Minecraft: Java Edition

Resolution Scaling

The measured FPS data across the three resolutions reveals a clear pattern of CPU limitation at lower resolutions and a gradual shift toward GPU involvement as pixel count increases. At 1080p with Low settings, the combination produces 477.6 FPS, while at 4K with the same settings, the frame rate drops to 310.8 FPS. This 166.8 FPS difference represents a 35% reduction from 1080p to 4K, which is relatively modest compared to what a pure GPU-bound scenario would show.

Examining the Ultra preset reveals a different scaling behavior. At 1080p Ultra, the system delivers 338.1 FPS, which falls to 231.9 FPS at 1440p and further to 121.8 FPS at 4K. The step from 1440p to 4K here is a 110.1 FPS drop, or roughly 47%, indicating that the Ultra settings introduce enough rendering load to make the GPU the dominant factor. In contrast, the Low preset's 4K-to-1080p ratio suggests that even at 4K, the Core i5-10400F's six cores and twelve threads are still able to feed frames faster than the visual load would demand on a lesser processor.

The Medium preset sits between these extremes, with 440.4 FPS at 1080p, 375.3 FPS at 1440p, and 243.1 FPS at 4K. The scaling from 1080p to 1440p for Medium is only a 65.1 FPS drop, while the jump to 4K costs 132.2 FPS. This nonlinear scaling points to a threshold where the RTX 5090's rendering pipeline becomes saturated — once pixel count reaches 4K, even medium complexity scenes require enough GPU work that the frame rate drops more steeply. The data suggests that for 1080p and 1440p, the limiting component is largely the CPU's single-thread performance, given how close the Low and High presets remain to each other at those resolutions.

At 1080p, the spread between Low and Ultra is 139.5 FPS, while at 4K that spread grows to 189.0 FPS. Wider gaps at higher resolutions are the signature of GPU-bound behavior, since each preset tier adds shading work that scales with pixel area. The RTX 5090 with its 32 GB of GDDR7 memory and 1.79 TB/s bandwidth is clearly capable of handling these loads, but the frame rate ceilings at 1080p (477.6 FPS Low) are likely constrained by the i5-10400F's boost clock of 4.30 GHz and the game's dependence on single-threaded logic.

FAQ

Q: What is the highest average FPS recorded in the measured data?

A: The maximum average FPS is 477.6, achieved at 1920x1080 resolution with Low settings. The next highest is 440.4 at the same resolution with Medium settings, showing a 37.2 FPS penalty for the additional visual features.

Q: How much performance is lost when moving from 1080p to 4K on the Ultra preset?

A: At Ultra settings, the average FPS drops from 338.1 at 1920x1080 to 231.9 at 2560x1440, then to 121.8 at 3840x2160. The total loss from 1080p to 4K is 216.3 FPS, representing a 64% reduction in frame rate.

Q: Does the RTX 5090's 32 GB VRAM make a difference in this game's measured results?

A: Minecraft: Java Edition is not a VRAM-intensive title in the measured data, but the 32 GB GDDR7 capacity means memory capacity is never a limiting factor. The GPU's 104.8 TFLOPS FP32 throughput and 1.79 TB/s bandwidth are what enable the high frame rates seen at 4K, particularly on the High preset which still manages 195.8 FPS.

Q: How does the i5-10400F compare to its nearest rival in the benchmark database?

A: The i5-10400F has an average benchmark score of 14168, which places it 0.4% ahead of the AMD EPYC 7552 (14115) and 0.8% behind the AMD Ryzen 3 7320C (14277). It also trails the Intel Core i5-8500 by 1.1% and the AMD Ryzen Embedded V2546 by 1.2%.

Q: What is the FPS difference between Low and High settings at 1440p?

A: At 2560x1440, Low settings produce 408.6 FPS while High settings produce 343.7 FPS, a difference of 64.9 FPS. The Medium preset sits between them at 375.3 FPS, while Ultra drops to 231.9 FPS.

Q: Is the performance gap between Medium and High settings consistent across resolutions?

A: No. At 1080p, Medium (440.4 FPS) is 34.7 FPS ahead of High (405.7 FPS). At 1440p, the gap narrows to 31.6 FPS (375.3 vs 343.7). At 4K, the gap becomes 47.3 FPS (243.1 vs 195.8). The larger gap at 4K indicates that the GPU has more headroom to process the extra effects at higher pixel counts.

CPU Role

The Intel Core i5-10400F is a 6-core, 12-thread processor based on the Comet Lake architecture, fabricated on Intel's 14 nm process. Its base clock is 2.90 GHz with a boost clock of 4.30 GHz, and it shares 12 MB of L3 cache across all cores. The CPU's benchmark profile shows strong multi-threaded performance: a Cinebench R23 multicore score of 10297 and a 3DMark 16-thread score of 4748. However, Minecraft: Java Edition is known for its reliance on a primary thread, and the single-core scores reflect this bottleneck: Cinebench R23 single-core is 1453, and 3DMark single-thread is 688.

The measured FPS data confirms a CPU ceiling around 477.6 FPS at 1080p Low. This is not a GPU limitation — the RTX 5090 is far more capable than that. Instead, it aligns with the i5-10400F's PassMark single-thread score of 2541 and its percentile ranking of 73 among all CPUs. The fact that 1080p Low (477.6 FPS) and 1080p High (405.7 FPS) differ by only 71.9 FPS suggests that the game logic, not the rendering, is the primary constraint at this resolution. The CPU's 42.7 GB/s memory bandwidth and DDR4 support are adequate for feeding the GPU, but the architecture's single-thread limits become apparent in the frame rate ceiling.

The CPU's position in the benchmark rankings reinforces this interpretation. With an average benchmark score of 14168, it sits in the 73rd percentile of all CPUs, and its nearest rivals include the AMD EPYC 7552 (14115, 0.4% slower) and the AMD Ryzen 3 7320C (14277, 0.8% faster). These are not high-end parts by modern standards, yet the combination still delivers over 300 FPS at 4K Low. This indicates that the game's engine, when not constrained by heavy shader work, can be fed adequately by a six-core Comet Lake part. The 12 threads help with chunk loading and background simulation, which is why the multicore scores (Cinebench R20 multicore 4324, PassMark multithread 12115) are respectable, but the single-thread ceiling is the binding constraint.

Measured FPS Breakdown

At 1920x1080, the full preset ladder is available. Low settings yield 477.6 FPS, Medium drops to 440.4 FPS, High further reduces to 405.7 FPS, and Ultra bottoms out at 338.1 FPS. The step from Low to Medium costs 37.2 FPS, from Medium to High costs 34.7 FPS, and from High to Ultra costs 67.6 FPS. The Ultra preset's larger penalty suggests that it enables features like advanced lighting or shadow quality that disproportionately impact the CPU's draw call processing.

Moving to 2560x1440, the same preset ordering holds. Low produces 408.6 FPS, Medium 375.3 FPS, High 343.7 FPS, and Ultra 231.9 FPS. The delta between Low and Ultra at 1440p is 176.7 FPS, which is larger than the 139.5 FPS gap at 1080p. This widening gap indicates that at higher resolutions, the GPU's rendering load becomes more significant relative to the CPU's fixed cost per frame. The 1440p Ultra figure of 231.9 FPS is still well above typical high-refresh displays, but the drop from 1440p High (343.7 FPS) to 1440p Ultra (231.9 FPS) is a substantial 111.8 FPS penalty.

At 3840x2160, the results are 310.8 FPS for Low, 243.1 FPS for Medium, 195.8 FPS for High, and 121.8 FPS for Ultra. The 4K Ultra figure is the only measured result below 200 FPS, and it represents a 64% reduction from the 1080p Low maximum. The scaling from 1440p to 4K for each preset is: Low drops 97.8 FPS, Medium drops 132.2 FPS, High drops 147.9 FPS, and Ultra drops 110.1 FPS. The High preset shows the largest absolute drop, indicating that its particular combination of effects scales poorly with pixel count. The Low preset's relatively smaller drop (97.8 FPS) suggests that at minimal settings, the game is still partially CPU-bound even at 4K.

GPU Role

The NVIDIA GeForce RTX 5090 is built on the Blackwell 2.0 architecture with a GB202 chip fabricated on TSMC's 5 nm process. It contains 92,200 million transistors spread across a 750 mm² die, with 21,760 shading units, 680 texture mapping units, and 176 ROPs. The GPU operates at a base clock of 2017 MHz and boosts to 2407 MHz, with memory running at 1750 MHz (28 Gbps effective) across a 512-bit bus. This configuration yields 1.79 TB/s of memory bandwidth from 32 GB of GDDR7 VRAM. The GPU's FP32 throughput is 104.8 TFLOPS, and its pixel rate is 423.6 GPixel/s.

The RTX 5090's benchmark results place it in the 94th percentile of all GPUs, with an average benchmark score of 84306. Its nearest rival is the NVIDIA GeForce RTX 5090 D at 84241 (0.1% slower), followed by the RTX 5050 Mobile at 84171 (0.2% slower) and the AMD Radeon PRO W6600 at 83209 (1.3% slower). The RTX 5090 D is essentially identical in performance, while the RTX 5050 Mobile's proximity is notable given the different market segments. The 3DMark Steel Nomad DX12 score of 14411 and Geekbench OpenCL score of 380114 demonstrate the GPU's raw compute capability.

In this game, the RTX 5090's role becomes decisive at 4K Ultra, where the 121.8 FPS result is the only figure that approaches a GPU-bound scenario. The fact that 4K Low still yields 310.8 FPS means that even at 4K, the GPU has substantial headroom when visual effects are minimal. The GPU's 176 ROPs and 423.6 GPixel/s pixel rate are more than sufficient for the game's relatively simple geometry, but the Ultra preset's additional pixel shader work engages the GPU's 21,760 shading units more fully. The 1.79 TB/s bandwidth ensures that texture streaming never becomes a bottleneck, even with the game's procedural chunk generation. The GPU's launch MSRP is 1,999 USD.

Settings Recommendations

For users prioritizing maximum frame rates, the Low preset at 1080p delivers 477.6 FPS, which is the highest measured result. However, this comes at the cost of visual fidelity. The Medium preset at 1080p offers 440.4 FPS, which is only 37.2 FPS lower than Low while providing better visuals — this represents the best trade-off for high-refresh 1080p displays. The High preset at 1080p (405.7 FPS) is also acceptable, but the 34.7 FPS cost over Medium is harder to justify unless the specific High settings features are desired.

At 1440p, the Medium preset at 375.3 FPS is the recommended choice for most users. It stays above 360 FPS, which covers the vast majority of high-refresh monitors, while the High preset (343.7 FPS) offers a 31.6 FPS penalty for marginal visual gains. The Ultra preset at 1440p (231.9 FPS) should be avoided unless the user specifically wants maximum effects, as the 143.4 FPS drop from Medium is substantial. For 4K gaming, the High preset at 195.8 FPS is the sweet spot — it maintains a smooth experience above 144 FPS while offering better visuals than the Medium preset's 243.1 FPS would suggest necessary for playability. The Ultra preset at 4K (121.8 FPS) is only recommended for users with 120Hz displays who prioritize visual quality over frame rate headroom.

The data indicates that the Medium preset provides the most consistent experience across all three resolutions, with only a 197.3 FPS drop from 1080p to 4K (compared to High's 209.9 FPS drop and Ultra's 216.3 FPS drop). The Low preset has the smallest resolution penalty (166.8 FPS drop), but its minimal visuals are unlikely to be worth the frame rate gain for most users. The optimal strategy is to match the preset to the display's refresh rate: Medium for 1440p high-refresh, High for 4K high-refresh, and Ultra only for 4K displays with lower refresh rates or for users who prioritize visual fidelity above all else. The measured data shows that the RTX 5090 and i5-10400F combination is capable of exceeding 300 FPS at 4K Low, so even conservative settings choices leave ample performance headroom.