Barony

# Barony with AMD Ryzen 9 9900X3D + Intel Arc B390: A Severely GPU-Limited Roguelike Experience

The AMD Ryzen 9 9900X3D paired with the Intel Arc B390 delivers playable but heavily constrained performance in Barony, ranking 2601 out of 2935 tested combos. The data reveals a system where the CPU is vastly overpowered for the task, while the GPU sits in the 9th percentile of all tested graphics cards, creating a bottleneck that dominates every resolution and settings tier. Benchmark results indicate that this combination is only suitable for 1080p play at medium settings or lower, with 4K Ultra proving effectively unplayable at 24 FPS.

FAQ

Q: What is the overall performance ranking of this combo in Barony?

A: This combination ranks 2601 out of 2935 tested combos, placing it in the bottom 11.4% of all tested systems for this game. The low rank is driven almost entirely by the GPU, which sits at the 9th percentile among all tested graphics cards.

Q: How does the Intel Arc B390 compare to its nearest rivals?

A: The Arc B390 scores 1482 in the 3DMark Steel Nomad DX12 test, placing it 1.3% ahead of the NVIDIA GeForce GT 520MX (1463), 1.5% ahead of the GeForce 800M (1460), 2.5% ahead of the GT 625 OEM (1446), and 2.7% ahead of the GT 710 (1443). These are extremely small margins against very low-end hardware.

Q: What is the best settings/resolution combination for playable frame rates?

A: The data shows that 1080p Low delivers 159 FPS average, 1080p Medium delivers 118 FPS, and 1080p High delivers 102 FPS. For 1440p, only Low (100 FPS) and Medium (75 FPS) remain comfortably playable, while High drops to 64 FPS. At 4K, no settings tier reaches 60 FPS.

Q: How does the Ryzen 9 9900X3D compare to its nearest CPU rivals?

A: The Ryzen 9 9900X3D has an average benchmark score of 54762, which is 0.6% below the Intel Core Ultra 5 245KF (55093), 1.3% above the Intel Core Ultra 5 245K (54053), 2% above the Intel Xeon 6505P (53701), and 2.1% above the Intel Core i7-14700F (53620). It sits at the 91st percentile among all CPUs.

Q: Is the Ryzen 9 9900X3D the limiting factor in this system?

A: No. The CPU ranks in the 91st percentile with 12 cores and 24 threads, while the GPU ranks in the 9th percentile. The massive disparity between the CPU's capabilities and the GPU's performance means the graphics card is the sole bottleneck in every scenario tested.

Q: Can this combo handle 4K gaming in Barony?

A: Technically yes, but not well. At 4K Low, the system averages 53 FPS, but Medium drops to 39 FPS, High to 34 FPS, and Ultra to 24 FPS. The 4K Medium setting shows a minimum of 33 FPS, indicating significant frame time variance.

GPU Role

The Intel Arc B390 is the dominant constraint in this system, and the benchmark data makes this unmistakably clear. Built on the Xe3-LPG architecture with a Panther Lake chip, this integrated graphics processor (IGP) features 1536 shading units, 48 texture mapping units, and 24 raster output units. The GPU's base clock runs at 300 MHz with a boost clock of 2500 MHz, producing a pixel rate of 60.00 GPixel/s and a texture rate of 120.0 GTexel/s. Its FP32 performance is rated at 7.680 TFLOPS, with FP16 at 15.36 TFLOPS.

The memory configuration is particularly telling: the card uses "System Shared" memory with "System Dependent" bandwidth, meaning it borrows from the CPU's DDR5 pool rather than having dedicated VRAM. The bus width is also listed as "System Shared," which introduces additional latency and bandwidth contention. The GPU's TDP is 80 W, and it uses no power connectors, consistent with its IGP classification.

In the 3DMark Steel Nomad DX12 benchmark, the Arc B390 scores 1482, placing it at the 9th percentile of all tested GPUs. Its nearest rivals are all ancient or entry-level parts: the GT 520MX, GeForce 800M, GT 625 OEM, and GT 710. The margin over these parts ranges from 1.3% to 2.7%, showing that the Arc B390 is not meaningfully faster than a decade-old entry-level discrete GPU.

For Barony, this GPU performance translates directly into the measured frame rates. The game's roguelike nature typically involves procedurally generated dungeons with relatively simple geometry, but the GPU still cannot maintain high frame rates at higher resolutions. The data shows that the GPU's limits are reached well before the CPU's capabilities are stressed, making this a pure graphics-bound scenario across all tested configurations.

Resolution Scaling

The frame rate progression from 1080p to 4K reveals a system that is entirely GPU-limited. At Low settings, the average FPS drops from 159 at 1080p to 100 at 1440p (a 37.1% decrease) and then to 53 at 4K (a 47% decrease from 1440p). This steep decline with resolution scaling is the classic signature of a GPU bottleneck, where each additional pixel places direct pressure on the rasterization and shading units.

At Medium settings, the progression is 118 FPS at 1080p, 75 FPS at 1440p, and 39 FPS at 4K. The percentage drops are 36.4% from 1080p to 1440p and 48% from 1440p to 4K. High settings show 102 FPS at 1080p, 64 FPS at 1440p, and 34 FPS at 4K, with drops of 37.3% and 46.9% respectively. Ultra settings yield 73 FPS at 1080p, 46 FPS at 1440p, and 24 FPS at 4K, representing drops of 37% and 47.8%.

The consistency of these percentage drops across all settings tiers is remarkable. Whether the system is running Low or Ultra, the resolution scaling penalty remains nearly identical, confirming that the bottleneck is purely resolution-driven. If the CPU were the limiting factor, we would expect to see smaller FPS differences between resolutions, particularly at lower settings where the CPU has more headroom. Instead, the data shows that the GPU is saturated at every resolution, with the 4K Ultra result of 24 FPS representing the absolute limit of this hardware.

The 1440p Medium setting provides the only measured minimum and maximum frame rates alongside the 4K Medium and 1080p Medium results. At 1440p Medium, the system ranges from 64 to 86 FPS, while 4K Medium ranges from 33 to 45 FPS. These ranges show that lower resolutions not only improve average FPS but also tighten the frame time distribution, making the experience smoother at 1440p compared to the wider swings seen at 4K.

Measured FPS Breakdown

At 1920x1080, the system delivers its best results. Low settings produce an average of 159 FPS, making this the only configuration that comfortably exceeds 144 Hz refresh rates. Medium settings average 118 FPS with a minimum of 101 and maximum of 136, providing a smooth experience for 100 Hz displays. High settings average 102 FPS, still excellent for most monitors. Ultra settings drop to 73 FPS, which remains playable but shows the GPU's limitations emerging even at 1080p.

Moving to 2560x1440, the frame rates become more moderate. Low settings average 100 FPS, which is still smooth, while Medium averages 75 FPS with a range of 64 to 86. High settings average 64 FPS, right at the edge of comfortable playability. Ultra settings fall to 46 FPS, which introduces noticeable stutter in fast-paced roguelike gameplay where quick reactions matter.

At 3840x2160, the system struggles across all settings. Low averages 53 FPS, barely above the 30 FPS playability threshold. Medium averages 39 FPS with a minimum of 33 and maximum of 45, showing that frame rates can dip into the low 30s during intense scenes. High averages 34 FPS, and Ultra averages just 24 FPS, which is below the threshold for acceptable playability in most titles.

The pattern across all resolutions shows that the settings tier has a consistent impact. From Low to Ultra at 1080p, the FPS drops from 159 to 73, a 54.1% reduction. At 1440p, the drop is from 100 to 46, a 54% reduction. At 4K, the drop is from 53 to 24, a 54.7% reduction. This consistency suggests that the GPU's architecture scales predictably with workload, but the absolute performance ceiling is simply too low for higher resolutions.

How This Combo Ranks

This combination ranks 2601 out of 2935 tested combos, placing it in the bottom 11.4% of all systems tested in Barony. This poor ranking is directly attributable to the GPU's 9th percentile standing among all graphics cards, which outweighs the CPU's 91st percentile performance. The data shows that the CPU is capable of far more than the GPU can deliver, but the overall system performance is capped by the weakest link.

The CPU's average benchmark score of 54762 places it just below the Intel Core Ultra 5 245KF (55093, 0.6% ahead) and above the Intel Core Ultra 5 245K (54053, 1.3% behind), Intel Xeon 6505P (53701, 2% behind), and Intel Core i7-14700F (53620, 2.1% behind). These tiny margins around the 54,000-55,000 score range show that the Ryzen 9 9900X3D is competitive with mid-range modern CPUs, but this CPU performance is irrelevant when the GPU cannot keep up.

The GPU's 3DMark Steel Nomad score of 1482 is only 1.3% above the GT 520MX, 1.5% above the GeForce 800M, 2.5% above the GT 625 OEM, and 2.7% above the GT 710. These rivals are all legacy parts from more than a decade ago, and the Arc B390 barely edges them out. In a game like Barony, which is not graphically demanding by modern standards, the GPU still cannot deliver high frame rates at higher resolutions.

Given the ranking data, this combo is best suited for 1080p gaming with modest settings. The 102 FPS average at 1080p High is the sweet spot, providing good visual quality with smooth performance. Pushing to 1440p requires dropping to Low settings for 100 FPS, and 4K is only marginally playable at Low settings with 53 FPS. The ranking reflects this reality: the system is adequate for entry-level gaming but cannot compete with even mid-range discrete GPU configurations.

CPU Role

The AMD Ryzen 9 9900X3D is a 12-core, 24-thread processor from the 9000 series, built on the Zen 5 architecture with the Granite Ridge codename. It operates at a base clock of 4.40 GHz with a boost clock of 5.50 GHz, and features a large 128 MB L3 cache alongside 1 MB L2 cache per core. The CPU supports DDR5 memory with dual-channel configuration and a memory bandwidth of 89.6 GB/s. It is manufactured on TSMC's 4 nm process with 16,630 million transistors across a dual-die design measuring 2x 70.6 mm².

In synthetic benchmarks, the CPU demonstrates strong performance. It scores 47737 in Cinebench R23 multicore, 20049 in Cinebench R20 multicore, and 4811 in Cinebench R15 multicore. Single-core scores are also respectable: 6739 in Cinebench R23, 2830 in Cinebench R20, and 679 in Cinebench R15. Geekbench results show 22884 multicore and 3041 single-core. The PassMark suite shows 56154 multithread and 4646 single-thread, with strong results in specific workloads like integer math (179727), floating point math (119622), and data compression (685074).

These CPU benchmarks place the processor at the 91st percentile among all CPUs, confirming that it is a high-end part. However, the CPU's role in Barony is minimal because the game is not CPU-intensive in the way that strategy or simulation titles might be. The 3DMark 16-thread score of 12490 and max-thread score of 13795 show that the CPU can handle heavy multi-threaded workloads, but Barony's roguelike gameplay likely doesn't stress more than a few cores.

The 128 MB L3 cache is a notable feature that could theoretically benefit game performance, but in this GPU-limited scenario, the cache advantage is never realized. The CPU is essentially idle waiting for the GPU to render frames, which is why the combo ranking is so low despite the CPU's high percentile. If the system were paired with a more capable GPU, the Ryzen 9 9900X3D would likely enable much higher frame rates, but with the Arc B390, the CPU's potential is wasted.

The benchmark data shows that the CPU's average score of 54762 is nearly 37 times higher than the GPU's score of 1482 in their respective benchmarks. This disparity explains why the system underperforms: the CPU is capable of supporting high-end gaming, but the GPU cannot deliver the rendering performance needed to take advantage of it. For users considering this combo, the takeaway is clear: the GPU is the definitive limiting factor, and upgrading it would yield far greater performance gains than any CPU change.