Overpower

The AMD Ryzen 9 5950X paired with the Intel Arc B370 produces a decisively GPU-limited experience in Overpower, with the measured frame rates indicating that the integrated graphics solution is the primary constraint across all tested configurations. The data shows a consistent pattern where the CPU's substantial multi-threaded capabilities, evidenced by a Cinebench R23 multi-core score of 26017, are left largely underutilized as the Arc B370 struggles to maintain playable frame rates at higher resolutions and settings.

Resolution Scaling

The frame rate data reveals a steep performance cliff as resolution increases, a classic symptom of a system bound by its graphics processor rather than its central processor. At 1920x1080 with Low settings, the combo achieves an average of 65 FPS, which drops by 10.8% to 58 FPS at 2560x1440, and then plummets by another 15.5% to 49 FPS at 3840x2160. This scaling pattern indicates that the GPU's pixel throughput is the limiting factor, as the resolution increase from 1080p to 4K represents a 4x increase in pixel count, yet the frame rate only drops by 24.6% — a clear sign that the workload is not purely pixel-bound but also heavily dependent on other GPU resources.

The gap widens dramatically when examining higher quality presets. At Ultra settings, the average frame rate falls from 31 FPS at 1080p to 28 FPS at 1440p, and finally to 24 FPS at 4K. This 22.6% total drop from 1080p to 4K at Ultra is proportionally smaller than the 24.6% drop at Low settings, suggesting that at Ultra, the GPU is already saturated at 1080p and additional resolution has diminishing impact on performance. The Medium preset shows a similar trend: 41 FPS at 1080p, 36 FPS at 1440p, and 31 FPS at 4K, representing a 24.4% reduction across the resolution spectrum.

The High preset follows the same trajectory, with 38 FPS at 1080p dropping to 34 FPS at 1440p and then 28 FPS at 4K — a 26.3% decline. Interestingly, the Low preset shows the most linear scaling, with each resolution step costing roughly 10-15% performance. This suggests that at Low settings, the game is more sensitive to raw pixel throughput, whereas at higher settings, other GPU bottlenecks like texture fill rate and shading complexity become more dominant. The fact that the 1080p to 1440p transition consistently costs less performance than the 1440p to 4K transition across all presets indicates that the Arc B370's memory subsystem, which is system-shared and system-dependent, becomes increasingly strained as the frame buffer demands grow.

GPU Role

The Intel Arc B370 is the definitive performance bottleneck in this configuration. Its benchmark results place it in the 5th percentile of all GPUs, with a 3DMark Steel Nomad DX12 score of just 1184. This places it in direct competition with legacy parts like the ATI Mobility Radeon HD 5570 (avg score 1186, delta -0.2%) and the AMD Radeon HD 7650M (avg score 1192, delta -0.7%), while trailing the ATI Radeon HD 5770 by 0.5%. These comparisons make it clear that the Arc B370 is not designed for modern gaming workloads, despite its 1280 shading units, 40 TMUs, and 20 ROPs.

The GPU's clock behavior further compounds its limitations. With a base clock of 300 MHz and a boost clock of 2400 MHz, the Arc B370 shows a wide operating range, but its 25 W TDP and IGP form factor indicate that thermal and power constraints severely cap sustained performance. The pixel rate of 48.00 GPixel/s and texture rate of 96.00 GTexel/s are modest figures that directly explain why the system cannot maintain 60 FPS at any resolution above 1080p Low. The FP32 compute throughput of 6.144 TFLOPS is similarly insufficient for modern game engines that rely heavily on shader complexity.

The system-shared memory architecture is another critical weakness. With no dedicated VRAM, the Arc B370 depends entirely on the host system's DDR4 memory, which is limited to 51.2 GB/s bandwidth via dual-channel configuration. This creates a significant bottleneck, as the GPU must compete with the CPU for memory bandwidth. The measured FPS data reflects this: at 4K Ultra, the combo manages only 24 FPS, while at 1080p Low it reaches 65 FPS — a 41 FPS difference that underscores how memory bandwidth constraints amplify at higher resolutions. The DirectX 12 Ultimate support with feature level 12_2 is technically present, but the hardware lacks the raw throughput to leverage these features meaningfully. The 10 ray tracing cores are essentially decorative at this performance level, as enabling any form of ray-traced effects would likely render the game unplayable.

FAQ

*Q: Can this system run Overpower at a stable 60 FPS?*

A: No. The highest average FPS recorded across all tested configurations is 65 FPS at 1920x1080 with Low settings. All other resolutions and settings combinations fall below 60 FPS, with Ultra settings at 4K dropping to just 24 FPS.

*Q: What is the best resolution for this CPU-GPU combo in Overpower?*

A: The data shows that 1920x1080 is the only resolution where playable frame rates are achievable. At 1080p, the combo delivers between 31 FPS (Ultra) and 65 FPS (Low), whereas 1440p and 4K see maximum averages of 58 FPS and 49 FPS respectively, both at Low settings.

Q: How does the Intel Arc B370 compare to its nearest GPU rivals?

A: The Arc B370 scores 1184 in 3DMark Steel Nomad DX12, placing it within 0.7% of the ATI Mobility Radeon HD 5570 (1186), ATI Radeon HD 5770 (1190), and AMD Radeon HD 7650M (1192). It outperforms the AMD FirePro M2000 (1168) by 1.4%.

Q: Is the AMD Ryzen 9 5950X the bottleneck in this configuration?

A: No. The CPU's benchmark scores, including a Cinebench R23 multi-core score of 26017 and a Passmark multithread score of 45424, are far above what Overpower requires. The 16-core, 32-thread processor sits in the 91st percentile of all CPUs, indicating it is not the limiting factor.

Q: What is the FPS difference between Low and Ultra settings at 4K?

A: At 3840x2160, the Low preset averages 49 FPS, while the Ultra preset averages 24 FPS. This represents a 25 FPS difference, meaning Ultra settings deliver 51% lower performance compared to Low at 4K.

Q: Does the GPU have dedicated video memory?

A: No. The Intel Arc B370 uses system-shared memory, with both the memory size, type, and bus width listed as "System Shared." The bandwidth is "System Dependent," meaning performance scales with the host system's DDR4 memory configuration.

How This Combo Ranks

The combination of the AMD Ryzen 9 5950X and Intel Arc B370 ranks 2742nd out of 2980 tested combos in Overpower, placing it in the bottom 8% of all configurations tested by this database. This ranking is almost entirely attributable to the GPU, as the CPU's percentile rank of 91st among all processors would normally place a system much higher in gaming performance charts. The massive discrepancy between the CPU's capability and the GPU's output creates a severely imbalanced pairing.

To contextualize this rank, the CPU alone performs competitively with other high-end processors. The Ryzen 9 5950X's average benchmark score of 51947 is within 0.7% of the Intel Core i7-14700 (52301, delta -0.7%) and within 0.4% of the Intel Core i9-13900F (51730, delta 0.4%). It also trails the Intel Core Ultra 5 235HX by just 0.2% (52073) and the AMD EPYC 8124P by 0.3% (52121). These margins are negligible in real-world gaming scenarios, yet the GPU's performance in the 5th percentile pulls the entire system down to the 2742nd rank.

The data indicates that this combo is fundamentally mismatched for gaming. The CPU is capable of driving much more powerful graphics cards, as evidenced by its high multi-threaded scores like the 11597 in 3DMark max threads and the 624347 in Passmark data compression. However, the Arc B370's 1184 score in Steel Nomad effectively caps the system's gaming potential. For Overpower specifically, the combo's rank of 2742 out of 2980 means that 238 other tested configurations also fail to achieve smooth gameplay, but the overwhelming majority of systems perform better. This pairing would only be sensible for non-gaming workloads where the CPU's 16 cores and 32 threads can be fully utilized without GPU dependency.

Measured FPS Breakdown

At 1920x1080, the lowest resolution tested, the combo shows its best performance. Low settings deliver an average of 65 FPS, which is the only configuration in the entire dataset to exceed 60 FPS. Medium settings average 41 FPS, with a minimum of 34 FPS and maximum of 47 FPS. High settings average 38 FPS, while Ultra settings drop to 31 FPS. The 34 FPS gap between Low and Ultra at 1080p illustrates how dramatically settings impact performance on this GPU-limited system.

Moving to 2560x1440, the performance degradation is immediate. Low settings average 58 FPS, still technically playable but below the 60 FPS threshold. Medium settings average 36 FPS, with a minimum of 31 FPS and maximum of 42 FPS. High settings average 34 FPS, and Ultra settings fall to 28 FPS. The gap between Low and Ultra narrows to 30 FPS, suggesting that at 1440p, the GPU is becoming saturated even at lower settings.

At 3840x2160, the system struggles across the board. Low settings average 49 FPS, which is the only 4K configuration that approaches playability. Medium settings average 31 FPS, with a minimum of 26 FPS and maximum of 35 FPS. High settings average 28 FPS, and Ultra settings produce just 24 FPS. The 25 FPS difference between Low and Ultra at 4K is proportionally similar to the 1080p gap, but the absolute numbers are so low that even the "playable" Low setting fails to provide a smooth experience.

The Medium preset consistently shows the most variance between minimum and maximum FPS. At 1080p, the range is 34-47 FPS (13 FPS spread); at 1440p, it's 31-42 FPS (11 FPS spread); and at 4K, it's 26-35 FPS (9 FPS spread). This suggests that Medium settings introduce enough GPU load to cause frame time inconsistencies, while Low and Ultra settings are either too light or too heavy to exhibit similar variance.

Settings Recommendations

Based on the measured FPS data, the only viable configuration for a remotely playable experience is 1920x1080 with Low settings, which averages 65 FPS. This is the sole combination that exceeds the 60 FPS threshold, making it the recommended preset for any user attempting to play Overpower with this hardware. The 1080p Low setting also provides the best minimum FPS headroom, although the dataset does not include min/max values for this specific configuration.

For users who prioritize visual quality over frame rate, the 1080p Medium preset offers a compromise at 41 FPS average, with a minimum of 34 FPS and maximum of 47 FPS. This is marginally better than the High preset's 38 FPS average and significantly better than Ultra's 31 FPS. However, the 13 FPS variance in Medium settings indicates potential stuttering, which could detract from the action genre gameplay. The High preset at 1080p is a slight downgrade from Medium with no variance data available, making it a less predictable choice.

At higher resolutions, the data strongly advises against anything above Low settings. At 1440p, Low averages 58 FPS, which is close to playable but still below 60. Medium, High, and Ultra all fall below 40 FPS, making them unsuitable for action gameplay. At 4K, even Low settings at 49 FPS will feel sluggish, and the 24-31 FPS range for Medium through Ultra is effectively unplayable. The system-shared memory architecture exacerbates this, as the GPU's performance is directly tied to the host system's DDR4 bandwidth, which is already strained at higher resolutions.

The pragmatic recommendation is to run Overpower at 1920x1080 with Low settings for competitive play, or accept the Medium preset at 1080p if visual fidelity is non-negotiable. Any attempt to play at 1440p or 4K should be abandoned unless the user is willing to tolerate sub-60 FPS gameplay with significant compromises. The Ultra preset is entirely impractical at any resolution, as the 24-31 FPS range it produces is far below the smoothness required for an action game. Given the GPU's 5th percentile ranking and its proximity to decade-old parts like the ATI Radeon HD 5770, the hardware simply lacks the capability to handle Overpower at modern quality standards.