Barony

The AMD Ryzen 9 9900X is a 12-core, 24-thread desktop processor built on the Zen 5 architecture, specifically the Granite Ridge die, and it sits in the 94th percentile of all CPUs tested. Its base clock is 4.40 GHz with a boost clock of 5.60 GHz, and it carries a 120W TDP. In Barony, a roguelike that is not typically considered a multi-threaded stress test, the CPU's raw single-thread strength is the primary driver of performance. The 3DMark single-thread score of 1286 and the Cinebench R23 single-core score of 6555 indicate strong per-core throughput, which directly translates to how quickly the game's logic and AI routines are processed. The processor's 64 MB of shared L3 cache and dual-channel DDR5 memory support further ensure that data flows quickly to the cores, minimizing stalls when the game needs to load new room layouts or enemy states.

The benchmark results show that the 9900X is a high-end part, with an average benchmark score of 54450 across all tested workloads. Its nearest rivals are the AMD Ryzen 9 9900X3D, which scores 54681 (a 0.4% difference), and the Intel Core i7-14700F, which scores 54097 (a 0.7% difference). This indicates that the 9900X sits in a tight pack at the top of the desktop CPU hierarchy; its performance in Barony will be virtually indistinguishable from these alternatives when the frame rate is not capped by the GPU. The processor’s high multi-threaded scores, such as the Cinebench R23 multi-core result of 46438, also mean it will handle any background tasks without compromising frame pacing, but in this specific game, the 12 cores are far more than necessary, making the single-core efficiency the deciding factor.

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

The Ryzen 9 9900X's 12 cores and 24 threads are a significant hardware resource, but Barony's gameplay—which involves procedural dungeon generation and turn-based or real-time combat with numerous entities—does not scale perfectly with such high core counts. The data shows that the CPU's single-thread performance is its key asset here. The 3DMark 2-thread score of 2519 and the 4-thread score of 4886 are strong, but the 8-thread score of 9114 and the max-thread score of 13929 highlight that scaling diminishes as more cores are engaged. This suggests that the game's primary thread is the bottleneck, and the 9900X's 5.60 GHz boost clock is what allows it to push high frame rates.

Looking at the measured FPS, at 1080p with Low settings, the combo achieves 388.1 FPS. This is a scenario where the CPU is fully untethered from GPU load, and the high clock speed of the 9900X is directly responsible for keeping the game logic running fast enough to not limit the frame rate. The Cinebench R15 single-core score of 660 and the Geekbench single-core score of 3341 confirm that this chip is among the fastest for single-threaded tasks. In contrast, when moving to Ultra settings, the frame rate at 1080p falls to 179.7 FPS, which is a 53.7% drop from the Low setting result. This drop is not due to CPU capability but rather the increased GPU load, showing that the CPU is capable of far higher throughput than the GPU can handle at maximum settings.

The processor's cache hierarchy also plays a role. The 80 KB L1 and 1 MB L2 per core, combined with the 64 MB shared L3, provide ample fast storage for the game's frequently accessed data structures. This reduces memory latency, which is crucial for a roguelike where every action depends on state checks. The 89.6 GB/s memory bandwidth is more than sufficient for the game's data needs, ensuring that the CPU never starves for data. In summary, the CPU role in this combo is to provide an extremely fast single-threaded baseline; the 9900X excels at this, making it a non-issue in Barony, allowing the GPU to become the primary determinant of performance at higher settings.

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

The NVIDIA GeForce RTX 3070 is the limiting factor in this pairing for Barony, particularly as resolution and settings increase. Based on the Ampere architecture and fabricated on Samsung's 8 nm process, this GPU features 8 GB of GDDR6 memory on a 256-bit bus, yielding 448.0 GB/s of bandwidth. Its boost clock is 1725 MHz, and it has 5888 shading units. The GPU's percentile standing is 72nd among all tested GPUs, which is respectable but far below the CPU's 94th percentile, indicating a clear imbalance in this combo for CPU-bound scenarios. The measured FPS data confirms this: at 1080p Low, the combo hits 388.1 FPS, which is an extreme load on the GPU's geometry and pixel pipelines, but at 4K Ultra, it falls to just 60.2 FPS.

The RTX 3070's 8 GB VRAM is a critical specification for Barony's higher settings. The game's texture resolution and shadow complexity increase with settings, and the data shows a significant performance cliff between Medium and Ultra at 4K. At 4K Medium, the combo achieves 98.2 FPS, but at 4K Ultra, it drops to 60.2 FPS. This 38 FPS drop is a 38.7% reduction, suggesting that the GPU is running out of memory bandwidth or capacity at Ultra settings. The 448.0 GB/s bandwidth is sufficient for 1080p and 1440p gaming, but at 4K, the pixel throughput demand is quadrupled, and the 8 GB frame buffer becomes a constraint. The GPU's FP32 performance of 20.31 TFLOPS is adequate for the game's shader complexity, but the data implies that memory limitations, rather than raw compute, are the bottleneck at high resolutions.

Compared to its rivals, the RTX 3070's average benchmark score is 28238. Its nearest rival, the AMD FirePro S7150, scores 28409 (a -0.6% difference), and the AMD Radeon RX 6600 XT scores 27985 (a 0.9% difference). This shows that the RTX 3070 is in a competitive tier, but it is not a top-tier card. The GTX 980 Ti, a much older card, scores 27956 (a 1% difference), which is remarkably close, indicating that the RTX 3070's performance advantage is slim. In Barony, the GPU's role is to handle the rendering pipeline; at lower settings, it can easily keep up with the CPU, but at 4K Ultra, it becomes the clear limiting component, preventing the high frame rates the CPU could otherwise support.

FAQ

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

A: The highest recorded average FPS is 388.1 at 1920x1080 with Low settings. This is the absolute peak measured performance, indicating a scenario where the GPU is lightly loaded and the CPU's speed is fully utilized.

Q: How does the combo perform at 4K resolution compared to 1080p?

A: At 4K with High settings, the average FPS is 81.1, which is significantly lower than the 249.8 FPS at 1080p High. This represents a 67.5% drop in frame rate, clearly demonstrating that the GPU becomes the primary bottleneck at higher resolutions.

Q: Is the Ryzen 9 9900X a bottleneck for the RTX 3070 in this game?

A: No, the data shows the opposite. At 1080p Low, the combo reaches 388.1 FPS, which is a very high frame rate. Since the CPU's single-thread score is in the 94th percentile, it is not limiting the GPU. The GPU's performance percentile of 72 is lower, indicating it is the limiting factor at higher settings.

Q: What are the FPS differences between Low and Ultra settings at 1440p?

A: At 2560x1440, the Low setting produces 245.9 FPS, while the Ultra setting produces 114.4 FPS. This is a 53.5% reduction in frame rate, showing the substantial impact of graphical settings on the GPU's workload.

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

A: The combo is ranked 762nd out of 1702 tested combinations. This places it in the upper-middle tier of all possible pairings, which is consistent with the RTX 3070 being a mid-range GPU paired with a high-end CPU.

Q: Does the 8 GB VRAM of the RTX 3070 affect 4K performance?

A: The data suggests it does. At 4K, the average FPS drops from 98.2 at Medium settings to 60.2 at Ultra settings, a 38.7% decrease. This steep drop indicates that the GPU is likely hitting memory capacity or bandwidth limits when handling the higher texture and shadow loads at 4K Ultra.

How This Combo Ranks

The AMD Ryzen 9 9900X + NVIDIA GeForce RTX 3070 combination achieves a rank of 762 out of 1702 tested combos in Barony. This places it in the 44.8th percentile of all combinations, meaning that roughly 55% of tested systems perform better in this specific game. This ranking is heavily influenced by the GPU's limitations. The CPU is a top-tier part, but the GPU's average benchmark score of 28238 is only in the 72nd percentile, pulling the combo's overall standing down. The data shows that at 1080p, where the CPU's speed can shine, the combo performs admirably, but at 4K, it falls behind many other pairings that feature more powerful GPUs.

The ranking is not static across all resolutions; it is a single overall rank, but the FPS data reveals the dynamic. At 1080p Low, the 388.1 FPS result would likely rank much higher than at 4K Ultra, where the 60.2 FPS result is more modest. The combo's position relative to its components' individual percentiles (CPU at 94th, GPU at 72nd) is a clear indicator that the GPU is the weak link. If the GPU were replaced with a higher-tier card, the combo's rank would improve substantially, as the CPU has ample headroom. However, as tested, the 762nd rank reflects a system that is excellent for high-refresh-rate 1080p and 1440p gaming but struggles to maintain high frame rates at 4K Ultra, which is a common trade-off for this type of mid-range GPU pairing.

Resolution Scaling

The resolution scaling data for this combo shows a clear and expected trend: as resolution increases, the average FPS decreases dramatically, and the limiting component shifts from the CPU to the GPU. At 1920x1080, the combo is largely CPU-bound at lower settings. The Low setting produces 388.1 FPS, while the High setting produces 249.8 FPS. The difference between these two is 138.3 FPS, which is a 35.6% reduction, indicating that even at 1080p, the GPU's workload is starting to impact performance. However, the CPU's high single-thread speed ensures that even at High settings, the frame rate is above 240 FPS, which is excellent for competitive play.

Moving to 2560x1440, the frame rates drop significantly. The Low setting yields 245.9 FPS, and the High setting yields 156.4 FPS. This is a 36.4% reduction from Low to High, similar to the 1080p scaling. The Ultra setting at 1440p produces 114.4 FPS, which is still playable but shows the GPU is now the primary constraint. Finally, at 3840x2160, the scaling becomes more severe due to the GPU's 8 GB VRAM and bandwidth limitations. The Low setting at 4K produces 127.6 FPS, but the High setting drops to 81.1 FPS, and the Ultra setting falls to 60.2 FPS. The drop from 1080p Low (388.1 FPS) to 4K Low (127.6 FPS) is a 67.1% reduction, which is a massive performance hit.

This scaling pattern reveals that the CPU is not the limiting factor at any resolution. The fact that the 4K Low setting (127.6 FPS) is higher than the 1080p Ultra setting (179.7 FPS) is counterintuitive but shows that the GPU's load is more dependent on settings than resolution. The data indicates that the RTX 3070 can handle the pixel count of 4K, but it cannot handle both high resolution and high settings simultaneously. The 60.2 FPS at 4K Ultra is a 84.5% drop from the 1080p Low result, demonstrating that the GPU's memory bandwidth and capacity are the primary bottlenecks. For users of this combo, the sweet spot is 1440p with High settings, where the 156.4 FPS result balances visual quality and performance, leveraging the CPU's strength without overburdening the GPU.