Doom

Resolution Scaling — how FPS drops from 1080p to 4K and what it says about the limiting component

The measured frame rates reveal a classic pattern of GPU-bound scaling as resolution climbs. At 1920x1080 with Low settings, the combo produces 288.8 FPS, but stepping up to 2560x1440 drops that to 188.1 FPS — a 34.9% reduction. Moving further to 3840x2160 yields 99.6 FPS, which is 47.0% lower than the 1440p result. This steep, consistent decline across each resolution step indicates the Radeon RX 6800 XT is the primary constraint at higher pixel counts, with the Core i9-14900K having ample headroom to feed frames at 1080p.

The scaling behavior becomes more telling when examining the Ultra preset. At 1080p Ultra, the system hits 173.5 FPS, which falls to 113.9 FPS at 1440p (−34.4%) and then to 62.7 FPS at 4K (−44.9%). The percentage drops are remarkably similar between Low and Ultra presets when moving from 1080p to 1440p, suggesting the GPU's shading and memory bandwidth are being taxed proportionally regardless of quality settings. However, the 4K step shows slightly steeper declines for Ultra compared to Low, implying that the combination of higher resolution and heavier effects pushes the GPU's 16 GB frame buffer and 512.0 GB/s bandwidth closer to their limits.

What is particularly interesting is the gap between Medium and High at each resolution. At 1080p, Medium delivers 253.3 FPS while High produces 214.2 FPS — a 15.4% difference. At 1440p, the gap narrows to 15.0% (165.0 vs 140.3 FPS), and at 4K it shrinks further to 18.3% (91.0 vs 74.3 FPS). This compression of performance differences at higher resolutions reinforces the GPU-bound narrative: when the render target grows, the cost of each quality tier becomes more evenly distributed across the pipeline, and the CPU's ability to pre-process frames becomes less relevant.

The data also exposes an interesting anomaly: the jump from High to Ultra at 1080p costs 40.7 FPS (214.2 to 173.5), but the same jump at 4K costs only 11.6 FPS (74.3 to 62.7). This suggests that Ultra settings introduce effects that are relatively more expensive at lower resolutions where the GPU has spare cycles, whereas at 4K the GPU is already saturated and the incremental cost of Ultra is partially hidden by the resolution bottleneck. For users targeting high refresh rates, the scaling curve indicates that 1080p is where this combo shines brightest, while 4K demands careful preset selection to maintain playable frame rates.

How This Combo Ranks — use comboRankInGame vs other tested combos

The Intel Core i9-14900K paired with the AMD Radeon RX 6800 XT achieves a combo rank of 594 out of 1681 tested combinations in Doom, placing it in the 64.7th percentile of all combos evaluated. This rank reflects a solidly above-average pairing, though not a top-tier configuration for this particular title. The position suggests that while the hardware is capable, there are numerous combos — likely featuring newer GPUs or higher-end RDNA 3 and Ada Lovelace parts — that outpace it in this game's specific rendering workload.

Looking at the individual components provides context for this ranking. The CPU holds a percentileVsAllCpus of 97, meaning it outperforms 97% of all processors in the database, and its nearest rival, the AMD EPYC 7413, scores nearly identically with a deltaPct of only −0.3%. The GPU, meanwhile, sits at percentileVsAllGpus of 87, with its closest competitor, the NVIDIA RTX A1000, coming in 1.7% higher in average benchmark score. The GPU's lower percentile relative to the CPU explains why the combo doesn't rank higher: the RX 6800 XT, while strong, is the limiting factor in this pairing, especially at higher resolutions where its 512.0 GB/s bandwidth and 20.74 TFLOPS of FP32 throughput are stretched.

The rank of 594 also implies that Doom is relatively forgiving on GPU resources compared to more modern titles, allowing the CPU's strengths to partially compensate. The i9-14900K's 24 cores and 32 threads, combined with a boost clock of 6.00 GHz, provide exceptional single-thread performance — evidenced by its Cinebench R23 single-core score of 7040 — which likely helps maintain high frame rates in the game's more CPU-sensitive moments. However, the data shows that at 4K Ultra, the combo drops to 62.7 FPS, which is below the 64.7th percentile rank might suggest for a "smooth" experience, indicating that the GPU's end-of-life status (productionStatus: "End-of-life") is beginning to show in demanding scenarios.

GPU Role — VRAM, clocks, and how they relate to this game's results

The Radeon RX 6800 XT brings 16 GB of GDDR6 memory on a 256-bit bus, yielding a bandwidth of 512.0 GB/s. In Doom, which was released in 2016 and is not particularly VRAM-hungry by modern standards, this capacity is more than sufficient — the measured frame rates show no signs of memory-related stuttering or texture thrashing at any resolution or preset. The GPU's base clock of 1825 MHz and boost clock of 2250 MHz, with a game clock of 2015 MHz, provide the raw throughput needed to sustain the high frame rates observed at 1080p and 1440p.

The GPU's architecture, RDNA 2.0 on a 7 nm process with 26,800 million transistors, delivers 20.74 TFLOPS of FP32 performance. This compute throughput is what drives the 288.8 FPS result at 1080p Low, but the game's rendering engine also relies on the GPU's 128 ROPs and 288 TMUs. The pixel rate of 288.0 GPixel/s and texture rate of 648.0 GTexel/s help explain why the transition from Medium to High at 1440p only costs 15.0% performance — the texture units have enough headroom to handle the increased detail without bottlenecking.

At 4K, the GPU's limitations become more apparent. The 512.0 GB/s bandwidth is likely the constraining factor when pushing 3840x2160 pixels, as the fill rate requirements scale quadratically with resolution. The fact that 4K Ultra yields 62.7 FPS while 4K Low produces 99.6 FPS — a 58.9% difference — indicates that the GPU's shading units are being pushed hard by the Ultra preset's additional lighting and post-processing effects, which demand more compute per pixel. The GPU's percentileVsAllGpus of 87 confirms it sits in the upper tier, but its nearest rivals, such as the NVIDIA RTX 4070 Ti SUPER (2.6% lower average score), suggest that newer architectures handle similar workloads more efficiently.

Settings Recommendations — which preset gives the best experience per the measured rows

Examining the measured FPS across all presets and resolutions, the Medium preset emerges as the most balanced choice for this combo. At 1080p, Medium delivers 253.3 FPS — a 15.4% improvement over High's 214.2 FPS and only 12.3% less than Low's 288.8 FPS. At 1440p, Medium's 165.0 FPS comfortably exceeds the 144 Hz refresh rate commonly found in gaming monitors, while High's 140.3 FPS sits just below that threshold. For 4K, Medium's 91.0 FPS provides a much smoother experience than High's 74.3 FPS, with the visual fidelity difference between Medium and High often being subtle in fast-paced FPS titles.

The Ultra preset is difficult to recommend for competitive play, as it drops to 173.5 FPS at 1080p — still high, but the 46.0% performance penalty compared to Low is significant for a game where frame rate directly impacts responsiveness. At 4K Ultra, the 62.7 FPS result may be acceptable for single-player exploration but falls short of the 90+ FPS target that many players prefer for smooth camera movement. The Low preset, while delivering the highest frame rates (288.8 FPS at 1080p), sacrifices too much visual detail for this hardware, which has the compute headroom to run higher settings without compromising playability.

The data suggests a tiered approach: for 1080p panels with 240 Hz refresh rates, High preset at 214.2 FPS is the sweet spot, as it stays above the refresh threshold while offering better image quality than Medium. For 1440p 144 Hz displays, Medium at 165.0 FPS is ideal, providing a 14.6% margin above the refresh rate to absorb frame-time spikes. For 4K 60 Hz displays, Medium at 91.0 FPS offers headroom above 60 FPS, allowing for V-Sync without drops. In all cases, the measured data shows that Medium provides the optimal balance between visual quality and performance for this GPU-CPU pairing.

Measured FPS Breakdown — resolution by resolution, settings by settings, exact numbers

At 1920x1080, the combo delivers its strongest performance. Low settings produce 288.8 FPS, Medium yields 253.3 FPS, High reaches 214.2 FPS, and Ultra bottoms out at 173.5 FPS. The spread from Low to Ultra is 115.3 FPS, representing a 66.5% performance range across presets. The step from Medium to High costs 39.1 FPS, while the step from High to Ultra costs 40.7 FPS — nearly identical penalties, suggesting that each quality tier adds a similar computational burden at this resolution.

At 2560x1440, the frame rates scale down predictably. Low produces 188.1 FPS, Medium delivers 165.0 FPS, High achieves 140.3 FPS, and Ultra drops to 113.9 FPS. The performance gap between Low and Ultra is 74.2 FPS, which is 39.4% of the Low value. The High to Ultra transition costs 26.4 FPS, while Medium to High costs 24.7 FPS — again showing a consistent per-tier penalty. Notably, 1440p Low still exceeds 1440p High at 1080p (188.1 vs 214.2 FPS), demonstrating that resolution has a larger impact on performance than preset quality.

At 3840x2160, the GPU becomes the clear bottleneck. Low yields 99.6 FPS, Medium reaches 91.0 FPS, High manages 74.3 FPS, and Ultra falls to 62.7 FPS. The Ultra result is the only setting below 60 FPS, but it's close to that threshold. The difference between Low and Medium at 4K is only 8.6 FPS, while Medium to High costs 16.7 FPS and High to Ultra costs 11.6 FPS. This non-linear scaling suggests that Medium and High presets at 4K are hitting different bottleneck points — possibly memory bandwidth for Medium and shading throughput for High. The 4K Low result of 99.6 FPS is particularly notable, as it shows the GPU can still push near-100 FPS at 4K when effects are minimized.

CPU Role — cores, clocks, and how they relate to this game's results

The Intel Core i9-14900K is a 24-core, 32-thread processor based on Raptor Lake architecture, with a base clock of 3.20 GHz and a boost clock of 6.00 GHz. Its performance in synthetic benchmarks is exceptional: Cinebench R23 multi-core score of 49872 and single-core score of 7040, along with a Geekbench multi-core score of 20370 and single-core score of 3064. These numbers place it in the 97th percentile of all CPUs, with its nearest rival, the AMD EPYC 7413, scoring only 0.3% lower on average. The CPU's L3 cache of 36 MB shared across cores and 2 MB L2 per core provide the low-latency data access that game engines like Doom's id Tech 6 rely on for draw call processing and physics calculations.

In the context of Doom, the CPU's role is primarily to feed the GPU with prepared frames. The game's Vulkan API support (the GPU lists Vulkan 1.4 support) allows for efficient multi-threading, which the i9-14900K's 32 threads can exploit. The measured frame rates at 1080p — where the CPU has more influence relative to the GPU — show that the processor is not a limiting factor: even at 1080p Low, the 288.8 FPS result is likely near the game's engine limits rather than a CPU bottleneck. The CPU's Passmark single-thread score of 4697 and multi-thread score of 58674 indicate it can handle the game's worst-case scenarios, such as large demon hordes or complex level geometry, without dropping frames.

The CPU's boost clock of 6.00 GHz is particularly relevant for Doom's single-threaded components, which handle player input and game logic. With a Cinebench R15 single-core score of 709, the i9-14900K outperforms most desktop processors in lightly-threaded workloads, ensuring that the game's critical path remains short. However, the data shows that at 4K, the CPU's performance becomes largely irrelevant — the GPU's 512.0 GB/s bandwidth and 20.74 TFLOPS are the determining factors, and the CPU's ample headroom does not compensate for the GPU's resolution-driven workload increase. The CPU's 125 W TDP and support for DDR4 and DDR5 memory (dual-channel) provide flexibility for system builders, but the measured results indicate that for Doom specifically, the GPU is the component to upgrade first if higher 4K frame rates are desired.

FAQ

Q: What is the best resolution for this combo in Doom?

A: The measured data shows 2560x1440 with Medium settings delivers 165.0 FPS, which is ideal for 144 Hz displays. At 1920x1080, High settings produce 214.2 FPS, suitable for 240 Hz panels. For 4K, Medium at 91.0 FPS is the highest preset that comfortably exceeds 60 FPS.

Q: How does this combo rank compared to other tested systems?

A: It ranks 594th out of 1681 tested combos in Doom, placing it in the 64.7th percentile. The CPU is in the 97th percentile of all CPUs, while the GPU is in the 87th percentile, indicating the GPU is the limiting factor for this pairing.

Q: Is the RX 6800 XT's 16 GB VRAM sufficient for Doom at 4K?

A: Yes, the measured results show no VRAM-related issues at any resolution or preset. The 16 GB GDDR6 memory with 512.0 GB/s bandwidth is more than enough for Doom, which was released in 2016 and does not require large texture pools.

Q: What frame rate can I expect at 4K with Ultra settings?

A: The measured average FPS at 3840x2160 with Ultra settings is 62.7 FPS. This is above 60 FPS but leaves little headroom for frame-time spikes, so Medium at 91.0 FPS is recommended for a more consistent experience.

Q: How does the i9-14900K compare to its closest CPU rivals?

A: The i9-14900K's average benchmark score is 79824, which is 0.3% lower than the AMD EPYC 7413 (80041), 0.7% higher than the Intel Core i9-14900KF (79245) and Core Ultra 7 265KF (79240), and 0.9% higher than the Core Ultra 7 265K (79102).

Q: Does raising settings from Low to Ultra affect performance more at 1080p or 4K?

A: At 1080p, the difference between Low (288.8 FPS) and Ultra (173.5 FPS) is 115.3 FPS, a 66.5% drop. At 4K, the difference is 36.9 FPS (99.6 vs 62.7), a 58.9% drop. The absolute penalty is larger at 1080p, but the relative penalty is actually smaller at 4K, showing the GPU is already saturated at higher resolutions.