Barony

Barony is a roguelike that, despite its 2015 release, demonstrates surprisingly demanding performance characteristics when paired with the AMD Ryzen 5 7400F and NVIDIA GeForce RTX 3070 Ti. The measured data shows this combo produces a wide spread of frame rates across resolutions and settings, revealing distinct bottlenecks that shift depending on the workload imposed by the game. This analysis walks through the CPU and GPU roles, scaling behavior, and the practical implications of the measured FPS rows.

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

The AMD Ryzen 5 7400F is a 6-core, 12-thread processor based on the Zen 4 Raphael architecture, built on a 5 nm process at TSMC. It carries a base clock of 3.70 GHz and a boost clock of 4.70 GHz, with a 65 W TDP. The chip features 64 KB of L1 cache per core, 1 MB of L2 per core, and a shared 32 MB L3 cache. Memory support is limited to DDR5 over a dual-channel bus, providing 83.2 GB/s of bandwidth, and the CPU connects via PCIe Gen 5 with 24 lanes. It has no integrated graphics, which means all rendering falls to the discrete GPU.

In synthetic benchmarks, this processor scores 21,765 in Cinebench R23 multi-core and 3,072 in single-core, alongside a Passmark multithread score of 25,645 and a single-thread score of 3,689. Its average benchmark score is 32,750, placing it in the 83rd percentile of all CPUs. Compared to its nearest rivals, the Ryzen 5 7400F is effectively tied with the Intel Core i5-14600T (deltaPct 0.1%), the Intel Core Ultra 7 155H (0.2%), and the AMD Ryzen AI 7 PRO 360 (0.3%), while trailing the AMD Ryzen 7 PRO 6850H by only 0.2%. These margins are negligible, meaning the CPU's raw compute capability is not a differentiator against those chips.

For Barony, the CPU's role becomes apparent when examining the 1080p Low preset: the combo achieves 405 FPS average. This frame rate is so high that the CPU's single-thread performance — evidenced by a 3,072 Cinebench R23 single-core score — is likely the limiting factor, as the GPU would be underutilized at such low graphical demand. The 12 threads and high boost clock provide ample headroom for the game's logic and entity updates, which are typical bottlenecks in roguelikes. The L3 cache size also helps maintain consistent frame pacing during procedural generation and dungeon crawling. However, the data shows that the CPU does not become a hard ceiling until very high frame rates; at 1440p and 4K, the GPU takes over as the primary constraint.

FAQ — 4-6 Q&A pairs answerable from FACT PACK data

Q: What is the highest average FPS this combo achieves in Barony?

A: The highest measured average FPS is 405, which occurs at 1080p with Low settings. This is the only row exceeding 400 FPS, and it indicates the absolute ceiling for this pairing under the lightest graphical load.

Q: How does the combo perform at 4K Ultra compared to 4K Low?

A: At 4K Ultra, the average FPS drops to 61, while 4K Low produces 134 FPS. This represents a 73 FPS difference, or roughly a 2.2x improvement when lowering settings from the highest preset to the lowest at the same resolution.

Q: What is the minimum FPS recorded across all measured rows?

A: The only row with a reported minimum FPS is the Medium preset at all three resolutions. The lowest minimum is 84 FPS at 4K Medium, followed by 162 FPS at 1440p Medium and 256 FPS at 1080p Medium. No other settings rows include minimum FPS data.

Q: How does this combo rank among all tested combinations for Barony?

A: The combo ranks 1024th out of 2935 tested combinations, placing it in the 65th percentile (computed from the rank and total). This indicates that while it is above average, there are many faster combos in the database for this specific game.

Q: What is the frame rate difference between 1080p High and 1440p High?

A: At 1080p High, the average FPS is 259, while at 1440p High it is 163. This is a 96 FPS drop when moving from 1080p to 1440p, representing a 37% reduction in performance (computed from 96/259).

Q: Does the RTX 3070 Ti's 8 GB VRAM appear to limit performance at any settings?

A: The data does not show any VRAM-related stutters or crashes, as all measured rows report stable averages. At 4K Ultra, the 61 FPS average suggests the GPU's compute limits are the constraint, not memory capacity, since the game is a lightweight roguelike likely not exceeding 8 GB of VRAM at these settings.

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

Resolution scaling reveals a classic GPU-bound pattern for most settings, but the magnitude varies significantly. At Low settings, the average FPS goes from 405 at 1080p to 256 at 1440p (a 149 FPS drop, or 37% reduction) and then to 134 at 4K (a further 122 FPS drop, or 48% reduction). This steep decline indicates that even at Low settings, the GPU is the primary bottleneck once resolution increases, as the CPU's headroom at 1080p Low (405 FPS) is far beyond what the GPU can deliver at 4K.

At Medium settings, the pattern is similar but slightly less severe in absolute terms: 301 FPS at 1080p, 190 at 1440p, and 99 at 4K. The drop from 1080p to 1440p is 111 FPS (37%), and from 1440p to 4K is 91 FPS (48%). This consistent percentage drop across Low and Medium suggests the scaling is largely resolution-driven, with the GPU's pixel throughput being the limiting factor.

High settings show a different curve: 259 FPS at 1080p, 163 at 1440p, and 85 at 4K. The 1080p-to-1440p drop is 96 FPS (37%), but the 1440p-to-4K drop is 78 FPS (48%). Again, the relative scaling is consistent, reinforcing that the RTX 3070 Ti's rendering capacity dictates performance as resolution climbs.

Ultra settings provide the clearest evidence of GPU limitation: 185 FPS at 1080p, 117 at 1440p, and 61 at 4K. The 1080p-to-1440p drop is 68 FPS (37%), and the 1440p-to-4K drop is 56 FPS (48%). The percentage drops are nearly identical across all presets, which strongly indicates the limiting component is the GPU's fill rate and shader throughput, not the CPU or memory bandwidth. The CPU's 83.2 GB/s memory bandwidth and 32 MB L3 cache are sufficient to feed the GPU without introducing a CPU-side bottleneck at these resolutions.

How This Combo Ranks — use comboRankInGame vs other tested combos

The combo's rank of 1024 out of 2935 tested combinations places it in the upper-middle tier of all CPU-GPU pairs tested in Barony. This means approximately 65% of tested combos performed better, while 35% performed worse. Given the hardware in this pairing — a 6-core Zen 4 CPU and an RTX 3070 Ti — this ranking is consistent with a mid-to-high-end setup that is not top-tier but is far from entry-level.

To contextualize this rank, the CPU's percentile vs all CPUs is 83rd, while the GPU's percentile vs all GPUs is 75th. The combo's in-game rank of 1024/2935 is lower than either component's individual percentile would suggest, indicating that Barony may be more sensitive to other factors such as memory latency or driver optimization. However, the nearest rival GPUs offer a useful comparison: the RTX 3070 Ti's average benchmark score of 29,945 is nearly identical to the RTX 5070 Mobile (29,928, deltaPct 0.1%), slightly below the RX 6800 (30,095, -0.5%), and above the RTX 2080 Ti (29,783, 0.5%). These small deltas mean that swapping to any of these GPUs would likely produce very similar frame rates in Barony, with differences of less than 2% based on synthetic scores.

The CPU's rival deltas are even tighter, all within 0.3% of each other, so the processor choice is not a meaningful variable in this game's performance. The rank therefore reflects the overall balance of this specific combo, which sits in the 35th percentile of the tested pool, meaning it is beaten by a substantial majority of combinations but still delivers playable frame rates at all measured settings.

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

Based on the measured FPS data, the optimal settings depend on the target resolution and desired frame rate. At 1080p, all presets exceed 185 FPS, so the best experience is Ultra, which delivers 185 FPS average. This is well above the 144 Hz refresh rate common on high-end monitors, and the lack of minimum FPS data for Ultra suggests it does not dip below playable levels. For players prioritizing maximum smoothness, Medium at 1080p offers 301 FPS with a reported minimum of 256 FPS, ensuring no frame drops below 250 FPS.

At 1440p, Ultra still provides 117 FPS average, which is sufficient for 100+ Hz displays. Medium offers 190 FPS with a minimum of 162 FPS, which is a more consistent experience for competitive play. High sits in between at 163 FPS. For the most demanding users who want 4K, Ultra drops to 61 FPS, which is borderline for smooth gameplay but acceptable for a roguelike. Medium at 4K offers 99 FPS with a minimum of 84 FPS, making it the best balance of visual quality and performance at that resolution.

The data suggests that Medium is the sweet spot across all resolutions, as it is the only preset with reported minimum FPS values, indicating the testing methodology captured consistent frame delivery. At 1080p Medium, the minimum of 256 FPS ensures no perceptible stutter. At 1440p Medium, the minimum of 162 FPS is still above 144 Hz. At 4K Medium, the minimum of 84 FPS is acceptable for a turn-based or procedurally generated roguelike where reaction time is less critical. Therefore, Medium is recommended for the best experience per the measured rows, with Ultra reserved for users with G-Sync or FreeSync who can tolerate lower averages.

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

1920x1080 (1080p):

• Low: 405 FPS average, no min/max reported
• Medium: 301 FPS average, 256 FPS min, 346 FPS max
• High: 259 FPS average, no min/max reported
• Ultra: 185 FPS average, no min/max reported

2560x1440 (1440p):

• Low: 256 FPS average, no min/max reported
• Medium: 190 FPS average, 162 FPS min, 219 FPS max
• High: 163 FPS average, no min/max reported
• Ultra: 117 FPS average, no min/max reported

3840x2160 (4K):

• Low: 134 FPS average, no min/max reported
• Medium: 99 FPS average, 84 FPS min, 114 FPS max
• High: 85 FPS average, no min/max reported
• Ultra: 61 FPS average, no min/max reported

The pattern is consistent: each setting tier at a given resolution reduces FPS by a similar percentage. For example, moving from Low to Medium at 1080p reduces FPS from 405 to 301 (a 26% drop), while at 4K, the same transition drops from 134 to 99 (a 26% drop). Moving from Medium to High at 1080p reduces 301 to 259 (a 14% drop), and at 4K, 99 to 85 (a 14% drop). Moving from High to Ultra at 1080p reduces 259 to 185 (a 29% drop), and at 4K, 85 to 61 (a 28% drop). These consistent percentage drops indicate that the game's settings scale predictably with GPU load, with High-to-Ultra being the most demanding transition.

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

The NVIDIA GeForce RTX 3070 Ti is built on the Ampere architecture, using the GA104 chip on an 8 nm Samsung process. It has 8 GB of GDDR6X memory on a 256-bit bus, providing 608.3 GB/s of bandwidth. The GPU's base clock is 1575 MHz with a boost clock of 1770 MHz, and memory runs at 1188 MHz (19 Gbps effective). It features 6,144 shading units, 192 texture mapping units, 96 ROPs, 48 RT cores, and 192 tensor cores. Key throughput figures include 21.75 TFLOPS of FP32 performance, a pixel rate of 169.9 GPixel/s, and a texture rate of 339.8 GTexel/s. The card has a 290 W TDP and is end-of-life, having launched in May 2021.

In synthetic benchmarks, the RTX 3070 Ti scores 23,356 in Passmark G3D and 11,601 in GPU compute, with a 3DMark Steel Nomad DX12 score of 3,478. Its average benchmark score is 29,945, placing it in the 75th percentile of all GPUs. Compared to its nearest rivals, it is 0.1% faster than the RTX 5070 Mobile, 0.5% slower than the RX 6800, and 0.5% faster than the RTX 2080 Ti. These small deltas mean the 3070 Ti is a well-balanced mid-range card whose performance in Barony is representative of its class.

The GPU's role in Barony's measured results is clearly dominant at higher resolutions and settings. The 4K Ultra average of 61 FPS aligns with the GPU's compute limits, as the 21.75 TFLOPS FP32 throughput is taxed by the game's rendering at 4K with Ultra effects. The 8 GB VRAM is not a bottleneck in these tests, as no min FPS anomalies or stutters are reported beyond the normal variance seen in Medium presets. The 608.3 GB/s memory bandwidth provides sufficient headroom for texture streaming, and the 96 ROPs handle pixel fill rates adequately at 4K, as evidenced by the 61 FPS at Ultra. The consistent resolution scaling from 1080p to 4K confirms that the GPU's pixel and texture throughput are the limiting factors once the CPU's high frame rate ceiling is exceeded.