Doom

Doom’s 2016 engine is famously scalable, and the measured results for the Intel Core i7-14700F paired with the NVIDIA GeForce RTX 5070 Ti confirm that this combination is heavily GPU-limited at higher resolutions, but the CPU still has enough headroom to push massive frame rates at 1080p. The data indicates a clear progression where lowering resolution yields disproportionately large FPS gains, which is the signature of a system where the graphics card is the primary bottleneck once pixel count rises.

Resolution Scaling

The frame rate scaling from 1080p to 4K reveals a textbook GPU-bound scenario, but with a twist at the highest settings. At High settings, the average FPS drops from 278 at 1920x1080 to 180 at 2560x1440, a 35% reduction in frame rate for a 77% increase in pixel count. Moving further to 3840x2160, the average falls to 98 FPS, which is a 45% drop from 1440p and a 65% drop from 1080p. This steep decline indicates that the RTX 5070 Ti’s rendering throughput is the limiting factor; the GPU simply cannot maintain the same level of output as pixel count quadruples from 1080p to 4K.

The pattern holds across all settings, but the magnitude of the drop varies. At Ultra settings, the average FPS goes from 227 at 1080p to 147 at 1440p, a 35% decrease, and then to 80 at 4K, a 46% drop from 1440p. The 4K Ultra result of 80 FPS is the lowest measured average in the entire dataset, showing that the combination of maximum visual fidelity and maximum resolution pushes the GPU to its limits. In contrast, at Low settings, the scaling is even more dramatic: 378 FPS at 1080p falls to 245 at 1440p (35% drop) and then to 133 at 4K (46% drop from 1440p). The fact that Low settings at 4K (133 FPS) is still faster than Ultra at 1440p (147 FPS) is a telling detail—the resolution has a larger impact on performance than the entire settings range.

The data strongly suggests that the RTX 5070 Ti is the limiting component at 1440p and 4K. If the CPU were the bottleneck, we would expect to see frame rates plateau as resolution decreases, because the CPU would be unable to feed the GPU with enough draw calls. Instead, we see continuous and significant FPS increases at lower resolutions, which means the GPU has spare capacity that is being utilized as the pixel workload shrinks. At 1080p, the frame rates are extraordinarily high—278 FPS at High, 329 FPS at Medium, and 378 FPS at Low—which suggests the GPU is still not fully saturated, but the CPU is likely starting to impose a ceiling around 380 FPS, as the jump from Medium (329 FPS) to Low (378 FPS) is the smallest relative gain between adjacent settings at that resolution.

FAQ

Q: Is the RTX 5070 Ti powerful enough for 4K gaming in Doom?

A: Yes, for smooth gameplay. The measured average FPS at 4K High is 98 FPS, which is well above the 60 FPS threshold for most displays. Even at 4K Ultra, the average is 80 FPS, which is still playable on high-refresh-rate monitors, though not at the highest refresh rates.

Q: What is the best resolution to use with a 240Hz monitor?

A: At 1920x1080, all settings except Ultra (227 FPS) are above 240 FPS. The 1080p Medium setting averages 329 FPS, and Low averages 378 FPS. At 2560x1440, only Low (245 FPS) exceeds 240 FPS, while Medium (213 FPS) and High (180 FPS) fall short. For a 240Hz display, 1080p is the safer choice for maintaining maximum frame rates.

Q: How much of a performance drop is there from High to Ultra settings?

A: The drop is consistent and substantial. At 1080p, the average FPS falls from 278 (High) to 227 (Ultra), a reduction of 18%. At 1440p, it drops from 180 to 147, a reduction of 18%. At 4K, it drops from 98 to 80, a reduction of 18%. The relative cost of Ultra over High is a consistent ~18% frame rate penalty across all resolutions.

Q: Does the CPU bottleneck at 1080p Low?

A: The data suggests it might. The jump from Medium (329 FPS) to Low (378 FPS) at 1080p is a 15% increase, while the jump from High (278 FPS) to Medium (329 FPS) is an 18% increase. The gains from lowering settings are starting to diminish, which hints that the CPU’s ability to process frames is becoming a limiting factor near 400 FPS, even though the Core i7-14700F has 20 cores and 28 threads.

Q: What is the frame time consistency like?

A: The only rows with min/max FPS data are the Medium settings. At 1080p Medium, the range is 280 to 379 FPS, with an average of 329 FPS. At 1440p Medium, the range is 181 to 245, with an average of 213. At 4K Medium, the range is 98 to 133, with an average of 115. The spread between min and max is roughly 30% of the average in each case, indicating some frame pacing variability, but the minimums are still high enough for smooth play.

Q: Should I use the Low preset to maximize competitive advantage?

A: Yes, if raw frame rate is the priority. The 1080p Low average of 378 FPS is the highest measured result in the dataset. At 1440p, Low (245 FPS) is also the best choice for high-refresh monitors. The trade-off is visual fidelity, but in a fast-paced FPS, the lower latency and smoother motion from higher FPS can be more important than graphical detail.

GPU Role

The NVIDIA GeForce RTX 5070 Ti is the dominant component in this system for Doom. Its 16 GB of GDDR7 memory on a 256-bit bus provides 896.0 GB/s of bandwidth, which is more than sufficient for the textures and effects in a 2016 title, even at 4K Ultra. The data shows no signs of a VRAM capacity or bandwidth limitation; the performance scaling is smooth and consistent with a purely compute-bound workload. The GPU’s 8960 shading units and 280 TMUs are the workhorses here, with the card’s boost clock of 2452 MHz ensuring high throughput.

The GPU’s benchmark scores align with its in-game performance. The passmark G3D score of 32974 places it in the 86th percentile of all GPUs, and its average benchmark score of 49957 is nearly identical to its nearest rival, the AMD Radeon RX Vega 64, which scores 50001 (a 0.1% difference). This close benchmark parity is not reflected in the game’s frame rates, which are specific to this title, but it does confirm that the RTX 5070 Ti is a high-end card. The 3DMark Steel Nomad DX12 score of 6604 further indicates strong DirectX 12 performance, which is relevant for Doom’s Vulkan and OpenGL paths, but the game’s engine is well-optimized enough to run at high FPS regardless.

The GPU’s role as the primary bottleneck is most evident at 4K Ultra, where the average FPS drops to 80. This is the only setting/resolution combo where the frame rate falls below 100 FPS, and it shows that the card’s raw compute power (43.94 TFLOPS FP32) is being fully utilized. In contrast, at 1080p Low, the GPU is clearly not the limiting factor, as the 378 FPS average suggests the card could push even higher if the CPU could keep up. The RTX 5070 Ti’s performance in this game is characterized by its ability to deliver high frame rates at 1440p and playable frame rates at 4K, with the 16 GB VRAM ensuring no texture pop-in or stutter from memory swapping.

Settings Recommendations

For the best balance of visual quality and performance, the Medium preset is the clear winner in this dataset. At 1440p, Medium averages 213 FPS, which is excellent for high-refresh 1440p monitors (typically 144Hz to 165Hz). The min FPS of 181 is still above 144Hz, ensuring a smooth experience. At 4K, Medium averages 115 FPS, which is a great target for VRR displays. The jump from Medium to High costs 18% of performance at every resolution, while the jump from Medium to Ultra costs even more, making Medium the sweet spot for this GPU.

If you have a 1080p 240Hz monitor, the High preset (278 FPS) is the best choice, as it provides a large visual upgrade over Medium (329 FPS) while still staying above 240 FPS. The Ultra preset at 1080p (227 FPS) is close to the 240Hz threshold but dips below it, which could cause screen tearing if VRR is not enabled. For 4K gaming, the High preset (98 FPS) is recommended over Ultra (80 FPS) because the visual difference between High and Ultra in Doom is often subtle, but the 18% performance gain is noticeable. The Low preset is only recommended for extreme competitive play where maximum FPS is the sole objective, as it sacrifices too much visual detail for a marginal FPS increase over Medium.

Measured FPS Breakdown

At 1920x1080, the frame rates are exceptionally high across the board. The Low preset leads with an average of 378 FPS, followed by Medium at 329 FPS (with a min of 280 and max of 379), High at 278 FPS, and Ultra at 227 FPS. The gap between Low and Medium is 49 FPS, while the gap between Medium and High is 51 FPS, and between High and Ultra is 51 FPS. This linear scaling in 50 FPS increments suggests the GPU is the primary limiter, but the smaller relative gain from Medium to Low (15%) compared to High to Medium (18%) hints at CPU involvement at the very top end.

At 2560x1440, the ranking remains the same but the numbers drop significantly. Low averages 245 FPS, Medium averages 213 FPS (min 181, max 245), High averages 180 FPS, and Ultra averages 147 FPS. The difference between Low and Medium is 32 FPS, Medium to High is 33 FPS, and High to Ultra is 33 FPS. The consistent 33 FPS gaps indicate a purely GPU-bound scenario, as the CPU has plenty of headroom at this resolution.

At 3840x2160, the frame rates are the lowest but still very playable. Low averages 133 FPS, Medium averages 115 FPS (min 98, max 133), High averages 98 FPS, and Ultra averages 80 FPS. Here, the gaps between settings shrink to 18 FPS between Low and Medium, 17 FPS between Medium and High, and 18 FPS between High and Ultra. This compression of the gaps at 4K is a classic sign of a GPU bottleneck, where the card’s compute resources are saturated and settings changes have a proportionally smaller absolute impact.

CPU Role

The Intel Core i7-14700F plays a supporting role in this system, but it is a highly capable one. With 20 cores and 28 threads, a boost clock of 5.40 GHz, and 33 MB of L3 cache, it has more than enough processing power to feed the RTX 5070 Ti in Doom. The CPU’s benchmark scores reinforce this: its Cinebench R23 multi-core score of 35122 and single-core score of 4958 are both strong, and its passmark multithread score of 41317 places it in the 91st percentile of all CPUs. The average benchmark score of 53620 is essentially tied with its nearest rival, the Intel Xeon 6505P (53701, -0.2% delta), showing that it is at the top of its class.

In Doom, the CPU’s influence is most apparent at 1080p Low, where the average FPS reaches 378. This is the only data point where the CPU could potentially be a limiting factor, as the GPU’s workload is light enough at Low settings that the CPU’s ability to issue draw calls and process game logic becomes more relevant. However, the fact that the frame rate still increases from Medium (329 FPS) to Low (378 FPS) suggests that the CPU is not a hard bottleneck, but rather that the system is balanced enough to scale well. The 20 cores and 28 threads are likely overkill for Doom, which is not heavily multithreaded, but the high single-core boost clock of 5.40 GHz ensures that the game’s main thread is well-served.

The CPU’s memory support for DDR4 and DDR5, with a dual-channel bus, provides ample bandwidth for game data. The 33 MB of shared L3 cache helps keep frequently accessed data close to the cores, reducing latency. The data shows that this CPU is not a limiting factor at 1440p and 4K, where the GPU is clearly the bottleneck. Even at 1080p High (278 FPS), the CPU has no trouble keeping up. The conclusion is that the Core i7-14700F is a well-matched companion for the RTX 5070 Ti in this game, providing a stable platform that allows the GPU to stretch its legs without any CPU-induced frame rate caps. The only scenario where the CPU might hold back the system is at 1080p Low, but even then, the 378 FPS average is far beyond what most displays can show.