Valorant

FAQ

Q: What is the average FPS for this AMD Ryzen 9 9900X + NVIDIA GeForce RTX 5090 D combo in Valorant at 1080p with Low settings?

A: At 1920x1080 with Low settings, this combo achieves an average of 711 FPS. This is the highest average FPS recorded across all tested resolution and settings combinations in this data set.

Q: How does the CPU's average benchmark score compare to its closest rival?

A: The AMD Ryzen 9 9900X has an average benchmark score of 57498. Its closest rival, the AMD EPYC 9015, scores 57555, which places the Ryzen 9 9900X just 0.1% behind. Another rival, the AMD EPYC 7313, scores 57399, placing it 0.2% ahead of the 9900X.

Q: What is the combo's ranking among tested systems for this game?

A: In Valorant, this specific CPU+GPU combination is ranked 48th out of 2883 tested combos. This puts it in the top 1.7% of all systems in the database for this game.

Q: What is the difference in average FPS between the Low and Ultra presets at 4K?

A: At 3840x2160, the Low preset averages 626 FPS, while the Ultra preset averages 516 FPS. This represents a performance drop of 110 FPS, or approximately 17.6%, when moving from the lowest to the highest quality preset at 4K resolution.

Q: Does the GPU have enough memory bandwidth for high-resolution gaming?

A: The GPU features 32 GB of GDDR7 memory on a 512-bit bus, providing a memory bandwidth of 1.79 TB/s. This exceptionally high bandwidth appears well-suited for the demands of 4K gaming, as the combo still maintains over 500 FPS at 4K Ultra.

Q: What are the boost clock speeds for the CPU and GPU in this combo?

A: The AMD Ryzen 9 9900X has a boost clock of 5.60 GHz across its 12 cores. The NVIDIA GeForce RTX 5090 D has a boost clock of 2407 MHz.

Settings Recommendations

The measured data provides a clear picture of how settings impact performance in Valorant for this high-end combo. At every resolution, the Low preset delivers the highest average FPS, followed by Medium, then High, with Ultra producing the lowest averages. This is the expected hierarchy, but the magnitude of the differences is revealing.

At 1080p, the gap between Low and Ultra is 125 FPS (711 vs 586). At 1440p, the gap narrows slightly to 118 FPS (674 vs 556). At 4K, the gap is the smallest at 110 FPS (626 vs 516). This trend shows that as resolution increases, the absolute frame rate penalty for using Ultra settings actually decreases, suggesting the GPU becomes more of a limiting factor at higher resolutions, reducing the relative impact of CPU-driven settings changes.

For competitive play, the Low preset appears to be the optimal choice. At 1080p Low, the system pushes 711 FPS, which is 20% higher than the High preset at the same resolution (596 FPS). The Low preset likely minimizes input latency and maximizes refresh rate headroom, which is critical for a fast-paced FPS like Valorant. The data shows that even at 4K Low, the system delivers 626 FPS, which exceeds the refresh rate of virtually all monitors on the market.

However, for players who prioritize visual fidelity, the Medium preset offers a compelling balance. At 4K Medium, the average FPS is 534, which is only 8 FPS lower than High but 92 FPS higher than the drop to Ultra would suggest. The Medium preset also provides the only complete min/max FPS data in the set — at 1080p, it shows a minimum of 515 FPS and a maximum of 697 FPS, indicating a narrower frame time distribution that could translate to more consistent pacing.

The Ultra preset, while visually demanding, still delivers over 500 FPS at all resolutions. The data suggests that for this particular combo, the settings choice is more about competitive advantage versus visual preference rather than achieving playable frame rates, since even the most demanding configuration easily exceeds 500 FPS on average. Given that Valorant is a tactical shooter where frame rate consistency matters, the Medium preset's combination of high averages and known minimums (454 FPS at 4K, 489 FPS at 1440p, 515 FPS at 1080p) makes it a strong recommendation for players who want both quality and performance without risking the unknown frame dips that could occur at Ultra settings.

CPU Role

The AMD Ryzen 9 9900X is a 12-core, 24-thread processor based on the Zen 5 architecture, built on a 4 nm process at TSMC. It has a base clock of 4.40 GHz and a boost clock of 5.60 GHz, with a 64 MB L3 cache. Its single-thread performance is exceptional — the 3dmark_single_thread score is 1286, and the Cinebench R23 single-core score is 2253. These numbers indicate that the CPU has immense per-core capability, which is crucial for Valorant's game engine that typically relies heavily on a few primary threads.

The benchmark results for the 9900X reveal a processor that scales well with thread count. The 3dmark scores progress from 2519 at 2 threads to 4886 at 4 threads, 9114 at 8 threads, and 13929 at max threads. This scaling pattern suggests that the CPU can efficiently distribute workload across its cores, though the diminishing returns beyond 8 threads (only 53% increase from 8 to 24 threads) indicate that the architecture is optimized for both low-thread and high-thread scenarios.

In Valorant, the CPU's role is likely dominant at lower resolutions. The data shows that at 1080p, the FPS difference between Low (711) and Ultra (586) is 125 FPS, which is the largest absolute gap across all resolutions. This suggests that at 1080p, the CPU is doing more of the heavy lifting, and turning down settings reduces the CPU-side workload (such as draw call processing and physics calculations), resulting in higher frame rates. At 4K, the same settings change only yields a 110 FPS difference, indicating that the GPU is becoming more of a bottleneck, reducing the CPU's influence on the final frame rate.

The CPU's percentile ranking of 92 against all CPUs, combined with its average benchmark score of 57498, places it in the upper echelon of processors. Its nearest rival, the Intel Core i9-14900, has an average score of 58115, which is 1.1% higher. This close margin in synthetic benchmarks suggests that real-world gaming performance would be nearly identical between these two processors, though the 9900X's 12-core/24-thread configuration and Zen 5 architecture may offer different characteristics in specific workloads.

The 9900X's boost clock of 5.60 GHz is notably high for a 12-core part, and this is likely a key factor in Valorant's performance. The game's engine benefits from high single-thread throughput, and the 3dmark_single_thread score of 1286 confirms that this CPU delivers top-tier performance in lightly-threaded scenarios. The fact that the combo achieves 711 FPS at 1080p Low — a scenario that is typically CPU-bound in most games — underscores the processor's capability to feed the GPU with enough frames to reach such extreme frame rates.

How This Combo Ranks

This specific combination of the AMD Ryzen 9 9900X and NVIDIA GeForce RTX 5090 D is ranked 48th out of 2883 tested combos in Valorant. This places it in the top 1.7% of all systems tested for this game. The rank is a composite measure that accounts for the combined performance of both components in the specific context of Valorant's engine and rendering requirements.

The combo's high ranking is particularly interesting given that neither the CPU nor the GPU individually holds the top percentile ranking. The CPU has a percentileVsAllCpus of 92, and the GPU has a percentileVsAllGpus of 92. Yet, when paired together, they achieve a rank that places them in the 98.3rd percentile of all combos. This suggests that the pairing is synergistic — the CPU's strengths (high single-thread performance) complement the GPU's strengths (massive rasterization throughput) in a way that is particularly effective for Valorant's specific workload.

The fact that there are 47 combos ranked higher indicates that there are some configurations that outperform this one in Valorant. Given that the CPU's closest rivals (Intel Core i9-14900, AMD EPYC 7313, etc.) have similar average benchmark scores, and the GPU's closest rivals (AMD Radeon RX 6850M XT, NVIDIA Tesla P100, etc.) have similar scores, it's likely that the combos ranked above this one feature slightly different component pairings that edge out this configuration in specific game scenarios. However, the difference is likely marginal, given the extremely high frame rates involved.

The combo's rank of 48 out of 2883 also implies that there are 2835 combos ranked below it. This puts the system in an elite tier, where the performance difference between rank 48 and rank 1 is likely imperceptible in actual gameplay, as the frame rates are all well above the refresh rates of any consumer monitor. The data suggests that for Valorant, once a system reaches this level of performance, the ranking becomes more of a theoretical exercise than a practical indicator of playability.

GPU Role

The NVIDIA GeForce RTX 5090 D is a monster of a graphics card. It features 32 GB of GDDR7 memory on a 512-bit bus, delivering 1.79 TB/s of memory bandwidth. The GPU has 21760 shading units, 680 TMUs, and 176 ROPs, with a boost clock of 2407 MHz. Its FP32 performance is rated at 104.8 TFLOPS, and it has a pixel rate of 423.6 GPixel/s. These specifications are among the highest in the database, and the GPU's percentileVsAllGpus of 92 confirms its elite status.

In Valorant, the GPU's role becomes more pronounced at higher resolutions. The data shows that at 4K, the average FPS across all settings is 550.25 (average of 525, 626, 534, and 516). At 1440p, the average is 592.75, and at 1080p, the average is 624.75. The progression from 1080p to 4K results in an 11.9% drop in average FPS, which is relatively small compared to what would be expected in more graphically demanding titles. This indicates that Valorant is not particularly GPU-intensive, and even at 4K, the RTX 5090 D is not being pushed to its limits.

The GPU's memory bandwidth of 1.79 TB/s is likely a significant factor in maintaining high frame rates at 4K. The 32 GB of GDDR7 memory ensures that there is no VRAM capacity limitation, even at Ultra settings with high-resolution textures. The 512-bit memory bus provides the bandwidth needed to feed the 21760 shading units, preventing any memory-bound bottlenecks that could occur in other cards.

Interestingly, the GPU's nearest rivals in the benchmark database are not other high-end desktop cards but rather mobile and data center parts: the AMD Radeon RX 6650M XT (1.1% higher score) and the AMD Radeon RX 6850M XT (1.6% lower score). This is unusual and suggests that the RTX 5090 D's benchmark score of 77712 may not fully capture its gaming performance, as the 3dmark_steel_nomad_dx12 score of 14326 and the geekbench_vulkan score of 376915 provide a more nuanced picture of its capabilities.

The GPU's boost clock of 2407 MHz is relatively modest compared to some competitor cards, but the sheer number of shading units (21760) compensates for this. The combination of high core count and high memory bandwidth allows the RTX 5090 D to maintain over 500 FPS even at 4K Ultra, which confirms its raw compute power. In Valorant, where the engine is not particularly demanding on the GPU, this card provides an immense amount of headroom, making it unlikely that the GPU will ever be the limiting factor in frame rate, even at the highest settings.

Measured FPS Breakdown

The measured FPS data provides a comprehensive picture of how this combo performs across different resolutions and settings in Valorant.

At 1920x1080, the system delivers its highest frame rates. The Low preset averages 711 FPS, which is the peak performance recorded in the entire dataset. The Medium preset averages 606 FPS, with a minimum of 515 FPS and a maximum of 697 FPS. The High preset averages 596 FPS, and the Ultra preset averages 586 FPS. The spread from Low to Ultra is 125 FPS, indicating that settings have a moderate impact on performance at this resolution.

At 2560x1440, the frame rates remain exceptionally high. The Low preset averages 674 FPS, which is 37 FPS lower than the 1080p Low result. The Medium preset averages 575 FPS, with a minimum of 489 FPS and a maximum of 661 FPS. The High preset averages 566 FPS, and the Ultra preset averages 556 FPS. The spread from Low to Ultra is 118 FPS, slightly narrower than at 1080p.

At 3840x2160, the system still maintains over 500 FPS in all presets. The Low preset averages 626 FPS, which is the second-highest average in the dataset after the 1080p Low result. The Medium preset averages 534 FPS, with a minimum of 454 FPS and a maximum of 614 FPS. The High preset averages 525 FPS, and the Ultra preset averages 516 FPS. The spread from Low to Ultra is 110 FPS, the narrowest of all resolutions.

The Medium preset is the only one that provides complete min/max data across all resolutions. At 1080p, the minimum is 515 FPS and the maximum is 697 FPS, giving a range of 182 FPS. At 1440p, the range is 172 FPS (489 to 661). At 4K, the range is 160 FPS (454 to 614). This narrowing range at higher resolutions suggests that frame time consistency improves as the GPU becomes more of a limiting factor, likely because the GPU's workload is more predictable than the CPU's bursty frame generation.

The data shows that the Low preset at 1080p is the peak performance configuration, delivering 711 FPS. However, the difference between Low and Ultra at 4K is only 110 FPS, which is 15 FPS less than the difference at 1080p. This suggests that at 4K, the GPU is more consistently the bottleneck, and the CPU's influence on frame rate is reduced.

Resolution Scaling

The resolution scaling behavior of this combo reveals important insights into which component is the primary limiter at different settings. From 1080p to 4K, the average FPS across all settings drops from 624.75 to 550.25, a decrease of 74.5 FPS or 11.9%. This is a relatively modest drop for a 4x increase in pixel count, indicating that the system is heavily CPU-bound at lower resolutions.

Looking at specific settings, the Low preset shows the largest absolute drop from 1080p to 4K: from 711 FPS to 626 FPS, a decrease of 85 FPS or 12.0%. The Ultra preset shows a smaller drop: from 586 FPS to 516 FPS, a decrease of 70 FPS or 11.9%. The similarity in percentage drops across settings suggests that the resolution scaling is consistent regardless of quality preset, which points to a fundamental system-level limiter rather than a settings-specific bottleneck.

The fact that the FPS drops by only 12% when going from 1080p to 4K is very telling. In a purely GPU-bound scenario, one would expect a much larger drop, closer to 50-60% or more, given the 4x increase in pixel count. The modest 12% drop indicates that the CPU is the primary bottleneck at 1080p, and even at 4K, the GPU is still not fully saturated. The RTX 5090 D's immense compute power (104.8 TFLOPS FP32) and memory bandwidth (1.79 TB/s) allow it to handle the 4K workload with relative ease, while the Ryzen 9 9900X's single-thread performance is the limiting factor in frame generation.

This interpretation is further supported by the settings scaling data. At 1080p, the difference between Low and Ultra is 125 FPS, which is 17.6% of the Low preset's frame rate. At 4K, the same settings difference is 110 FPS, which is 17.6% of the 4K Low frame rate. The consistent percentage impact of settings changes across resolutions suggests that the settings are primarily affecting the CPU-side workload (such as draw calls and shadow calculations) rather than the GPU-side workload, as the relative impact doesn't change with resolution.

The data implies that for this combo, the optimal configuration for maximum frame rate is 1080p Low, which achieves 711 FPS. However, the frame rate at 4K Ultra (516 FPS) is still more than sufficient for any display refresh rate currently available. The limiting component at 1080p is clearly the CPU, as the GPU has massive headroom that is only partially utilized at lower resolutions. At 4K, the GPU begins to take on a more significant role, but it still has enough headroom to maintain over 500 FPS, suggesting that even at 4K, the system is not fully GPU-bound. This is a remarkable result that highlights the extraordinary performance of both components in this pairing.