Barony

The Intel Core i5-10400F paired with the NVIDIA GeForce RTX 5090 D produces a fascinating performance profile in Barony, one defined by a stark contrast between raw GPU capability and the limitations of a 6-core Comet Lake processor. The data shows that this combination is capable of delivering extremely high frame rates, yet the scaling pattern across resolutions reveals that the CPU, not the GPU, becomes the decisive bottleneck as settings and resolution change. This analysis breaks down the measured FPS data to explain what this pairing means for players.

Resolution Scaling

The measured FPS data reveals a clear and unusual scaling pattern that directly identifies the limiting component in this system. At 1080p with Low settings, the combo achieves 474 FPS average. Moving to 1440p Low drops this to 406 FPS, and at 4K Low it falls to 239 FPS. The drop from 1080p to 4K at Low settings is a reduction of 235 FPS, or roughly a 50% decrease. This substantial decline when lowering resolution indicates that the GPU is being increasingly taxed at higher pixel counts, which is expected. However, the more telling metric comes from the High settings row.

At 1080p High, the average FPS is 403. At 1440p High, it is 292 FPS, and at 4K High, it is 152 FPS. The scaling from 1080p to 1440p shows a 111 FPS drop, and from 1440p to 4K, a further 140 FPS drop. The pattern here shows the GPU becoming the primary limiter as resolution increases, with the 4K High result being less than half of the 1080p High result. This is typical for a high-end GPU, where pixel throughput dominates at higher resolutions.

Yet the most revealing data point is the comparison between Low and High settings at the same resolution. At 1080p, the difference between Low (474 FPS) and High (403 FPS) is only 71 FPS. At 1440p, the difference between Low (406 FPS) and High (292 FPS) is 114 FPS. At 4K, the difference between Low (239 FPS) and High (152 FPS) is 87 FPS. The relatively small delta at 1080p suggests that the CPU is holding the system back from reaching even higher frame rates, as the GPU is capable of far more than 474 FPS in this scenario. The fact that 1080p Low and 1080p High are only 71 FPS apart, while 1440p Low and 1440p High are 114 FPS apart, indicates that at 1080p, the CPU is saturated and cannot feed the GPU enough frames to differentiate between settings significantly.

The scaling behavior points to a dual-phase bottleneck. At 1080p, the Intel Core i5-10400F with its 6 cores and 12 threads becomes the limiting factor, capping performance around 474 FPS regardless of how light the GPU load is. As resolution increases to 4K, the GeForce RTX 5090 D takes over as the primary constraint, evidenced by the steep drop to 152 FPS at High settings. This is not a balanced pairing; it is a system where the CPU limits at low resolutions and the GPU limits at high resolutions, with the crossover occurring somewhere between 1440p and 4K.

How This Combo Ranks

The combination of the Intel Core i5-10400F and the NVIDIA GeForce RTX 5090 D ranks 147th out of 2935 tested combos in Barony. This places it in the top 5% of all tested configurations, which is a strong overall position. However, this rank is somewhat misleading given the individual components involved. The RTX 5090 D alone sits in the 92nd percentile of all GPUs, with an average benchmark score of 77712. The CPU, by contrast, sits in the 68th percentile of all CPUs, with an average benchmark score of 14185. The combination rank of 147 suggests that the pairing performs well, but the CPU is clearly dragging the system down from what the GPU alone could achieve.

Looking at the nearest rivals for the CPU provides context for this bottleneck. The Intel Xeon 6756E has an average score of 14163, which is 0.2% lower than the i5-10400F. The AMD EPYC 7552 scores 14115, 0.5% lower. The Intel Core 7 160UL scores 14232, 0.3% higher, and the AMD Ryzen 3 7320C scores 14277, 0.6% higher. These rivals are all within a single percentage point of the i5-10400F, indicating that this CPU is a mid-pack performer. In the context of Barony, where the game relies heavily on single-threaded performance, this puts the system at a significant disadvantage when paired with a GPU that is in the top 8% of all GPUs.

The GPU's nearest rivals further illustrate the mismatch. The AMD Radeon RX 6650M XT scores 76904, which is 1.1% lower than the RTX 5090 D. The AMD Radeon RX 6850M XT scores 78940, 1.6% higher. The NVIDIA Tesla P100 PCIe 12 GB scores 79396, 2.1% higher, and the Tesla P100 PCIe 16 GB scores 79605, 2.4% higher. These rivals are all in the same performance tier as the RTX 5090 D, confirming that the GPU is a top-tier component. The fact that this top-tier GPU is paired with a CPU that sits in the 68th percentile explains why the combo rank of 147 is not higher. In a game like Barony, the CPU's single-thread performance is critical, and the i5-10400F's 3DMark single-thread score of 688 and Cinebench R23 single-core score of 1451 are modest figures that limit overall performance.

FAQ

Q: Does the Intel Core i5-10400F bottleneck the NVIDIA GeForce RTX 5090 D in Barony?

A: Yes, the data indicates a significant CPU bottleneck at lower resolutions. At 1080p Low, the system achieves 474 FPS, which is only 71 FPS higher than at 1080p High (403 FPS). This small delta suggests the CPU is capping performance, as the GPU is capable of much higher frame rates. The CPU's 68th percentile ranking versus the GPU's 92nd percentile ranking confirms this imbalance.

Q: What is the best resolution to play Barony with this combo?

A: The 1440p High setting offers the best balance between performance and visual quality. It delivers 292 FPS average, which is more than sufficient for smooth gameplay, while the scaling data shows that the GPU is not as heavily taxed as at 4K, where High settings drop to 152 FPS. At 1080p, the CPU bottleneck prevents the system from fully utilizing the GPU.

Q: How does this combo compare to other tested configurations?

A: This combo ranks 147th out of 2935 tested combos, placing it in the top 5%. This is a strong rank, but it is driven by the GPU's performance. The CPU is only in the 68th percentile of all CPUs, and its nearest rivals, such as the AMD EPYC 7552, are within 0.5% of its average score, indicating that it is not a top-tier processor.

Q: Is the RTX 5090 D the main contributor to the high FPS numbers?

A: Yes, the GPU is the primary driver of the high frame rates. The RTX 5090 D has an average benchmark score of 77712 and sits in the 92nd percentile of all GPUs. Its nearest rivals, such as the AMD Radeon RX 6850M XT, are within 1.6% of its score, confirming its top-tier status. The CPU, in contrast, is a mid-range part.

Q: What is the impact of the CPU's 6-core, 12-thread configuration on performance?

A: The CPU's configuration is adequate but not optimal for Barony. The 6 cores and 12 threads provide enough parallelism for the game, but the single-thread performance is the limiting factor. The Cinebench R23 single-core score of 1451 and the 3DMark single-thread score of 688 are modest, and in a game like Barony, which relies on single-threaded logic, this limits the maximum achievable FPS.

Q: What is the difference between Low and Ultra settings at 4K?

A: At 4K, Low settings average 239 FPS, while Ultra settings average 109 FPS. This is a difference of 130 FPS, or roughly a 54% reduction in performance. This large delta indicates that the GPU is heavily loaded at 4K, and the Ultra preset demands significantly more from the RTX 5090 D.

Settings Recommendations

Based on the measured FPS data, the optimal settings for this combo depend on the player's priority for frame rate versus visual fidelity. The data shows that the High preset at 1440p delivers 292 FPS average, which is an excellent experience for a fast-paced roguelike. This setting provides a significant visual upgrade over Low (406 FPS) while maintaining a frame rate that is well above the threshold for smooth gameplay. The Medium preset at 1440p offers 340 FPS, but the 48 FPS gain over High is marginal and likely not worth the reduction in visual quality.

At 1080p, the High preset at 403 FPS is the recommended choice. The Low preset at 474 FPS offers a 71 FPS increase, but the game's visual clarity benefits greatly from the High settings. The Ultra preset at 1080p drops to 331 FPS, which is still playable, but the 72 FPS reduction from High makes it a less attractive option unless the player demands maximum visual detail. The 1080p Medium preset at 425 FPS is a reasonable middle ground, but High provides better visuals for only a 22 FPS penalty.

At 4K, the situation is different. The High preset at 152 FPS is playable, but the Ultra preset at 109 FPS may cause noticeable dips in demanding scenes. The Low preset at 239 FPS is the best option for competitive play, but it sacrifices significant visual fidelity. For most players, the 4K Medium preset at 178 FPS offers the best balance, as it provides a solid frame rate while maintaining acceptable visuals. The data shows that the Medium preset has a min FPS of 151 and a max FPS of 204, indicating a stable experience, whereas the High and Ultra presets lack min/max data, suggesting they may be less consistent.

The best experience per the measured rows is the High preset at 1440p, delivering 292 FPS with no recorded min or max FPS data, implying a stable performance profile. This is the sweet spot where the GPU is not overwhelmed, and the CPU bottleneck is less pronounced than at 1080p.

Measured FPS Breakdown

The measured FPS data provides a comprehensive view of performance across three resolutions and four settings presets. At 3840x2160 (4K), the Low preset averages 239 FPS. The Medium preset averages 178 FPS, with a minimum of 151 FPS and a maximum of 204 FPS. The High preset averages 152 FPS, and the Ultra preset averages 109 FPS. This creates a clear hierarchy where each increase in settings results in a substantial FPS drop, with the difference between Low and Ultra being 130 FPS.

At 2560x1440 (1440p), the Low preset averages 406 FPS. The Medium preset averages 340 FPS, with a minimum of 289 FPS and a maximum of 391 FPS. The High preset averages 292 FPS, and the Ultra preset averages 209 FPS. The scaling here is less severe than at 4K, with the difference between Low and Ultra being 197 FPS. The Medium preset's min and max FPS data shows a range of 102 FPS, indicating some variability, but the average is solid.

At 1920x1080 (1080p), the Low preset averages 474 FPS. The Medium preset averages 425 FPS, with a minimum of 361 FPS and a maximum of 488 FPS. The High preset averages 403 FPS, and the Ultra preset averages 331 FPS. The difference between Low and Ultra is 143 FPS, which is smaller than at 1440p or 4K. This is the key indicator of the CPU bottleneck, as the GPU is not being fully utilized at this resolution. The Medium preset's min FPS of 361 and max FPS of 488 show the widest range across all settings, suggesting that the CPU is struggling to maintain consistent frame delivery at higher settings.

The data shows that the 1080p Low result of 474 FPS is the highest measured average across all configurations, while the 4K Ultra result of 109 FPS is the lowest. The spread between these two extremes is 365 FPS, which highlights the significant impact of both resolution and settings on this particular combo.

CPU Role

The Intel Core i5-10400F is a 6-core, 12-thread processor from the Core 10th Gen series, based on the Comet Lake architecture. It has a base clock of 2.90 GHz and a boost clock of 4.30 GHz, with a 65 W TDP. The CPU's cache hierarchy includes 64 KB of L1 cache per core, 256 KB of L2 cache per core, and 12 MB of shared L3 cache. This configuration is typical for a mid-range desktop processor, and its benchmark scores reflect that positioning.

In terms of raw compute, the CPU scores 10283 in Cinebench R23 multicore and 1451 in Cinebench R23 single-core. The Geekbench scores are 6257 for multicore and 1420 for single-core. The 3DMark tests show a single-thread score of 688, a 2-thread score of 1350, a 4-thread score of 2560, an 8-thread score of 3929, and a max-thread score of 4735. These scores indicate that the CPU scales well with additional threads, but its single-thread performance is modest. The PassMark single-thread score of 2541 and the multithread score of 12115 further confirm this profile.

The CPU's role in Barony is critical because the game, as a roguelike, likely has a significant amount of game logic that runs on a single thread. The data shows that at 1080p, the system is CPU-bound, with the 474 FPS Low result being the ceiling. The fact that the 1080p High result is only 71 FPS lower than Low suggests that the CPU is fully saturated and cannot deliver more frames regardless of the GPU's capacity. The 3DMark single-thread score of 688 is a key metric here, as it is below average for a modern desktop CPU, and this directly impacts the game's ability to process game state updates quickly.

The CPU's nearest rivals, such as the AMD EPYC 7552 with an average score of 14115, are all within 0.6% of the i5-10400F's score. This places the CPU in a crowded mid-range tier, where small differences in single-thread performance can have a noticeable impact on frame rates in CPU-bound scenarios. The 12 MB of shared L3 cache is sufficient for the game's data, but the 14 nm process node and the 2.90 GHz base clock are not current-generation. The CPU's 68th percentile ranking among all CPUs means that while it is not a weak processor, it is far from the top tier, and pairing it with a top-tier GPU creates an imbalance that is clearly visible in the measured FPS data. The system would benefit from a CPU with higher single-thread performance to fully unlock the RTX 5090 D's potential, especially at 1080p.