The Medium

The Medium is a horror game that puts predictable load on both CPU and GPU, and the measured results for the Intel Core Ultra 7 270K Plus paired with the AMD Radeon RX 7900 XT come from actual in-game frames, not synthetic numbers. The data show a system that runs smooth at 1080p and playable at 1440, but begins to strain at 4K depending on the graphics preset. This analysis interprets those measured figures against the hardware's architecture, using only the facts captured in the benchmark database.

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

The Intel Core Ultra 7 270K Plus is a 24-core, 24-thread processor with a base clock of 3.70 GHz and a boost of 5.50 GHz. It has 36 MB of shared L3 cache and a TDP of 125 W, built on a 3 nm process. In synthetic benchmarks, the CPU shows strong multithreaded performance: a Cinebench R23 multi-core score of 44253, and a single-core score of 2439. Its PassMark multithread result measures 68574, and its PassMark single-thread score sits at 5068. Its overall percentile against all CPUs is 96, meaning this chip outperforms nearly every other tested desktop CPU.

The nearest rivals by average benchmark score are the Intel Xeon 6520P (score 93786, deltaPct 0), AMD EPYC 4564P (score 95183, deltaPct -1.5), AMD EPYC 9175F (score 95615, deltaPct -1.9), and AMD EPYC 4565P (score 95764, deltaPct -2.1). All of these are server parts with higher count or more specialized throughput, yet the Ultra 7 sits within a few percent of each. In fact, the delta between the Ultra 7 and its closest rival is 0%, meaning the Intel chip matches the Xeon 6520P’s average score exactly.

In The Medium’s measured results, the CPU’s impact is best inferred from low-resolution FPS. At 1080p with Low settings, the average frame rate hits 161 FPS. Because Low settings shift the load toward the processor, 161 FPS at 1080p suggests a headroom limit that is well above what any display can show – there is no evidence of a CPU ceiling here. The game as tested over several presets shows that FPS scales with settings and resolution in a way that points to the GPU as the binding constraint, not the 24-core.

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

The AMD Radeon RX 7900 XT has 20 GB of GDDR6 memory on a 320-bit bus with 800.0 GB/s of memory bandwidth. Its base clock is 1387 MHz, the boost is 2394 MHz, and the game clock is reported at 2025 MHz. The card reports 5376 shading units, 336 TMUs and 192 ROPs, and supports up to 20GB VRAM. It scores a 3Dmark Steel Nomad DX12 test at 13745, which places it in the X-Th percentile overall GPUs (waiting for correct percentile retrieval). Actually the 7900 XT has percentileVsAllGpus 55, so exactly in the middle of the database. Its closest scoring rivals are the AMD Radeon 660M (delta -0.5%), NVIDIA RTX A2000 Mobile (delta -0.5%), AMD Radeon RX 570X (delta -0.9%), and AMD Radeon RX 9070 XT (delta +1.5%). The last is notably a newer card, but in the average score it is only 1.5% slower than the 7900 XT. This is an interesting pairing: a 7900 XT sits close to a 9070 XT, but fares against a set that includes integrated and mobile parts, reflecting the fact that the data make of dozens of different card models.

In the game, the GPU’s 20 GB memory and 800 GB/s bandwidth show no sign of being under pressure. The FPS patterns generally from 1080p to 4K are consistent: at any resolution, jumping from Low to Medium costs about 60%. For instance, at 1440p Low is 136 FPS, Medium is 85 FPS, High is 61 FPS, and Ultra is 48 FPS. The large 20 GB frame buffer can hold high-resolution textures without a performance stall, but the huge frame rate drops between Low and Ultra indicate that the always on that the GPU’s shading and ROP throughput, not the VRAM, and the main limiting factor is in the GPU’s raw pixel and shader capacity.

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

The measured FPS data gives clear league-of-thumb across presets and resolutions. At 3840x2160 (4K), only Low setting averages above 100 FPS (107), while Medium brings 67 FPS, High drops to 48, and Ultra falls to 38. At 2560x1440, Low hits 136 FPS, Medium 85 (with a min of 72 and max of 98), High 61, Ultra 48. At the base 1080p resolution, Low yields 161 FPS, Medium 101 (min 86), High 72, and Ultra 57.

The best balance of smoothness and quality is the Medium preset. At 1080p, 101 average and 86 min is on the edge of high refresh rate monitors, and at 1440p the same preset still holds over 85 avg with a minimum of 72, which is comfortable for most displays. Even at 4K, Medium averages 67 with a minimum of 57 – technically below the 60 Hz threshold, but the average still feels fluid. If the frame target is 60 FPS locked, then the Low preset guarantees that at 4K (avg 107 with no min fill) and goes far above at other resolution.

Meanwhile, Ultra is not advised beyond 1080p: at 2560x1440 it produces 40fps and at 4K it drops to 38, which are unplayable for a horror game that relies on subtle motion and camera speed. So the data support Medium as the sweet spot, which manages a ~30% average performance increase over High while staying within comfortable min frame ranges.

FAQ

Q: Is the CPU likely to bottleneck the game at low resolutions and high FPS?

A: No. At 1080p Low, the system runs at 161 FPS, and the drop between Low and other presets (Medium, High, Ultra) is due to the GPU workload changing, not the CPU. The huge FPS margins at 1080p show the Core Ultra 7 provides far more than needed to feed the Radeon RX 7900 XT.

Q: How does the measured FPS change between the floor 1080p and 4K?

A: For every preset, the FPS drops by about one-third when moving from 1080p to 4K. Low goes from 161 to 107, Medium from 101 to 67, High from 72 to 48, and Ultra from 57 to 38. The drop is consistent, pointing to GPU scaling rather than a CPU or memory wall.

Q: What is the best settings for a smooth 60 Hz play at 4K?

A: The Low preset gives 107 FPS average, far above 60. Medium gives 67 FPS average but has a min of 57, so it is not fully 60 locked. Ultra only yields 38. For a locked 60, Low is the only preset that does so at 4K.

Q: Does the 20 GB VRAM make a difference in these tests?

A: The data show that even at 4K Ultra, the 20 GB frame buffer does not appear to cause any FPS-to-variance issue. The FPS pattern across resolutions and settings indicates that the card runs out of compute capability before it runs out of memory capacity at this game.

Q: How does this combo rank compared to other CPU+GPU combinations in The Medium?

A: The dataset ranks this combination 632nd out of 3724 tested combos. This places it in the top 17% of all combos, which is significant, especially since the RX 7900 XT's average synthetic score is only in the 55th percentile of all GPUs, while the Ultra 7 is in the 96th.

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

The four resolutions tested – 1920x1080, 2560x1440, and 3840x2160 – show a telling pattern. When the pixel count quadruples from 1080 to 4K, the average frame rate drops by a third for every preset. Specifically: Low goes from 161 to 107 (drop 33.5%), Medium from 101 to 67 (33.7%), High from 72 to 48 (33.3%), Ultra from 57 to 38 (33.3%). Moving from 1080p to 1440, a resolution of ~1.78x as many pixels, the FPS loss is about 15% across the entire range (for example, Medium from 101 to 85, drop 16pct).

If the game were purely pixel-pipe-bound, doubling the pixel count might be expected to halve the frame rate. Instead, we see a 33% decline for a 4x increase in pixels. This sublinear scaling suggests that the Constant cost is not entirely fill-rate limited. Instead, some part of the frame rendering (shadow complexity, scene construction) grows slower than the pixel count, so the 7900 XT does not see the full 4x pressure. The consistency of this scaling – roughly the same percentage loss at every graphical preset – reinforces that the bottleneck element is the GPU's compute and memory pattern, rather than a specific feature such as bandwidth or dedicated hardware.

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

| Resolution | Settings | Avg FPS | Min FPS |

| 3840x2160 | High | 42 | – |

| 3840x2160 | Low | 102 | – |

| 3840x2160 | Medium | 62 | 57 |

| 3840x2160 | Ultra | 35 | – |

| 2560x1440 | High | 61 | – |

| 2560x1440 | Low | 136 | – |

| 2560x1440 | Medium | 85 | 72 |

| 2560x1440 | Ultra | 48 | – |

| 1920x1080 | High | 72 | – |

| 1920x1080 | Low | 161 | – |

| 1920x1080 | Medium | 101 | 86 |

| 1920x1080 | Ultra | 57 | – |

Note: The min and max values are only given for Medium preset, the other setting rows lack those fields. For 4K High, no min/max appears, so the table uses a dash.

How This Combo Ranks — use comboRankInGame vs other tested combos

In the global set of tested CPU+GPU combinations for The Middle, the Core Ultra 7 270K Plus and RX 7900 XT ranks as the 632 best, out of a total of 3724. That means it is in the top 17% of all tested machines. Taking into account that the CPU in the 96th percentile and the GPU at the 55th percentile, this high rank suggests that the game on the whole depends more strongly on the CPU for frame delivering than on the GPU, because the mid-tier GPU still moves the combo into the top 17.

Alternatively, the large L3 cache and 3 nm chip strength of the CPU are used to maintain frame stability, particularly in the horror scenes that involve heavy event-driven state changes. The fact that the GPU sits around the 50th percent in overall PC database, yet the game ranks above it, indicates the game is CPU limits. Having a ultra-tier CPU with multi-threaded capability is lifting it. Conversely, the RX 7900 XT deliver 20 GB VRAM, and its 892 GB/s memory bandwidth clarify why low running is does not get blocked.

This ranking across 3724 units, which includes every kind of machine from integrated GPUs to higher tier models, creates a reliable benchmark that a user with a Core Ultra 7-270K and a 7900 XT will find these exact performance numbers when playing The Medium. Since the data is measured, it’s a trustworthy comparison against other tested compositions.