Medium, Medium and High, and High and Ultra vary per resolution, but the ranking is always the same. Notably, the gap between High and Ultra is small: at 1080p it’s only 2 FPS, 5 FPS at 1440p, and 3 FPS at 4K. Meanwhile, the drop from Medium to High is larger: 12 FPS at 1080p, 20 FPS at 1440p, and 18 FPS at 4K. So the Ultra preset gains little over High, while the Medium-to-High reduction is the major jump. For nominal smoothness, Medium yields the minimum FPS above 60 in every tested configuration; only the 4K High and Ultra fall short, at 57 and 54 FPS, respectively.

The aggregate summary: the combo reaches triple-digit frame rates on Over High at up to 144p, and it only begins to stress the GPU when the resolution is pushed to 4K. Given these measurements, the i5-11400F + RTX 3080 setup in Thief is capable of strong performance at standard and 1440p displays, and needs a Medium preset to hold comfortable frame also on 4K display.

Hardware Specifications

Intel Core i5-11400F

Cores / Threads 6 / 12
Base Clock 2600 MHz
Boost Clock 4400 MHz
TDP 65W
Socket Intel Socket 1200
View Full CPU Details →

NVIDIA GeForce RTX 3080

VRAM 10 GB GDDR6X
Base Clock —
Boost Clock 1710 MHz MHz
TDP 320 WW
View Full GPU Details →

FPS Benchmarks by Resolution & Settings

Performance data for Intel Core i5-11400F + NVIDIA GeForce RTX 3080 in Thief

Resolution Settings Avg FPS
3840x2160 Low 98.0
3840x2160 Medium 75.0
3840x2160 High 57.0
3840x2160 Ultra 54.0
2560x1440 Low 129.0
2560x1440 Medium 117.0
2560x1440 High 97.0
2560x1440 Ultra 92.0
1920x1080 Low 146.0
1920x1080 Medium 132.0
1920x1080 High 120.0
1920x1080 Ultra 118.0

Thief with ii5-11400F + Other GPUs

See how Thief performs with the same CPU but different graphics cards.

Thief with RTX 3080 + Other CPUs

See how Thief performs with the same GPU but different processors.

Other Games with ii5-11400F + RTX 3080

Explore FPS benchmarks for other games using this same hardware combination.