Barony

The Intel Core Ultra 5 125H paired with the NVIDIA GeForce RTX 4090 produces a combination that ranks 610th out of 2,935 tested system configurations for Barony. This places the combo in the 79th percentile of all tested setups, indicating that it outperforms the majority of systems in this specific roguelike title. The ranking reflects a system that is heavily weighted toward GPU capability, as the RTX 4090 sits in the 88th percentile among all GPUs, while the Core Ultra 5 125H holds a 79th percentile position among all CPUs. The data shows a configuration where the graphics card is the dominant component, and the processor is capable enough to avoid creating a significant bottleneck in this particular game.

GPU Role

The NVIDIA GeForce RTX 4090 is the central component driving performance in Barony. Its 24 GB of GDDR6X memory on a 384-bit bus provides a bandwidth of 1.01 TB/s, which is more than sufficient for the graphical demands of this roguelike title. The GPU operates at a base clock of 2235 MHz and boosts to 2520 MHz, with memory running at 1313 MHz (21 Gbps effective). These specifications contribute to a pixel rate of 443.5 GPixel/s and a texture rate of 1,290.2 GTexel/s, figures that explain why the system achieves such high frame rates even at demanding settings.

The RTX 4090's 16,384 shading units, 512 texture mapping units, and 176 raster output units provide substantial raw compute power. Its FP32 performance of 82.58 TFLOPS ensures that even the most particle-heavy or procedurally generated scenes in Barony are rendered without difficulty. The GPU also features 128 ray tracing cores and 512 tensor cores, though these are less relevant for a game of this genre, which typically relies more on rasterization and CPU-side logic. The card's 5 nm process node from TSMC and 76,300 million transistors on a 609 mm² die allow for this high level of performance within a 450 W TDP envelope.

Benchmark results for the RTX 4090 show a Passmark G3D score of 38,194 and a Passmark GPU Compute score of 26,613, with a Geekbench OpenCL score of 255,416 and a Vulkan score of 271,631. These synthetic benchmarks align with the measured in-game performance, confirming that the GPU is operating near its expected capability. The GPU's percentile ranking of 88th among all graphics cards means it outperforms 87% of tested GPUs, and its average benchmark score of 60,347 places it in close competition with the Intel Arc Pro A60 (deltaPct of 0%) and the AMD Radeon Pro Vega 48 (deltaPct of 0.3%). The data indicates that for Barony, the RTX 4090 provides an exceptionally high ceiling for frame rates, with the limiting factor being the CPU's ability to feed it with draw calls and game logic updates.

Resolution Scaling

The measured FPS data reveals how the system scales across resolutions, highlighting the relationship between GPU load and CPU capability. At 1080p with Low settings, the system achieves 471 FPS, while at 4K with Low settings, it drops to 156 FPS. This represents a 67% reduction in frame rate when moving from 1080p to 4K, which is a substantial drop but still results in highly playable performance. The scaling from 1080p to 1440p at Low settings shows a decrease from 471 FPS to 298 FPS (a 37% reduction), while the 1440p to 4K transition at Low settings drops from 298 FPS to 156 FPS (a 48% reduction).

At Medium settings, the frame rates are 351 FPS at 1080p, 221 FPS at 1440p, and 116 FPS at 4K. The drop from 1080p to 1440p is 37%, and from 1440p to 4K is 47%, following a similar pattern to Low settings. High settings show 301 FPS at 1080p, 190 FPS at 1440p, and 99 FPS at 4K, with reductions of 37% and 48% respectively. Ultra settings produce the lowest frame rates: 216 FPS at 1080p, 136 FPS at 1440p, and 71 FPS at 4K, with drops of 37% and 48%.

The consistent percentage drops across all settings, regardless of the absolute FPS value, suggest that the scaling is primarily driven by GPU workload rather than CPU limitations. When the resolution increases, the rendering resolution scales the GPU workload directly, and the frame rate drops proportionally. The fact that the system can still produce 71 FPS at 4K Ultra indicates that the GPU is not being fully saturated at lower resolutions; instead, the CPU becomes the limiting factor at 1080p and 1440p. At 1080p Low, the 471 FPS figure suggests the CPU is reaching its maximum game logic throughput, as the GPU could likely render more frames if the processor could provide the necessary draw calls. This is corroborated by the 471 FPS at 1080p Low versus the 156 FPS at 4K Low, a 3.0x difference, while the resolution pixel count increases by 4.0x, indicating that the GPU is scaling closer to linear at higher resolutions.

Measured FPS Breakdown

The measured frame rates for this combination in Barony are as follows, organized by resolution and settings:

1920x1080 (1080p):

• Low settings: 471 FPS average
• Medium settings: 351 FPS average, with a minimum of 298 FPS and maximum of 403 FPS
• High settings: 301 FPS average
• Ultra settings: 216 FPS average

2560x1440 (1440p):

• Low settings: 298 FPS average
• Medium settings: 221 FPS average, with a minimum of 188 FPS and maximum of 254 FPS
• High settings: 190 FPS average
• Ultra settings: 136 FPS average

3840x2160 (4K):

• Low settings: 156 FPS average
• Medium settings: 116 FPS average, with a minimum of 98 FPS and maximum of 133 FPS
• High settings: 99 FPS average
• Ultra settings: 71 FPS average

The minimum FPS data available for Medium settings at each resolution provides insight into frame time consistency. At 1080p Medium, the minimum of 298 FPS versus the average of 351 FPS indicates a 15% drop below the average during demanding scenes. At 1440p Medium, the minimum of 188 FPS versus the average of 221 FPS shows a similar 15% drop. At 4K Medium, the minimum of 98 FPS versus the average of 116 FPS represents a 16% dip. These consistent percentages suggest that the frame time variance scales with the overall performance level, maintaining a similar relative stability across resolutions.

The progression from Low to Ultra at each resolution shows a clear relationship. At 1080p, moving from Low to Medium reduces FPS by 25%, from Medium to High by 14%, and from High to Ultra by 28%. At 1440p, the reductions are 26%, 14%, and 28% respectively. At 4K, the reductions are 26%, 15%, and 28%. The consistent percentage reductions across resolutions indicate that each settings tier imposes a similar relative cost on the GPU, regardless of the absolute resolution.

FAQ

Q: What is the best resolution to run Barony with this CPU/GPU combination?

A: The data shows that at 4K Ultra settings, the system achieves an average of 71 FPS, which is above the 60 FPS threshold for smooth gameplay. If frame rates above 100 FPS are preferred, 4K Medium provides 116 FPS, while 1440p High delivers 190 FPS.

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

A: At Low settings, the average FPS drops from 471 at 1080p to 156 at 4K, a 67% reduction. At Ultra settings, the drop is from 216 FPS to 71 FPS, a 67% reduction as well. The scaling is consistent across all settings tiers.

Q: Is the RTX 4090 being fully utilized at 1080p?

A: No. At 1080p Low, the system reaches 471 FPS, but the GPU is capable of higher performance. The CPU, with its 14 cores and 18 threads, becomes the limiting factor at lower resolutions where the GPU is not fully saturated.

Q: How does the Intel Core Ultra 5 125H compare to its nearest CPU rivals?

A: The Core Ultra 5 125H has an average benchmark score of 27,507, which is 0.1% below the AMD Ryzen 7 5800 (27,535) and 0.3% below the Intel Core i5-12600K (27,578). It is 0.5% above the AMD Ryzen 3 PRO 5355G (27,382).

Q: What settings provide the most consistent frame pacing?

A: The Medium settings at each resolution have the only available minimum FPS data. At 1080p Medium, the minimum of 298 FPS is 15% below the average; at 1440p Medium, the minimum of 188 FPS is 15% below average; and at 4K Medium, the minimum of 98 FPS is 16% below average. This consistency suggests Medium settings provide predictable performance.

Q: Can this system handle Barony at 4K with high refresh rates?

A: At 4K High settings, the average FPS is 99, which exceeds the 60 Hz refresh rate but falls short of 144 Hz. For 144 Hz at 4K, the system would need to run at Low settings, which achieves 156 FPS.

Settings Recommendations

Based on the measured FPS data, the optimal settings depend on the target refresh rate and resolution. For users with a 60 Hz display at 4K, Ultra settings at 71 FPS provide the highest visual quality while staying above the refresh rate threshold. For those with a 120 Hz display at 4K, Medium settings at 116 FPS are the best choice, as they exceed the refresh rate and offer a significant visual improvement over Low settings, which achieve 156 FPS.

At 1440p, the data shows that High settings at 190 FPS provide a balanced experience for 144 Hz displays, while Ultra at 136 FPS is suitable for 120 Hz panels. For 165 Hz or higher refresh rate displays at 1440p, Medium settings at 221 FPS or Low settings at 298 FPS would be necessary. The 1440p Medium settings also offer the only minimum FPS data in this resolution range, with a floor of 188 FPS, which maintains a comfortable margin above most high refresh rate thresholds.

At 1080p, the system is clearly CPU-limited, as even Ultra settings at 216 FPS exceed the refresh rate capabilities of most monitors. The data suggests that for 1080p, users can select Ultra settings without concern for frame rate drops below 200 FPS, making it the recommended choice for maximum visual fidelity. The Medium settings at 1080p provide a minimum FPS of 298, which ensures frame rates never drop below 298 FPS even in the most demanding scenes, making it the safest choice for competitive play where frame time consistency matters. However, the visual difference between Medium and Ultra at this resolution is likely more noticeable than the difference between 216 FPS and 351 FPS, so Ultra is recommended for most users.

CPU Role

The Intel Core Ultra 5 125H plays a supporting role in this configuration, providing sufficient processing power to keep the RTX 4090 fed with data. This processor features 14 cores and 18 threads, with a base clock of 3.60 GHz and a boost clock of 4.50 GHz. Its architecture is Meteor Lake, built on Intel's 7 nm process node, and it has an 18 MB shared L3 cache along with 2 MB L2 cache per core and 112 KB L1 cache per core. The CPU's TDP is 28 W, which is modest for a mobile processor, yet it delivers benchmark scores that place it in the 79th percentile of all CPUs.

In Cinebench R23, the Core Ultra 5 125H scores 12,804.5 in multi-core and 1,647.5 in single-core tests. Its Passmark multithread score is 21,100, with a single-thread score of 3,399. The CPU's average benchmark score of 27,507 places it in close competition with several desktop processors, including the AMD Ryzen 7 5800 (27,535, deltaPct of -0.1%) and the Intel Core i5-12600K (27,578, deltaPct of -0.3%). These comparisons show that despite its mobile form factor and 28 W TDP, the Core Ultra 5 125H performs on par with older desktop CPUs, providing adequate single-threaded performance for Barony's game logic.

The CPU's memory bandwidth is 89.6 GB/s via dual-channel DDR5 support, which is sufficient for the data throughput required by this roguelike genre. The 18 threads allow for background tasks and game logic threads to run concurrently without contention. In Barony, which is a turn-based roguelike with real-time elements, the CPU's role is to handle pathfinding, procedural generation, and game state updates. The measured FPS at 1080p Low (471 FPS) indicates that the CPU can sustain very high game logic update rates, though the exact frame rate ceiling is not documented in the data. The fact that the system achieves 216 FPS at 1080p Ultra, where the GPU is more heavily loaded, suggests that the CPU is not a limiting factor for this game at standard resolutions. However, the 471 FPS at 1080p Low likely approaches the CPU's maximum throughput, as further increasing the frame rate would require faster single-threaded performance.