Doom

This CPU and GPU pairing delivers a surprisingly consistent 1080p experience, but the data shows exactly where that consistency ends. The measured results for Doom indicate that the RTX 3070 Ti is the primary driver of frame rates at higher resolutions, while the Ryzen 5 5500 holds its own at lower settings.

Resolution Scaling

The frame rate drop from 1080p to 4K is steep and clearly points to the GPU as the limiting factor at higher resolutions. At High settings, the average FPS falls from 200 at 1920x1080 to 129 at 2560x1440, a 35.5% reduction. Moving from 1440p to 4K brings another drop to 70 FPS, which is another 45.7% decrease. The pattern is consistent across all settings presets, with the gap between 1080p and 4K being roughly 65-70% of the 1080p value.

The scaling behavior changes depending on the preset. At Low settings, the drop from 1080p to 1440p is 95 FPS (from 271 to 176), but the drop from 1440p to 4K is 81 FPS (from 176 to 95). At Ultra settings, the corresponding drops are 57 FPS (163 to 106) and 49 FPS (106 to 57). The absolute reduction in FPS is smaller at Ultra, but the percentage loss is larger because the starting frame rate is already lower.

The 4K Medium result shows a minimum of 70 FPS and a maximum of 95 FPS, which brackets the average of 83 FPS. This range is wider than the 1440p Medium range (130-176 FPS) and the 1080p Medium range (201-272 FPS), suggesting that at 4K, the GPU is working harder and frame times are less consistent. The fact that the 4K High average of 70 FPS matches the 4K Medium minimum exactly indicates that the GPU is nearing its practical limit at this resolution, regardless of the small step between Medium and High.

The data indicates that 1440p is the sweet spot for this pairing. The 1440p High average of 129 FPS is well above the 60 FPS threshold for smooth gameplay, and even the 1440p Ultra result of 106 FPS leaves substantial headroom. At 4K, the GPU becomes the bottleneck, with even Low settings only reaching 95 FPS, which is below the 1080p Medium result of 236 FPS. The 1080p results, on the other hand, are so high that the CPU likely becomes the limiting factor at Low settings, where the 271 FPS average is likely constrained by the processor's single-thread performance rather than the GPU.

Settings Recommendations

The measured rows show that Medium settings offer the best balance between visual quality and frame rate. At 1080p, Medium delivers 236 FPS average, which is 36 FPS higher than High and 73 FPS higher than Ultra. The jump from Medium to High costs 36 FPS, while the jump from High to Ultra costs another 37 FPS. The difference between Medium and High is nearly identical to the difference between High and Ultra, so there is no particular advantage to choosing High over Medium beyond the visual fidelity.

At 1440p, Medium provides 153 FPS average, which is 24 FPS more than High and 47 FPS more than Ultra. The relative penalty for Ultra is larger at 1440p than at 1080p, with the 106 FPS Ultra result being 30.7% below the Medium result. At 4K, Medium delivers 83 FPS, which is 13 FPS more than High and 26 FPS more than Ultra. The 4K Ultra result of 57 FPS is below the 60 FPS threshold, making Ultra at 4K a poor choice for smooth gameplay.

For competitive play, Low settings at 1080p provide the highest frame rate at 271 FPS, which is 35 FPS higher than Medium and 71 FPS higher than High. The gap between Low and Medium is larger than the gap between Medium and High, suggesting diminishing returns from Low to Medium. For single-player experiences, Medium at 1440p offers the best compromise, with 153 FPS being more than double the 4K High result and still well above refresh rates of common monitors.

The data suggests avoiding Ultra at 4K entirely, as the 57 FPS average is below the 60 FPS baseline. At 1440p, Ultra is viable at 106 FPS, but the jump from High to Ultra costs 23 FPS, which is a 17.8% reduction. At 1080p, all settings are playable, but the 163 FPS Ultra result is 37 FPS below High, making High a better choice if the visual difference is not critical.

FAQ

Q: What is the best resolution for this pairing?

A: The data shows 2560x1440 as the sweet spot. At High settings, the average FPS is 129, which is 84.3% higher than the 4K High result of 70 FPS. Even Ultra at 1440p (106 FPS) beats 4K Medium (83 FPS).

Q: How much FPS does Ultra cost compared to Medium?

A: At 1080p, Ultra is 73 FPS lower than Medium (163 vs 236). At 1440p, the gap is 47 FPS (106 vs 153). At 4K, the gap is 26 FPS (57 vs 83). The percentage penalty grows with resolution.

Q: Is 4K playable with this hardware?

A: At 4K, the best result is 95 FPS at Low settings. Medium delivers 83 FPS, High drops to 70 FPS, and Ultra falls to 57 FPS. For 60 FPS gameplay, Medium is the highest preset that stays above the threshold.

Q: Does the CPU limit performance at any resolution?

A: The 1080p Low result of 271 FPS is the highest measured value, and the gap between 1080p Low and 1440p Low is 95 FPS, while the gap between 1440p Low and 4K Low is 81 FPS. The proportional drop from 1080p to 1440p is 35.1%, while the drop from 1440p to 4K is 46.0%, indicating that the GPU is the limiting factor at all resolutions.

Q: What settings should I use for a 144Hz monitor?

A: At 1440p, High settings deliver 129 FPS, which is below 144 FPS. Medium delivers 153 FPS, which exceeds it. At 1080p, Low (271 FPS), Medium (236 FPS), and High (200 FPS) all exceed 144 FPS, but Ultra falls short at 163 FPS.

Q: How does the 3070 Ti compare to its nearest rivals in this context?

A: The RTX 3070 Ti has an average benchmark score of 29945, which places it 0.1% ahead of the RTX 5070 Mobile, 0.5% ahead of the RTX 2080 Ti, 0.5% behind the AMD Radeon RX 6800, and 1.6% behind the AMD Radeon RX 6700.

Measured FPS Breakdown

At 1920x1080, the results are consistently high across all presets. Low settings produce an average of 271 FPS, Medium delivers 236 FPS with a minimum of 201 and maximum of 272, High yields 200 FPS, and Ultra drops to 163 FPS. The spread from Low to Ultra is 108 FPS, with each step up in settings costing roughly 35-37 FPS. The Medium range of 201-272 FPS shows that the minimum frame rate is still well above the maximum of most standard monitors.

At 2560x1440, the averages are 176 FPS at Low, 153 FPS at Medium (minimum 130, maximum 176), 129 FPS at High, and 106 FPS at Ultra. The drop from 1080p to 1440p is 95 FPS at Low, 83 FPS at Medium, 71 FPS at High, and 57 FPS at Ultra. The absolute FPS cost of moving to 1440p decreases as settings increase, but the percentage cost increases. At Low, the 1440p result is 64.9% of the 1080p result, while at Ultra, it is 65.0% of the 1080p result.

At 3840x2160, the results are 95 FPS at Low, 83 FPS at Medium (minimum 70, maximum 95), 70 FPS at High, and 57 FPS at Ultra. The 4K Medium minimum of 70 FPS is equal to the 4K High average, indicating frame time inconsistency at this resolution. The 4K Ultra result of 57 FPS is the only measured value below 60 FPS, making it the only preset and resolution combination that fails to maintain smooth gameplay. The 4K Medium maximum of 95 FPS is also the same as the 4K Low average, suggesting that Medium at 4K can occasionally match Low performance but with better visuals.

The progression from 1080p to 1440p to 4K shows a consistent pattern of diminishing returns. The 1080p Medium result of 236 FPS is 54.2% higher than the 1440p Medium result of 153 FPS, which is 84.3% higher than the 4K Medium result of 83 FPS. The gap between 1080p and 1440p is larger than the gap between 1440p and 4K in absolute terms, but the relative performance loss is similar across presets.

GPU Role

The NVIDIA GeForce RTX 3070 Ti features 8 GB of GDDR6X memory on a 256-bit bus, delivering 608.3 GB/s of bandwidth. The memory clock runs at 1188 MHz with 19 Gbps effective speed. The GPU has 6144 shading units, 192 texture mapping units, and 96 raster output units, with 48 RT cores and 192 tensor cores. The base clock is 1575 MHz and the boost clock is 1770 MHz, producing 21.75 TFLOPS of FP32 performance.

The benchmark results place this GPU in the 75th percentile of all GPUs, with an average score of 29945. Its nearest rivals include the RTX 5070 Mobile (29928, 0.1% ahead), the RTX 2080 Ti (29783, 0.5% ahead), the RX 6800 (30095, 0.5% behind), and the RX 6700 (30433, 1.6% behind). These close margins indicate that the 3070 Ti sits in a competitive performance band, and the measured FPS results in Doom reflect this positioning.

The 8 GB VRAM capacity is sufficient for this game at the measured settings. The 4K Ultra result of 57 FPS is more likely limited by raw compute performance than memory capacity, as the 4K Medium result of 83 FPS shows the GPU can handle higher resolutions when the settings are reduced. The memory bandwidth of 608.3 GB/s appears adequate for the texture and geometry loads in Doom, as the FPS scaling from 1080p to 4K follows a predictable pattern based on pixel count rather than hitting an unexpected memory bottleneck.

The GPU's 169.9 GPixel/s pixel rate and 339.8 GTexel/s texture rate provide context for the resolution scaling. The drop from 1080p to 1440p increases the pixel count by 78%, and the measured FPS drop at Medium settings is 83 FPS (35.2%). The drop from 1440p to 4K increases the pixel count by 125%, and the FPS drop is 70 FPS (45.8%). The larger percentage drop at 4K indicates the GPU is approaching its fill-rate limits at that resolution.

CPU Role

The AMD Ryzen 5 5500 is a 6-core, 12-thread processor from the 5000 series, built on the Zen 3 architecture (Cezanne) using TSMC's 7 nm process. It has a base clock of 3.60 GHz and a boost clock of 4.20 GHz, with 16 MB of L3 cache and 512 KB of L2 cache per core. The CPU runs on the AM4 socket with DDR4 memory support and a dual-channel memory bus providing 51.2 GB/s of bandwidth. The 65 W TDP is modest for a desktop processor.

The CPU's benchmark results show it ranks in the 73rd percentile of all CPUs, with a single-thread score of 836 in 3DMark and 2319 in Cinebench R23. Its multi-thread scores include 16429 in Cinebench R23 multicore and 19330 in Passmark multithread. The nearest rivals are the AMD Ryzen AI 5 430 (19617, 0.1% ahead), the Intel Core i5-12500 (19668, 0.4% ahead), the Intel Core i7-10700F (19499, 0.5% behind), and the Intel Core i5-11600KF (19723, 0.7% ahead). These small deltas indicate the 5500 is competitive with mid-range processors from both AMD and Intel.

In Doom, the CPU's role becomes apparent at lower resolutions and settings. The 1080p Low result of 271 FPS is the highest measured value, and the gap between 1080p Low and 1440p Low is 95 FPS. If the CPU were the bottleneck at 1080p Low, the FPS would not scale as dramatically when moving to 1440p. The proportional drop of 35.1% from 1080p to 1440p at Low settings is similar to the 35.2% drop at Medium, suggesting the GPU is handling the increased pixel load consistently.

The 1080p Medium result of 236 FPS (minimum 201, maximum 272) shows that even at lower settings, the CPU has enough headroom to feed the GPU. The maximum of 272 FPS at 1080p Medium is nearly identical to the 1080p Low average of 271 FPS, which could indicate that the CPU is reaching its limit around 270 FPS. However, the 1080p Low average of 271 FPS and the 1080p Medium maximum of 272 FPS being nearly identical suggests that the CPU is the limiting factor for frame rates above 270 FPS in this game.

The multi-thread performance of the 5500, with a Passmark multithread score of 19330, is sufficient for Doom's workload. The game benefits from multiple cores, and the 12 threads provide adequate parallelism. The single-thread performance, indicated by a 3DMark single-thread score of 836, is also adequate, as the 1080p results show no signs of CPU stutter or frame pacing issues. The 4K results are uniformly GPU-limited, with even Low settings at 4K (95 FPS) being far below the CPU's apparent ceiling around 270 FPS.