Doom

The AMD Ryzen 7 5800XT paired with the NVIDIA GeForce RTX 3070 Ti delivers a robust performance profile in Doom, demonstrating a clear shift in bottleneck behavior as resolution increases. The measured frame rates across three resolutions and four quality presets show a system that is heavily CPU-bound at 1080p, transitioning to a GPU-bound state at 4K, with the 1440p tier representing the optimal balance for this hardware combination.

Resolution Scaling

The frame rate deltas across resolutions reveal a classic scaling curve. At 1080p with Ultra settings, the combo produces 163.9 FPS. Moving to 1440p Ultra, the average drops to 103.8 FPS, a reduction of 60.1 FPS. The jump to 4K Ultra further compresses performance to 57.1 FPS, a 46.7 FPS decline from the 1440p figure. This pattern indicates that at 1080p, the CPU’s 16 threads are capable of feeding the GPU well beyond the display’s refresh rate, but the RTX 3070 Ti’s rendering load increases substantially with pixel count.

The scaling from Low to Ultra settings at each resolution reinforces this. At 1080p, the difference between Low (270.1 FPS) and Ultra (163.9 FPS) is a substantial 106.2 FPS, suggesting that even at the lowest settings, the CPU is not the limiting factor—the GPU is still processing a massive number of frames. However, at 4K, the spread narrows considerably: Low yields 97.4 FPS, while Ultra drops to 57.1 FPS. The 40.3 FPS gap at 4K is proportionally smaller than at 1080p, confirming that the GPU’s fill rate and shading capabilities become the primary constraint at higher resolutions.

Data indicates the transition point is between 1440p and 4K. At 1440p High, the system achieves 131.6 FPS, which is still high enough to suggest the CPU is doing significant work. At 4K High, the 68.4 FPS result shows the GPU is now the dominant bottleneck, as the resolution increase of roughly 2.25 times the pixel count (from 1440p to 4K) yields a frame rate that is nearly halved. This is a textbook sign of a GPU that is saturated by pixel throughput, rather than a CPU that is starved for instructions.

How This Combo Ranks

Within the benchmark database of 1,681 tested combos for Doom, this specific pairing ranks at position 665. This places it in the upper-middle tier of all tested configurations. The percentile data for the individual components provides context: the Ryzen 7 5800XT sits at the 86th percentile among all CPUs, while the RTX 3070 Ti ranks at the 74th percentile among all GPUs. The composite rank of 665 out of 1,681 suggests that the pairing is well-matched, though the GPU’s lower percentile relative to the CPU means the graphics card is likely the constraining factor in more demanding scenarios.

The CPU’s nearest rivals in the database include the AMD Ryzen 7 PRO 5755GE, which scores within 0.1% of the 5800XT’s average benchmark score. The AMD Ryzen 7 6800HS and AMD Ryzen 7 7840H are also close, with the 5800XT trailing the 7840H by 0.3%. These small deltas indicate that the 5800XT’s Zen 3 architecture provides competitive multi-threaded performance for its segment. The GPU’s closest rivals are the AMD Radeon RX 590 GME (0.3% ahead) and the AMD Radeon RX 6700 (0.8% behind), showing that the RTX 3070 Ti’s average benchmark score is tightly clustered with these cards, despite the architectural differences.

The 4K Ultra result of 57.1 FPS is the only measured configuration that falls below the 60 FPS threshold commonly associated with smooth gameplay. This suggests that for 4K displays, the combo is marginally insufficient for maximum quality settings, but the 4K High setting at 68.4 FPS provides a viable alternative. The rank of 665 is heavily influenced by the 4K results, as higher-resolution performance carries more weight in the overall ranking due to the greater number of combos that can achieve high frame rates at 1080p.

Settings Recommendations

The measured data supports a clear hierarchy of settings based on resolution targets. For 1080p displays, the Ultra preset at 163.9 FPS is the recommended choice, as the frame rate remains far above the refresh rate of most monitors, and the visual fidelity gains are substantial. Dropping to Medium at 236.2 FPS or Low at 270.1 FPS offers diminishing returns, as the extra frames are unlikely to be perceptible on standard 144Hz or 165Hz panels.

At 1440p, the High preset at 131.6 FPS is the sweet spot. It provides a significant visual upgrade over Medium (153 FPS) while maintaining a frame rate that is well-suited for high-refresh-rate gaming. The Ultra preset at 103.8 FPS is still playable, but the 27.8 FPS cost over High may not justify the marginal quality increase in shadow and texture filtering. Low at 178.5 FPS is unnecessary for most users, as the CPU and GPU are both capable of high performance without sacrificing quality.

For 4K, the recommendations shift to prioritize framerate stability. The High preset at 68.4 FPS is the best experience, as it stays above the 60 FPS baseline while offering a substantial quality improvement over Medium (83.3 FPS). The Ultra preset at 57.1 FPS is borderline, and the data suggests that the 11.3 FPS drop from High to Ultra is too steep for a 4K panel. Low at 97.4 FPS is viable for competitive play, but the visual compromise is severe for a single-player title like Doom.

Measured FPS Breakdown

At 1920x1080, the measured averages are: Low 270.1 FPS, Medium 236.2 FPS, High 201.2 FPS, and Ultra 163.9 FPS. The scaling from Low to Ultra shows a consistent 30-40 FPS drop per tier, indicating that the GPU is the primary driver of performance even at this resolution, as the CPU can easily sustain these frame rates.

At 2560x1440, the results are: Low 178.5 FPS, Medium 153 FPS, High 131.6 FPS, and Ultra 103.8 FPS. The absolute deltas between settings are smaller than at 1080p, with the largest gap being 74.7 FPS between Low and Ultra. This suggests that the GPU is starting to feel the load, but the system remains fluid across all presets.

At 3840x2160, the averages are: Low 97.4 FPS, Medium 83.3 FPS, High 68.4 FPS, and Ultra 57.1 FPS. The scaling here is more linear, with each preset tier costing roughly 10-15 FPS. The 4K Ultra result is the only sub-60 FPS measurement in the entire dataset, which underscores the GPU’s limitations at this extreme resolution. The 4K Low result of 97.4 FPS is still respectable, but it represents a 172.7 FPS drop from the 1080p Low figure, highlighting the massive impact of pixel count on the RTX 3070 Ti’s rendering pipeline.

GPU Role

The RTX 3070 Ti’s specifications directly influence these results. The 8 GB of GDDR6X memory on a 256-bit bus provides 608.3 GB/s of bandwidth, which is sufficient for Doom’s textures at 4K, but the card’s 6144 shading units and 96 ROPs are more critical for the observed frame rates. The boost clock of 1770 MHz, combined with a base clock of 1575 MHz, allows the GPU to maintain high throughput in this title, which is known for its efficient id Tech engine.

The GPU’s 21.75 TFLOPS of FP32 performance is the theoretical ceiling, and the measured data shows that the card approaches this limit at 1080p but falls short at 4K due to memory bandwidth and ROP throughput constraints. The 48 RT cores and 192 tensor cores are present in the hardware, but Doom (2016) does not utilize ray tracing or DLSS, meaning the raw rasterization performance is the primary factor. The 169.9 GPixel/s pixel rate and 339.8 GTexel/s texture rate are the key metrics for this game’s rendering workload.

The GPU’s percentile rank of 74 among all GPUs is lower than the CPU’s 86th percentile, which explains why the combo’s overall rank of 665 out of 1681 is not higher. The 290 W TDP and 600 W suggested PSU indicate that the card has ample power headroom, but in this specific title, the bottleneck is clearly the GPU’s shading and pixel processing capabilities at higher resolutions. The 4K Ultra result of 57.1 FPS is a direct consequence of the 96 ROPs being tasked with filling a 3840x2160 buffer, while the 1080p results show the card can easily exceed 200 FPS when the pixel count is reduced by a factor of four.