The Medium

The pairing of the Intel Core i5-9400 with the NVIDIA GeForce RTX 3080 Ti presents a fascinating study in bottleneck dynamics within The Medium. While the GPU is a high-end Ampere part with a substantial 12 GB frame buffer, the data indicates that the 6-core, 6-thread Coffee Lake CPU often limits performance, particularly at lower resolutions. This analysis breaks down the measured FPS rows to understand where the system excels and where it falls short.

Resolution Scaling

The measured FPS data reveals a distinct pattern of scaling across the three tested resolutions. At 1080p, the system achieves its highest raw frame rates, but the gap between settings is relatively narrow. For instance, moving from Low (87 FPS) to Ultra (53 FPS) at 1080p results in a 34 FPS drop. At 1440p, the spread widens slightly, with Low at 82 FPS and Ultra at 45 FPS, a 37 FPS difference. The most significant change occurs at 4K, where the Ultra preset falls to 35 FPS and High dips to 44 FPS, while Low still manages 76 FPS.

This scaling behavior suggests the CPU is a dominant factor at 1080p and 1440p. The i5-9400, which sits at the 47th percentile among all CPUs, appears to cap the frame rate before the RTX 3080 Ti can fully stretch its legs. The fact that 1080p Low (87 FPS) is only 5 FPS faster than 1440p Low (82 FPS) is telling; a GPU-bound scenario would typically show a larger delta when pixel count drops by over 40%. The modest gains from lowering resolution indicate the processor is the limiting component at these settings.

As resolution increases to 4K, the tables turn. The 4K Ultra average of 35 FPS is a 34% reduction from the 1440p Ultra score of 45 FPS. This steep drop confirms that the GPU becomes the primary bottleneck at 4K, as the pixel throughput demand finally exceeds what the RTX 3080 Ti can deliver in this title. The 4K Low result of 76 FPS is noteworthy—it equals the 1440p Medium score (68 FPS) plus 8 FPS, showing that even at 4K with Low settings, the CPU still has headroom to feed the GPU adequately.

The most telling statistic is the 4K Medium result, which shows an average of 61 FPS with a minimum of 52 and maximum of 71. This 19 FPS variance between the min and max indicates frame pacing instability, a common symptom when both CPU and GPU are under heavy load but neither is fully dominant. At 1080p Medium, the variance is smaller (62 to 83 FPS), suggesting the CPU is more consistently the limiting factor there.

GPU Role

The RTX 3080 Ti is a formidable piece of silicon, built on the Ampere architecture with a GA102 chip. Its 10,240 shading units and 112 ROPs are backed by 12 GB of GDDR6X memory on a 384-bit bus, yielding a bandwidth of 912.4 GB/s. These specifications place the card at the 83rd percentile among all GPUs, with an average benchmark score of 41,187. The nearest rival, the AMD Radeon Pro 5300, scores 40,870, putting the 3080 Ti just 0.8% ahead, while the NVIDIA Tesla M40 24 GB trails by 1.2%.

In The Medium, the GPU's role is heavily modulated by the CPU's capabilities. At 4K Ultra, the card is clearly the bottleneck, as the 35 FPS average is well below what the card's raw compute would suggest. The FP32 performance of 34.10 TFLOPS and texture rate of 532.8 GTexel/s are more than sufficient for most scenes, but the game's fixed render path seems to limit utilization. The 4K High result of 44 FPS further confirms this, as the GPU is still struggling with the pixel count despite the less demanding settings.

The memory subsystem plays a critical role at higher resolutions. The 12 GB frame buffer is ample for 4K textures, and the 912.4 GB/s bandwidth prevents any memory-related stutters in the measured data. However, the card's boost clock of 1665 MHz, while respectable, is not the differentiator here. The core clock scaling from 1365 MHz base to 1665 MHz boost appears adequate, but the data suggests the game is not purely GPU-core limited at 4K either.

The most revealing comparison is the 4K Low result of 76 FPS versus the 1080p Ultra result of 53 FPS. This 23 FPS gap shows that even with dramatically reduced GPU load at 4K Low, the CPU still cannot push the frame rate significantly higher than 76 FPS. This suggests the i5-9400 has a practical ceiling around 80-90 FPS in this game, regardless of GPU headroom. The GPU's role, therefore, is to provide the visual fidelity headroom—the 12 GB VRAM and 384-bit bus allow for Ultra textures at 4K without crashing, but the actual frame rate is capped by other factors.

FAQ

Q: Why is the 4K Ultra frame rate so low compared to 4K Low?

A: The 4K Ultra average of 35 FPS versus the 4K Low average of 76 FPS represents a 41 FPS drop. This massive delta indicates that the Ultra preset introduces rendering features that heavily tax the GPU's shading units and ROPs, moving the bottleneck firmly onto the RTX 3080 Ti.

Q: Is the RTX 3080 Ti being fully utilized at 1080p?

A: No. The 1080p Low score of 87 FPS and 1080p Ultra score of 53 FPS are both lower than what the GPU's 83rd percentile ranking would suggest. The i5-9400, with its 6 cores and 6 threads, appears to limit the frame rate, preventing the GPU from reaching its full potential at lower resolutions.

*Q: What is the best resolution for this combo in The Medium?*

A: The data shows 1440p Medium provides a balanced experience with an average of 68 FPS and a minimum of 58 FPS. This is above the 60 FPS threshold while maintaining reasonable visual quality, and the frame variance (58-79 FPS) is stable enough for smooth gameplay.

Q: How does the 4K Medium result compare to the 1440p Ultra result?

A: The 4K Medium average of 61 FPS is 16 FPS higher than the 1440p Ultra average of 45 FPS. This indicates that lowering settings at 4K yields better performance than raising settings at a lower resolution, suggesting the GPU is the limiting factor at high pixel counts.

Q: Does the CPU bottleneck disappear at 4K?

A: Partially. At 4K Low, the system hits 76 FPS, which is close to the 1080p Low score of 87 FPS. The 11 FPS gap shows the CPU still has some influence, but the 4K Ultra result of 35 FPS demonstrates that the GPU becomes the primary constraint when settings are maxed out.

Q: Why is there no min/max FPS data for most settings?

A: The measured FPS table only includes min and max values for the Medium preset at each resolution. This suggests the testing methodology captured frame time variance only for that specific preset, while other settings were measured for average FPS only.

How This Combo Ranks

The combination of the i5-9400 and RTX 3080 Ti ranks 1619th out of 3,724 tested combos in The Medium, placing it in the 43rd percentile of all configurations. This ranking is notably lower than what the GPU's individual 83rd percentile would suggest, and it is closer to the CPU's 47th percentile. This discrepancy is a direct result of the CPU bottleneck observed at lower resolutions.

The ranking tells a clear story: this is a system that is held back by its processor in this specific game. The i5-9400's nearest rivals in synthetic benchmarks include the Intel Core i7-1068NG7 (0.2% slower), the Intel Core i7-10710U (0.3% faster), and the AMD Ryzen 5 PRO 1500 (0.4% faster). These are all similarly performing chips, none of which are known for high-end gaming throughput. The CPU's average benchmark score of 2,254 places it just below the 2,259 of the i7-1068NG7 and above the 2,246 of the Ryzen 5 PRO 1500.

The GPU's individual ranking is far more impressive. With an average benchmark score of 41,187, the RTX 3080 Ti sits just 0.8% above the Radeon Pro 5300 (40,870) and 1.2% above the Tesla M40 24 GB (41,707). The nearest rival, the RTX 5070, scores 40,377, placing it 2% below the 3080 Ti. This suggests the GPU is more than capable of handling demanding titles, but its potential is squandered in this pairing.

The 1619th rank out of 3,724 combos means the system outperforms roughly 56% of tested configurations. This is not a poor result, but it is underwhelming for a GPU that was launched with a 1,199 USD MSRP. The data indicates that upgrading the CPU would likely yield significant improvements in The Medium, particularly at 1080p and 1440p where the processor bottleneck is most pronounced.

Settings Recommendations

Based on the measured FPS data, the optimal settings for this combo depend heavily on the target resolution and the user's preference for frame rate versus visual fidelity. The data provides four presets at each resolution, allowing for a clear recommendation.

At 1080p, the Medium preset delivers an average of 72 FPS with a minimum of 62 and maximum of 83. This is the only 1080p setting with min/max data, and it provides a smooth experience above the 60 FPS threshold. The High preset at 64 FPS is also playable, but the 8 FPS drop from Medium suggests diminishing returns. The Low preset at 87 FPS offers the highest frame rate but sacrifices visual quality. The Ultra preset at 53 FPS is not recommended, as the 19 FPS drop from High is substantial for marginal graphical improvements.

At 1440p, the Medium preset again stands out with an average of 68 FPS and a minimum of 58 FPS. This is nearly as fast as the 1080p High result, indicating that the resolution increase is well-tolerated. The High preset at 56 FPS is borderline, while the Low preset at 82 FPS is excellent for competitive play. The Ultra preset at 45 FPS is too low for a smooth experience and should be avoided.

At 4K, the Medium preset is the clear winner with an average of 61 FPS and a minimum of 52 FPS. This is the only 4K setting that approaches 60 FPS, and the variance (52-71 FPS) is acceptable for most users. The Low preset at 76 FPS is the fastest option but looks noticeably worse. The High preset at 44 FPS and Ultra at 35 FPS are both below the 45 FPS threshold that most consider the minimum for playable frame rates in a horror title.

The overall recommendation is to use the Medium preset at any resolution. It provides the best balance of frame rate and visual quality, and it is the only setting with measured minimum and maximum FPS data, indicating the testing methodology found it to be the most stable. For users with 1440p monitors, this is the sweet spot. For 4K users, the Medium preset is the only viable option that maintains playable performance.

Measured FPS Breakdown

The following table summarizes the exact measured FPS data for all tested combinations:

| Resolution | Settings | Avg FPS | Min FPS | Max FPS |

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

| 3840x2160 | High | 44 | - | - |

| 3840x2160 | Low | 76 | - | - |

| 3840x2160 | Medium | 61 | 52 | 71 |

| 3840x2160 | Ultra | 35 | - | - |

| 2560x1440 | High | 56 | - | - |

| 2560x1440 | Low | 82 | - | - |

| 2560x1440 | Medium | 68 | 58 | 79 |

| 2560x1440 | Ultra | 45 | - | - |

| 1920x1080 | High | 64 | - | - |

| 1920x1080 | Low | 87 | - | - |

| 1920x1080 | Medium | 72 | 62 | 83 |

| 1920x1080 | Ultra | 53 | - | - |

At 1080p, the system scales from 87 FPS at Low to 53 FPS at Ultra, a 34 FPS spread. The Medium preset's 72 FPS average with a 62 FPS minimum provides a stable experience. The High preset at 64 FPS is 11% slower than Medium, while Ultra is 17% slower than High.

At 1440p, the Low preset drops to 82 FPS, a 5 FPS reduction from 1080p Low. The Medium preset at 68 FPS is 4 FPS slower than its 1080p counterpart, and the High preset at 56 FPS is 8 FPS slower. The Ultra preset at 45 FPS is 8 FPS slower than 1080p Ultra, showing a consistent degradation.

At 4K, the Low preset at 76 FPS is 6 FPS slower than 1440p Low. The Medium preset at 61 FPS is 7 FPS slower than 1440p Medium, and the High preset at 44 FPS is 12 FPS slower than 1440p High. The Ultra preset at 35 FPS is 10 FPS slower than 1440p Ultra, representing the largest performance cliff in the entire dataset.

The 4K Medium result of 61 FPS is particularly noteworthy, as it outperforms the 1440p High result of 56 FPS. This counterintuitive finding suggests that the Medium preset at 4K is less demanding than the High preset at 1440p, possibly due to different texture streaming or post-processing requirements. The min FPS of 52 at 4K Medium is still above the 50 FPS threshold for acceptable frame pacing, making it the most reliable 4K option.