Barony

# Barony with Intel Core Ultra 5 235 + NVIDIA GeForce RTX 2070 SUPER

This combo lands at rank 1678 out of 2935 tested combinations in Barony, placing it in the 43rd percentile of all tested systems. That mid-pack positioning is somewhat surprising given the hardware involved—the RTX 2070 SUPER is a capable Turing-era card, and the Core Ultra 5 235 is a modern Arrow Lake processor. The data suggests that Barony’s roguelike engine rewards raw throughput in ways that don’t perfectly align with this pairing’s strengths, leaving meaningful headroom on both sides of the equation.

How This Combo Ranks

At rank 1678 of 2935 combos, this system sits just below the middle of the tested field. The percentile placement reveals a curious story: the GPU alone sits at the 65th percentile among all GPUs, while the CPU reaches the 89th percentile among all CPUs. Neither component is weak, yet their combined result in Barony underperforms what those individual percentiles might suggest. This gap implies that Barony’s workload—likely a mix of procedural generation, physics, and rendering—does not scale perfectly with this particular pairing.

The CPU’s nearest rivals in aggregate benchmarks show how tight the competition is at this performance tier. The AMD Ryzen AI 9 HX 375 scores within 0.1% of the Core Ultra 5 235, while the Intel Core i9-13900HX is 0.1% behind. The AMD EPYC 4364P and EPYC 7303 sit 0.2% ahead and behind, respectively. These sub-percent deltas mean that swapping the CPU for any of these alternatives would produce nearly identical Barony frame rates, all else being equal. The ranking position is therefore more a function of the GPU’s mid-tier status than any CPU bottleneck.

For the GPU, the nearest rivals cluster within 1.5% of the RTX 2070 SUPER’s average score. The Quadro M4000M is 1% ahead, the RTX 3070 Mobile is 1.2% behind, and the Intel Arc B570 and Arc A750 trail by 1.3% and 1.5%, respectively. In Barony, these small deltas translate to single-digit FPS differences at most settings, meaning the 2070 SUPER’s performance class is well understood: it’s a solid 1440p performer that struggles at 4K Ultra.

The ranking data also hints at diminishing returns. This combo at 1080p Low hits 316 FPS, which is far beyond any display’s refresh rate, yet at 4K Ultra it drops to 48 FPS—barely playable. The rank of 1678 reflects this uneven experience, where the system excels at lower resolutions but falls behind at the top end. Compared to the top combos in Barony, which likely pair high-end CPUs with flagship GPUs, this system’s mid-pack rank makes sense given the 8 GB VRAM ceiling and the 14-core CPU’s single-thread focus.

GPU Role

The RTX 2070 SUPER brings 8 GB of GDDR6 memory on a 256-bit bus, delivering 448.0 GB/s of bandwidth. The memory clock runs at 1750 MHz, or 14 Gbps effective, which is adequate for Barony’s texture-heavy roguelike environments. The GPU’s base clock is 1605 MHz with a boost of 1770 MHz, and it packs 2560 shading units, 160 TMUs, and 64 ROPs. These specifications place it firmly in the upper-midrange of the Turing generation, with 40 RT cores and 320 tensor cores available for ray tracing and DLSS—though Barony does not appear to leverage these features based on the measured results.

The VRAM capacity is the most consequential spec here. At 4K Ultra, the combo drops to 48 FPS, and the gap between 4K High (67 FPS) and 4K Medium (78 FPS) suggests that texture memory pressure begins to bite at higher settings. The 8 GB frame buffer is sufficient for 1080p and 1440p at any settings, but at 4K, Ultra presets likely exceed what the VRAM can comfortably hold, causing the frame rate to plummet by 29 FPS compared to High. The measured data shows a clear pattern: from Medium to Ultra at 4K, FPS drops from 78 to 48, a 38% reduction that points to memory bandwidth and capacity limits rather than pure compute.

The GPU’s compute capabilities are modest by modern standards—9.062 TFLOPS FP32 and 18.12 TFLOPS FP16 (2:1). Barony’s rendering workload appears to be fill-rate bound at higher resolutions, as evidenced by the pixel rate of 113.3 GPixel/s and texture rate of 283.2 GTexel/s. At 1080p, these rates are more than sufficient, allowing the GPU to hit 316 FPS on Low. But at 4K, the pixel throughput becomes a bottleneck, particularly with Ultra settings that enable additional effects. The 12 nm process node and 215 W TDP also mean this GPU runs hot and power-hungry by today’s standards, though the measured FPS data does not include thermal throttling metrics.

The GPU’s benchmark scores reinforce its mid-tier status. The PassMark G3D score of 18169 and 3DMark Steel Nomad DX12 score of 1651 are respectable but not exceptional. The Geekbench OpenCL score of 83358 and Vulkan score of 90637 indicate balanced compute performance across APIs. Notably, the PassMark DirectX 12 score of 67 is surprisingly low compared to DirectX 11’s 151, suggesting that Barony—which likely runs on an older engine—may not fully utilize the GPU’s DX12 capabilities, potentially leaving performance on the table.

CPU Role

The Intel Core Ultra 5 235 is a 14-core, 14-thread processor with no hyperthreading, which is a critical detail for Barony’s workload. The base clock of 3.40 GHz and boost clock of 5.00 GHz provide strong single-thread performance, as evidenced by the PassMark single-thread score of 4516 and Cinebench R23 single-core score of 2085. Barony, being a roguelike with procedural generation, likely benefits more from high single-thread speeds than from multi-thread throughput, since much of the game logic runs on one or two threads.

The CPU’s cache hierarchy includes 192 KB L1 per core, 3 MB L2 per core, and 24 MB shared L3. This configuration supports the high boost clocks effectively, as frequent cache hits reduce memory latency. The memory support for DDR5 with dual-channel bandwidth of 102.4 GB/s is ample for Barony’s needs, though the game’s data streaming might not fully utilize this bandwidth. The 14 threads (one per core) mean that heavily threaded sections of the engine—if any—could see contention, but the measured FPS suggests this is not a limiting factor.

The CPU’s multi-thread scores are strong but not class-leading. Cinebench R23 multi-core yields 14769, and PassMark multithread scores 37816. These numbers place it near the top of the desktop CPU hierarchy, with a percentile of 89 among all CPUs. However, the nearest rivals show that this performance tier is crowded: the Ryzen AI 9 HX 375 and i9-13900HX are within 0.1% in aggregate benchmarks. For Barony specifically, the CPU’s role is likely secondary to the GPU at higher resolutions, but at 1080p Low, the CPU must keep up with 316 FPS, which requires strong per-core performance.

The 65 W TDP and 3 nm process node (TSMC) indicate efficient power delivery, allowing sustained boost clocks without excessive heat. The integrated Arc Xe-LPG Graphics 24EU is present but irrelevant for gaming with a discrete GPU. The CPU’s launch MSRP is $257, though pricing is not a factor in performance analysis. The PCIe Gen 5 interface with 20 lanes provides ample bandwidth for the GPU, though the RTX 2070 SUPER uses PCIe 3.0 x16, so no bottleneck exists there.

In Barony, the CPU’s 14 cores may be underutilized if the game doesn’t scale beyond 4-6 threads. The single-thread performance is what matters most, and the boost clock of 5.00 GHz is among the highest available. The data shows that at 1080p Low, the combo hits 316 FPS, which is likely CPU-bound at that point—meaning the CPU can sustain very high frame rates when the GPU isn’t the constraint. At 4K Ultra, the CPU becomes irrelevant as the GPU takes over as the primary bottleneck, with FPS dropping to 48.

FAQ

Q: Is this combo better suited for 1080p or 1440p gaming in Barony?

A: The measured data shows 1080p High at 202 FPS and 1440p High at 127 FPS, both excellent. At 1080p Low, it reaches 316 FPS, while 1440p Low hits 199 FPS. Both resolutions are highly playable, but 1440p offers a better balance of visual quality and frame rate, with the Medium preset at 148 FPS being particularly strong.

Q: Why does the FPS drop so dramatically from 1440p to 4K?

A: At 1440p High the combo delivers 127 FPS, but 4K High drops to 67 FPS—a 47% reduction. This is primarily due to the GPU’s 8 GB VRAM and 256-bit memory bus, which struggle to sustain texture bandwidth at 4K. The 448.0 GB/s bandwidth is insufficient for the higher pixel count, especially with Ultra settings.

Q: Does Barony benefit more from the CPU or GPU in this setup?

A: At lower resolutions (1080p), the CPU’s 5.00 GHz boost clock drives frame rates up to 316 FPS, indicating CPU-bound behavior. At higher resolutions (4K), the GPU becomes the limiting factor, with FPS dropping to 48-104 depending on settings. The CPU’s 89th percentile ranking is higher than the GPU’s 65th percentile, so the GPU is the weak link.

Q: What is the best settings preset for a smooth experience at 4K?

A: The 4K Medium preset averages 78 FPS with a minimum of 66 FPS, which is the only 4K setting that maintains playable frame rates above 60 FPS. High drops to 67 FPS, and Ultra falls to 48 FPS. Medium provides the best balance of visual quality and performance at 4K.

Q: How does the RTX 2070 SUPER compare to its nearest GPU rivals in aggregate benchmarks?

A: The Quadro M4000M is 1% ahead, while the RTX 3070 Mobile is 1.2% behind, the Intel Arc B570 is 1.3% behind, and the Arc A750 is 1.5% behind. These sub-2% differences mean in Barony, the FPS variance between these GPUs would be negligible, typically 1-3 FPS at any given setting.

Q: Is the CPU’s lack of hyperthreading a problem for Barony?

A: The 14 threads (one per core) are not a hindrance in the measured results. Barony’s roguelike engine appears to scale well with single-thread performance, and the 5.00 GHz boost clock compensates for the lack of SMT. The 1080p Low result of 316 FPS demonstrates that thread count is not a bottleneck.

Resolution Scaling

The FPS scaling from 1080p to 4K tells a clear story about where the bottleneck lies. At High settings, 1080p delivers 202 FPS, 1440p delivers 127 FPS (37% drop), and 4K delivers 67 FPS (47% drop from 1440p). This scaling is roughly proportional to pixel count, indicating the GPU is the limiting factor at higher resolutions. The pixel count from 1080p to 1440p increases by 78%, yet FPS only drops by 37%, suggesting some headroom in the GPU’s fill rate. From 1440p to 4K, the pixel count increases by 125%, but FPS drops by 47%, which is more severe and points to VRAM bandwidth constraints.

At Low settings, the scaling is more dramatic: 1080p Low hits 316 FPS, 1440p Low drops to 199 FPS (37% drop), and 4K Low reaches 104 FPS (48% drop from 1440p). This pattern shows that even at low graphical load, the GPU cannot maintain frame rates at 4K, confirming that the memory bandwidth of 448.0 GB/s is the ceiling. The fact that 4K Low (104 FPS) is higher than 4K High (67 FPS) indicates that reducing texture quality and effects alleviates some memory pressure, but the 8 GB capacity still limits performance.

The Medium settings row includes min and max FPS data, which provides insight into frame time consistency. At 1080p Medium, the average is 235 FPS with a minimum of 200 and maximum of 270. At 1440p Medium, the average is 148 FPS with min 126 and max 171. At 4K Medium, the average is 78 FPS with min 66 and max 89. The min-to-max spread is consistent across resolutions (about 30% of average), indicating stable frame pacing without major stutters. However, the 4K Medium minimum of 66 FPS is close to the 60 FPS threshold, suggesting that any scene complexity spikes could cause noticeable dips.

The resolution scaling reveals that this combo is optimized for 1080p and 1440p, with 4K playable only at Medium settings. The CPU’s role diminishes as resolution increases: at 1080p Low, the CPU is likely the bottleneck at 316 FPS, but at 4K, the GPU’s pixel rate and memory bandwidth dominate. The 113.3 GPixel/s pixel rate is sufficient for 1080p at over 300 FPS but falls short for 4K at high settings, where the pixel count requires more than double the throughput.

Settings Recommendations

The measured data points to Medium as the sweet spot for this combo across all resolutions. At 1080p Medium, the average of 235 FPS with a minimum of 200 FPS provides a massive margin above any display refresh rate, ensuring no frame drops. At 1440p Medium, 148 FPS average with 126 FPS minimum remains well above 120 Hz displays, making it the ideal choice for high-refresh monitors. At 4K Medium, 78 FPS average with 66 FPS minimum is the only 4K preset that consistently stays above 60 FPS, making it the recommended setting for 4K gaming.

High settings offer diminishing returns. At 1080p High, 202 FPS is still excellent, but the jump from Medium to High provides minimal visual improvement in a roguelike game while costing 33 FPS. At 1440p High, 127 FPS is playable but the minimum FPS is not recorded, so the Medium preset’s guaranteed 126 FPS minimum makes it safer. At 4K High, 67 FPS average risks dropping below 60 FPS during intense scenes, making Medium the better choice.

Ultra settings are not recommended for this combo. At 1080p Ultra, 145 FPS is still smooth, but the visual gains over High are marginal in Barony’s pixel-art style. At 1440p Ultra, 91 FPS is acceptable but lacks the headroom of Medium. At 4K Ultra, 48 FPS is below playable thresholds for most users, and the lack of min/max data suggests potential instability. The 8 GB VRAM is insufficient for Ultra textures at 4K, as evidenced by the 29 FPS drop from High to Ultra.

Low settings are only useful for competitive play or extreme frame rate chasing. At 1080p Low, 316 FPS exceeds any display’s capabilities, making it pointless for most users. At 1440p Low, 199 FPS is still far above 144 Hz, but the visual downgrade is unnecessary. At 4K Low, 104 FPS is playable, but the Medium preset’s 78 FPS is worth the visual quality improvement.

Measured FPS Breakdown

1920x1080 (1080p):

• Low: 316 FPS average
• Medium: 235 FPS average, 200 FPS minimum, 270 FPS maximum
• High: 202 FPS average
• Ultra: 145 FPS average

At 1080p, the combo demonstrates its maximum capability. The Low preset’s 316 FPS shows the CPU’s single-thread strength, while Medium’s 235 FPS with a 200 FPS minimum indicates excellent frame pacing. High drops to 202 FPS, and Ultra to 145 FPS—still smooth but showing the GPU’s limits as settings increase.

2560x1440 (1440p):

• Low: 199 FPS average
• Medium: 148 FPS average, 126 FPS minimum, 171 FPS maximum
• High: 127 FPS average
• Ultra: 91 FPS average

1440p is where this combo shines. The Medium preset’s 148 FPS average and 126 FPS minimum make it ideal for 144 Hz monitors. Low’s 199 FPS is overkill, while High’s 127 FPS is still excellent. Ultra at 91 FPS is playable but the minimum is not recorded, so Medium remains the safest choice.

3840x2160 (4K):

• Low: 104 FPS average
• Medium: 78 FPS average, 66 FPS minimum, 89 FPS maximum
• High: 67 FPS average
• Ultra: 48 FPS average

4K performance is limited by the GPU’s 8 GB VRAM and 448.0 GB/s bandwidth. Medium is the only preset that stays above 60 FPS, with a 78 FPS average and 66 FPS minimum. High at 67 FPS is borderline, and Ultra at 48 FPS is not recommended. Low at 104 FPS shows that even minimal settings cannot overcome the memory bandwidth ceiling at 4K.

The data reveals a consistent pattern: Medium settings offer the best balance of visual quality and performance at every resolution. The 2070 SUPER’s 8 GB VRAM is the primary constraint at 4K, while the Core Ultra 5 235’s high boost clock ensures no CPU bottleneck at lower resolutions. This combo is best suited for 1440p Medium/High gaming, with 4K playable only at Medium settings.