Doom

# Doom with AMD Ryzen 9 5900X + NVIDIA GeForce RTX 4090

This combination of the AMD Ryzen 9 5900X and NVIDIA GeForce RTX 4090 places at rank 156 out of 3712 tested combos in Doom, putting it in the top 4.2% of all systems benchmarked for this title. The pairing of a 12-core Zen 3 processor with a 24 GB Ada Lovelace flagship produces frame rates that are exceptionally high across every resolution tested, though the data reveals interesting scaling behavior that suggests where the bottleneck shifts as settings and resolution change.

How This Combo Ranks

The combo’s rank of 156 out of 3712 tested combinations indicates that only about 4.2% of systems outperform it in Doom, a game from 2016 that was designed before either of these components existed. The CPU itself sits at the 87th percentile among all processors, with an average benchmark score of 41376, while the GPU holds the 88th percentile among all graphics cards with an average score of 60347. These percentile positions are remarkably close, suggesting a balanced pairing where neither component dramatically outclasses the other in general compute terms.

Looking at the CPU’s nearest rivals, the Ryzen 9 5900X is essentially tied with three Intel and AMD mobile processors: the Intel Core Ultra 7 366H (0.3% higher score), Intel Core Ultra 7 356H (0.4% higher), and AMD Ryzen AI 5 PRO 440 (0.4% higher). The Intel Core i7-14650HX trails by 0.5%. This clustering within a 0.9% band demonstrates that the 5900X remains competitive in synthetic multi-threaded workloads despite being a desktop part from late 2020.

For the GPU, the RTX 4090’s nearest rivals show a peculiar pattern. The Intel Arc Pro A60 matches it exactly at 0% delta, while the AMD Radeon Pro Vega 48 is 0.3% behind and the AMD Radeon PRO V710 is 2.9% behind. The AMD Radeon Pro W6600M actually scores 2.5% higher. These workstation-oriented comparisons highlight that the RTX 4090’s massive 82.58 TFLOPS FP32 throughput doesn’t always translate to dominance in every synthetic benchmark category.

In Doom specifically, the measured frame rates far exceed what the component percentiles alone would suggest. The game’s id Tech 6 engine clearly responds well to both the CPU’s high single-thread performance (3dmark single-thread score of 941, Cinebench R23 single-core of 1527) and the GPU’s raw rasterization power. This is not a case where a high rank comes from one component carrying the other — both contribute meaningfully across the tested configurations.

FAQ

Q: Can this system maintain 100+ FPS at 4K Ultra settings in Doom?

A: Yes, but barely. The measured average at 3840x2160 Ultra is 94 FPS, which is just under the 100 FPS threshold. Dropping to High settings at the same resolution yields 115 FPS average, providing a comfortable margin above 100 FPS.

Q: How does the frame rate change when going from 1080p to 1440p to 4K?

A: At High settings, the average drops from 326 FPS at 1920x1080 to 211 FPS at 2560x1440, then to 115 FPS at 3840x2160. This represents a 35% reduction from 1080p to 1440p, and a further 45% reduction from 1440p to 4K.

Q: What is the best preset for competitive play at 1440p?

A: The Low preset at 2560x1440 delivers 287 FPS average, while Medium achieves 250 FPS and High reaches 211 FPS. For maximum responsiveness on high-refresh monitors, Low at 1440p is the strongest option, though Medium still provides over 250 FPS with better visual fidelity.

Q: Is the CPU or GPU more likely to be the limiting factor at 1080p?

A: At 1920x1080, the frame rates are extraordinarily high (363 FPS at Low, 326 FPS at High), suggesting the CPU’s single-thread performance is becoming relevant. The Ryzen 9 5900X’s boost clock of 4.80 GHz and 3dmark single-thread score of 941 indicate strong per-core capability, but the RTX 4090’s pixel rate of 443.5 GPixel/s is still more than adequate for this resolution.

Q: Does the 24 GB VRAM provide any advantage in this game?

A: No. Doom from 2016 does not require anywhere near 24 GB of memory, and even the Ultra preset at 4K runs within the GPU’s available VRAM without issue. The memory bandwidth of 1.01 TB/s is more relevant for frame pacing at high resolutions.

Q: How does the minimum FPS compare to the average at Medium settings?

A: At Medium settings, the minimum FPS is 115 at 4K, 213 at 1440p, and 294 at 1080p. The gap between minimum and average grows at higher resolutions: 21 FPS at 4K, 37 FPS at 1440p, and 52 FPS at 1080p, indicating more consistent frame delivery at lower resolutions.

Settings Recommendations

For 4K Ultra, the 94 FPS average is technically playable but leaves little headroom for demanding scenes. The High preset at 4K provides 115 FPS average, which is a 22% improvement over Ultra while maintaining high visual quality. This is the recommended balance for 4K displays.

At 1440p, the Medium preset offers 250 FPS average with a minimum of 213 FPS, which exceeds the refresh rate of most high-end monitors. The High preset at 211 FPS is also excellent. For those with 240 Hz or higher refresh displays, the Low preset’s 287 FPS average is the only option that fully saturates such panels.

At 1080p, every preset exceeds 260 FPS, with Low reaching 363 FPS and Medium at 346 FPS. The Ultra preset’s 267 FPS is still well above any mainstream monitor’s refresh rate. Given the CPU’s 24 threads and the GPU’s capabilities, the Medium preset at 1080p appears to be the sweet spot — it delivers 346 FPS average with a 294 FPS minimum, offering improved visual quality over Low without sacrificing meaningful performance.

The measured data shows diminishing returns from High to Ultra at every resolution: at 4K the drop is 21 FPS (115 to 94), at 1440p it is 38 FPS (211 to 173), and at 1080p it is 59 FPS (326 to 267). This consistent pattern suggests Ultra settings impose a similar percentage penalty regardless of resolution.

Measured FPS Breakdown

At 3840x2160, the four settings produce a clear hierarchy: Low averages 156 FPS, Medium 136 FPS (minimum 115, maximum 156), High 115 FPS, and Ultra 94 FPS. The spread from Low to Ultra is 62 FPS, representing a 40% performance cost for the highest visual quality. The Medium preset’s maximum of 156 FPS matches the Low average exactly, indicating that even at 4K, the engine has substantial headroom on the lowest settings.

The 2560x1440 results show a similar pattern with higher absolute values: Low averages 287 FPS, Medium 250 FPS (minimum 213, maximum 288), High 211 FPS, and Ultra 173 FPS. The gap between Low and Ultra is 114 FPS, a 40% reduction that mirrors the 4K scaling. The Medium preset’s maximum of 288 FPS is one frame higher than the Low average, suggesting the Medium settings occasionally allow higher peak performance than sustained Low settings.

At 1920x1080, the frame rates become exceptionally high: Low averages 363 FPS, Medium 346 FPS (minimum 294, maximum 398), High 326 FPS, and Ultra 267 FPS. The difference between Low and Ultra is 96 FPS, which is proportionally smaller (26%) than at higher resolutions. The Medium preset’s maximum of 398 FPS is the highest single measurement in the entire dataset, exceeding the Low average by 35 FPS.

Across all resolutions, the Medium preset consistently shows both minimum and maximum values, while other presets only report averages. This suggests the Medium tier was tested more thoroughly or exhibits more variance in frame delivery.

Resolution Scaling

The transition from 1920x1080 to 2560x1440 at High settings reduces the average FPS from 326 to 211, a 35% decrease. Moving from 1440p to 3840x2160 at High drops the average from 211 to 115, a 45% decrease. The cumulative drop from 1080p to 4K is 65%, meaning 4K delivers only about one-third of the 1080p frame rate.

At Low settings, the scaling is slightly different: 363 FPS at 1080p drops to 287 FPS at 1440p (21% decrease), then to 156 FPS at 4K (46% decrease). The total drop from 1080p to 4K at Low is 57%. At Medium settings, the progression is 346 to 250 to 136 FPS, representing 28% and 46% decreases respectively, for a total of 61%.

The Ultra preset shows the steepest scaling: 267 FPS at 1080p, 173 FPS at 1440p (35% decrease), and 94 FPS at 4K (46% decrease), totaling 65% reduction. This consistent pattern across settings — where the 1440p to 4K transition always costs more than the 1080p to 1440p transition — indicates the GPU’s fill rate and memory bandwidth become increasingly strained at higher pixel counts.

The RTX 4090’s 1.01 TB/s memory bandwidth and 443.5 GPixel/s pixel rate are substantial, but the 4K pixel count (8.3 million pixels) is 2.25 times higher than 1440p (3.7 million) and 4 times higher than 1080p (2.1 million). The frame rate drops align more closely with pixel count increases than with linear scaling, confirming that the GPU is the primary limiter at higher resolutions.

CPU Role

The AMD Ryzen 9 5900X brings 12 cores and 24 threads based on the Zen 3 architecture, codenamed Vermeer. Its base clock of 3.70 GHz boosts to 4.80 GHz, with 64 MB of L3 cache spread across the chip. The Cinebench R23 multi-core score of 16262 and single-core score of 1527 demonstrate strong performance in both threaded and single-threaded workloads.

In Doom, which was released in 2016 and is not known for heavy multi-threading, the CPU’s single-thread performance is more relevant than its core count. The 3dmark single-thread score of 941 and 2-thread score of 1852 indicate per-core efficiency that allows the game engine to run without CPU-induced frame pacing issues. The PassMark single-thread score of 3469 further confirms this capability.

The CPU’s memory bandwidth of 51.2 GB/s over dual-channel DDR4 is modest by modern standards, but Doom’s memory access patterns do not appear to stress this limitation. At 1080p where frame rates exceed 360 FPS, the CPU’s ability to feed the GPU becomes more critical. The 24 threads provide ample headroom for background tasks and game engine threads, though the game likely uses far fewer cores.

The nearest rival comparisons show the 5900X trading blows with newer mobile processors within a 0.5% band, which is remarkable for a 2020 desktop chip. The 87th percentile ranking among all CPUs means it outperforms the vast majority of processors in general benchmarks, and in Doom specifically, the combination’s rank of 156 out of 3712 suggests the CPU is not holding back the RTX 4090 at any tested setting.

GPU Role

The NVIDIA GeForce RTX 4090 is built on the Ada Lovelace architecture with the AD102 chip, manufactured on a 5 nm process at TSMC with 76,300 million transistors. Its 24 GB of GDDR6X memory on a 384-bit bus delivers 1.01 TB/s of bandwidth, while the boost clock of 2520 MHz drives 82.58 TFLOPS of FP32 compute through 16384 shading units.

In Doom, the GPU’s raw rasterization power is the dominant factor at higher resolutions. The 4K Ultra average of 94 FPS represents the most demanding configuration tested, and the GPU handles it with a comfortable margin above playable frame rates. The pixel rate of 443.5 GPixel/s and texture rate of 1,290.2 GTexel/s are more than sufficient for this 2016 title, even at maximum settings.

The GPU’s PassMark G3D score of 38194 and 3DMark Steel Nomad DX12 score of 9223 place it at the 88th percentile among all GPUs. Its nearest rivals include the Intel Arc Pro A60 with a 0% delta, meaning the synthetic benchmark scores are effectively identical. This is surprising given the RTX 4090’s specifications, but the comparison reflects average scores across multiple tests rather than gaming-specific performance.

The 24 GB VRAM is far beyond what Doom requires, and the memory bandwidth of 1.01 TB/s ensures that texture streaming and frame buffer operations never become bottlenecks. The GPU’s boost clock of 2520 MHz provides consistent performance across scenes, as evidenced by the measured FPS data showing no dramatic minimum frame dips at any configuration. The 450 W TDP and triple-slot cooler design indicate this is a high-power component, but the game’s performance justifies the power draw in terms of absolute frame rates.