Valorant

The pairing of AMD’s Ryzen 9 9900X and NVIDIA’s GeForce RTX 4090 produces a fascinating dataset in Valorant, a game where frame rates are already astronomically high. The measured results reveal a system that is almost entirely unconstrained by traditional bottlenecks, yet the scaling patterns across resolutions and settings still tell a clear story about where the workload resides. With a combo rank of 12 out of 1,650 tested combinations, this setup sits firmly in the elite tier, but the data suggests that its absolute ceiling is dictated more by the CPU’s single-thread velocity than the GPU’s raw throughput.

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

The transition from 1920x1080 to 3840x2160 produces a surprisingly modest frame rate decline, which is the first clue that the GPU is not the primary constraint. At Low settings, the average FPS falls from 711.2 at 1080p to 625.5 at 4K, a drop of just 12%. This is a remarkably shallow curve for a resolution quadrupling pixel count. The RTX 4090, with its 24 GB of GDDR6X memory and 1.01 TB/s of bandwidth, is so far ahead of Valorant’s requirements that the increased pixel workload barely registers.

Moving from 1080p to 1440p at Low settings, the average FPS only dips from 711.2 to 674.4, a 5.2% reduction. The subsequent jump to 4K brings another 7.2% reduction. This pattern—small, linear declines rather than a cliff—indicates that the rendering pipeline is not stalling on pixel throughput. If the GPU were the limiting factor, the 4K score would likely be a fraction of the 1080p number, not 88% of it.

At High settings, the scaling is similar but slightly more pronounced. The average FPS goes from 595.1 at 1080p to 563.1 at 1440p (5.4% drop) and then to 526.9 at 4K (6.4% drop from 1440p). The Ultra preset follows the same trajectory: 587 at 1080p, 556.4 at 1440p, and 513.8 at 4K. The consistency of these deltas across all settings suggests that resolution scaling in Valorant is a secondary factor; the game’s engine is not heavily taxing the GPU’s shading units or texture fill rate at any of these configurations.

The most telling insight is the gap between Low and Ultra at the same resolution. At 4K, Low produces 625.5 FPS while Ultra produces 513.8 FPS, a difference of 21.7%. At 1080p, the same comparison yields 711.2 versus 587, an 18.4% gap. The fact that settings have a larger relative impact than resolution indicates that the game’s visual effects (shadows, lighting, post-processing) are more demanding than raw pixel count. However, even the lowest 4K Ultra score of 513.8 FPS is far beyond any display’s refresh rate capability, making the entire scaling curve more academic than practical.

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

The Ryzen 9 9900X is a 12-core, 24-thread processor based on the Zen 5 architecture, with a base clock of 4.40 GHz and a boost clock of 5.60 GHz. In a game like Valorant, which is notorious for its reliance on a few fast threads, the multi-core configuration is largely a spectator. The benchmark data confirms this: the CPU’s 3DMark single-thread score is 1286, while its 2-thread score is 2519, showing nearly perfect scaling across two threads. However, the 4-thread score of 4886 and 8-thread score of 9114 show diminishing returns, and the max-thread score of 13929 is only 12% higher than the 8-thread result.

Valorant’s frame rates in this dataset are so high that the game’s main thread is likely the bottleneck. The CPU’s Cinebench R23 single-core score of 6555 is exceptional, placing it in the top percentile of all CPUs (94th percentile overall). This single-thread prowess is what enables the 711.2 FPS average at 1080p Low. The data implies that increasing the core count beyond what is needed for the game’s logic and render thread yields no additional FPS; the 12 physical cores are more than sufficient, and the remaining cores are idle while two or three do the heavy lifting.

Interestingly, the CPU’s nearest rival is the AMD Ryzen 9 9900X3D, which scores 54681 on average versus the 9900X’s 54450, a delta of -0.4%. This suggests that the extra 3D V-Cache in the rival part provides almost no benefit in synthetic benchmarks, and in Valorant, the 9900X’s high boost clock is likely the more relevant factor. The Intel Core i7-14700F trails by 0.7%, while the Core Ultra 5 245K leads by 1.5%, indicating that at this performance tier, the CPU choice has minimal impact on the final frame count. The measured FPS figures are so close across settings that the CPU’s 5.60 GHz boost clock appears to be the single most important specification for this game.

The CPU’s 64 MB of shared L3 cache is also noteworthy. Valorant’s maps and entity data are relatively small, so the large cache likely helps maintain consistent frame pacing by keeping frequently accessed data on-die. The processor’s 89.6 GB/s of dual-channel DDR5 memory bandwidth is more than adequate for the game’s data streaming needs, and the PCIe Gen 5 interface (24 lanes) ensures no bottleneck when communicating with the GPU, even at 4K.

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

The RTX 4090 is an extreme overkill for Valorant, but the data shows how its specifications still influence the outcome. With a boost clock of 2520 MHz and 16,384 shading units, the GPU has a theoretical FP32 throughput of 82.58 TFLOPS. Yet, in this game, the GPU’s utilization is likely very low, as evidenced by the minimal FPS differences between Low and Ultra settings. At 1080p, the gap between Low and Ultra is 124.2 FPS (711.2 vs 587), which is 17.5% lower for Ultra. This is a settings-driven cost, not a resolution-driven one.

The 24 GB of GDDR6X memory on a 384-bit bus is far beyond what Valorant requires. At 4K Ultra, the game likely uses less than 8 GB of VRAM, so the GPU’s 1.01 TB/s bandwidth is never stressed. The pixel rate of 443.5 GPixel/s and texture rate of 1,290.2 GTexel/s are similarly overprovisioned. The data indicates that the GPU’s role is to render frames as fast as the CPU can feed them, and the 4090’s massive compute resources are simply not engaged by this title.

The GPU’s nearest rival in the benchmark database is the NVIDIA Tesla T4, with a delta of -0.4%, which is surprising given the massive architectural and performance gap between a data-center inference card and a flagship gaming GPU. This suggests that the average benchmark scores for the RTX 4090 are heavily weighted by synthetic tests that do not reflect Valorant’s workload. In practice, the 4090’s 128 RT cores and 512 tensor cores are irrelevant here; the game does not use ray tracing or DLSS in these measurements.

The GPU’s end-of-life production status and 450 W TDP are specs, but the measured FPS shows that the card is never the limiting factor. The fact that the 4K High score (526.9) is only 11.5% lower than the 1080p High score (595.1) confirms that the GPU is idling, waiting for draw calls from the CPU. The 4090’s 5 nm process node and 76,300 million transistors provide immense headroom, but Valorant simply cannot utilize it.

FAQ — 4-6 Q&A pairs answerable from FACT PACK data

Q: What is the maximum average FPS achieved by this combo?

A: The highest average FPS is 711.2, recorded at 1920x1080 with Low settings.

Q: How does the combo rank against other tested configurations?

A: It ranks 12th out of 1,650 tested combos in Valorant, placing it in the top 1% of all pairings.

Q: Does increasing resolution from 1080p to 4K significantly hurt performance?

A: No. At Low settings, the FPS drops from 711.2 to 625.5, a 12% decrease, which is minor given the 4x pixel count increase.

Q: Which is more impactful on FPS: resolution or settings?

A: Settings have a larger impact. At 4K, Low yields 625.5 FPS while Ultra yields 513.8 FPS, a 21.7% gap. The resolution gap from 1080p to 4K at Low is only 12%.

Q: Is the CPU or GPU the bottleneck in this game?

A: The CPU is the primary constraint. The GPU’s score changes little across resolutions, while the CPU’s single-thread performance (Cinebench R23 score of 6555) appears to set the frame rate ceiling.

Q: How does the Ryzen 9 9900X compare to its closest rival in synthetic benchmarks?

A: It scores 54450 on average, which is 0.4% lower than the Ryzen 9 9900X3D (54681) and 0.7% higher than the Intel Core i7-14700F (54097).

How This Combo Ranks — use comboRankInGame vs other tested combos

The combo’s rank of 12 out of 1,650 tested combinations places it in the 99.3rd percentile of all configurations. This is an exceptional position, but the data suggests that the ranking is more a function of the CPU’s single-thread speed than the GPU’s dominance. The 11 combos ahead of this one likely feature CPUs with even higher boost clocks or more efficient architectures for Valorant’s engine. The Ryzen 9 9900X’s 94th percentile ranking among all CPUs indicates it is near the top, but not the absolute best, for this workload.

The GPU’s 91st percentile ranking among all GPUs is slightly lower than the CPU’s, yet in Valorant, the GPU’s rank is almost irrelevant. The measured FPS at 1080p Low (711.2) is likely near the engine’s practical limit, meaning that any combo ranking above this one is achieving marginal gains, possibly through faster memory or a CPU with a higher single-thread score. The 0.4% deficit to the Ryzen 9 9900X3D in CPU benchmarks is not enough to explain a ranking difference, so the top 11 combos must leverage CPUs with superior single-core performance, such as a higher-binned chip or one with a more aggressive boost algorithm.

The data also shows that the combo’s rank is consistent across settings. At 4K Ultra, the FPS is 513.8, which is still likely higher than most other combos at 1080p. This means the rank of 12 is not fragile; even a demanding setting like 4K Ultra keeps this combo in the upper echelon. The performance gap to the 1st place combo is unknown from the data, but given the shallow scaling curves, it is likely less than 10% at any resolution.

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

At 1920x1080, the measured average FPS values are: 711.2 for Low, 606.1 for Medium, 595.1 for High, and 587 for Ultra. The spread from Low to Ultra is 124.2 FPS, representing an 17.5% reduction. This is the largest relative spread across all resolutions, indicating that at 1080p, the CPU is so fast that the GPU’s settings work has a more noticeable impact.

At 2560x1440, the averages are 674.4 for Low, 574.4 for Medium, 563.1 for High, and 556.4 for Ultra. The Low-to-Ultra gap is 118 FPS, or 17.5%, nearly identical to 1080p. The resolution increase from 1080p to 1440p costs about 5.2% at Low (711.2 to 674.4) and 5.4% at High (595.1 to 563.1), showing a consistent scaling factor.

At 3840x2160, the averages are 625.5 for Low, 530.7 for Medium, 526.9 for High, and 513.8 for Ultra. The Low-to-Ultra gap is 111.7 FPS, or 17.9%. The resolution increase from 1440p to 4K costs about 7.2% at Low (674.4 to 625.5) and 6.4% at High (563.1 to 526.9). The 4K Low score of 625.5 is still higher than the 1080p Ultra score of 587, demonstrating that the combo’s performance headroom is massive.

The pattern across all rows is clear: Medium, High, and Ultra settings cluster closely together, while Low stands apart. At 1080p, Medium (606.1) is only 1.8% higher than High (595.1), and High is only 1.4% higher than Ultra (587). This suggests that the Medium preset is the sweet spot for visual quality, as it captures nearly all of the performance benefit of Low while offering better graphics.

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

Given the measured data, the Low preset offers the highest frame rates, but those numbers are already far beyond any display’s refresh rate. At 1080p Low, the combo produces 711.2 FPS, which is unnecessary for a 144Hz or even 360Hz monitor. The Ultra preset at 1080p produces 587 FPS, which is still 63% higher than a 360Hz display’s maximum. Therefore, players should prioritize visual fidelity over raw FPS, as the performance cost is negligible in absolute terms.

The Medium preset at 4K produces 530.7 FPS, which is 3.3% lower than Low (625.5) but 2.3% higher than Ultra (513.8). This makes Medium the balanced choice for 4K users who want improved shadows and effects without sacrificing frame rate. At 1440p, Medium (574.4) is 1.9% slower than Low (674.4) but 3.2% faster than Ultra (556.4), offering the same relative advantage.

For competitive play, the Low preset is traditionally favored to reduce visual clutter and input lag. However, the data shows that the difference between Low and Ultra at 1080p is only 124.2 FPS, which is a 17.5% reduction. Since the absolute values are so high (587 FPS at Ultra), the gameplay experience will be identical in terms of smoothness. The recommendation is to use Ultra at 1080p or 1440p if the display supports high refresh rates, as the GPU has ample headroom. At 4K, Medium is the better choice, as it provides 530.7 FPS versus 513.8 for Ultra, a 3.3% improvement with minimal visual compromise. The High preset offers no significant advantage over Medium, as the scores are nearly identical (526.9 vs 530.7 at 4K), making High the least efficient setting in the dataset.