Control

The AMD Ryzen 5 5500X3D and Intel Arc B370 pairing lands at a very specific performance tier in Control, one defined by the Arc B370’s integrated-graphics-class output. The CPU is a 6-core, 12-thread Zen 3 part with a 3.00 GHz base clock and a 4.00 GHz boost clock, built on a 7 nm process with 96 MB of shared L3 cache. That large cache is the headline feature for gaming, but in this combo, the processor is rarely the limiting factor. The Arc B370, with 1280 shading units, 40 TMUs, and 20 ROPs, is an IGP-class GPU with a 25 W TDP and system-shared memory. Its 3DMark Steel Nomad DX12 score sits at 1184, placing it in the 5th percentile of all GPUs.

The benchmark results confirm this imbalance. Across every measured resolution and settings preset, the frame rates are consistently low, and the pattern of scaling points directly at the GPU as the bottleneck. The CPU’s PassMark multithread score of 20363 and single-thread score of 2941 are adequate for this title, but the Arc B370 cannot push enough pixels to utilize that headroom. The data shows a system where CPU capability is held far in reserve, with the integrated GPU dictating a ceiling that the processor never approaches.

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

The Ryzen 5 5500X3D’s role in Control is best described as a spectator. With 6 cores and 12 threads, it offers the kind of parallel throughput that modern action-adventure titles can exploit, but the measured FPS numbers never rise high enough to tax it. The CPU’s boost clock of 4.00 GHz provides solid per-thread performance, and the 96 MB L3 cache is designed to reduce memory latency in gaming workloads. However, in this particular pairing, the cache advantage is irrelevant because the GPU cannot deliver the frame rate needed to make CPU cache misses a factor.

The PassMark physics score of 2282 and the integer math score of 60033 indicate a processor capable of handling the game’s simulation and AI workloads without breaking a sweat. The floating-point math score of 34511 further reinforces that the CPU’s compute throughput is not the constraint. In contrast, the Arc B370’s FP32 performance of 6.144 TFLOPS is the real limiter. The GPU’s pixel rate of 48.00 GPixel/s and texture rate of 96.00 GTexel/s sound respectable on paper, but they are achieved at very low clock speeds, with a base of 300 MHz and a boost of 2400 MHz, and the system-shared memory bandwidth is the Achilles’ heel.

The benchmark data shows that at 1080p Low, the combo reaches 59 FPS, which is the highest measured average in the entire set. This suggests that even at the lowest settings, the GPU is barely managing to stay ahead of the CPU’s minimum frame delivery. The CPU’s single-thread score of 2941 is more than enough for a game like Control, which is not known for being CPU-bound. The nearest rival comparison for the CPU shows it trading blows with the AMD Ryzen 5 5600F (0.2% delta) and the Intel Core Ultra 5 225 (0.2% delta), meaning the processor is competitively positioned against its peers. But those peers are not the issue here; the Intel Arc B370 is.

The Arc B370’s nearest rivals, based on 3DMark Steel Nomad scores, are ancient parts like the ATI Mobility Radeon HD 5570 and the ATI Radeon HD 5770. The deltaPct values are all within -0.7% to 1.4% of the Arc B370’s score, which places this Intel GPU in the same performance bracket as hardware from over a decade ago. That context is crucial: the CPU is a modern, capable part, but the GPU anchors the entire system to legacy-level performance. In Control, the CPU’s job is to feed the GPU with draw calls and physics data, and it does so flawlessly. The problem is that the GPU can only render a fraction of what the CPU sends.

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

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

A: The highest measured average FPS is 59, achieved at 1920x1080 with Low settings. This is the only setting and resolution pairing that reaches near-60 FPS territory, and it represents the absolute ceiling for this configuration.

Q: How does the CPU compare to its closest rivals?

A: The Ryzen 5 5500X3D has an average benchmark score of 37018, which puts it 0.2% ahead of the AMD Ryzen 5 5600F and the Intel Core Ultra 5 225, and 0.2% behind the AMD Ryzen AI 7 PRO 450. It is also 0.3% behind the Intel Core i9-12900T, showing that it is competitively matched with a wide range of processors.

Q: Is the GPU or the CPU more responsible for the low frame rates?

A: The GPU is overwhelmingly the limiting component. The Arc B370’s 3DMark Steel Nomad score of 1184 places it in the 5th percentile of all GPUs, and its nearest rivals are the ATI Mobility Radeon HD 5570 and ATI Radeon HD 5770. The CPU’s 85th percentile ranking among all CPUs indicates it is far more capable than the GPU.

Q: What is the FPS drop from 1080p to 4K at High settings?

A: At High settings, the average FPS drops from 37 at 1920x1080 to 28 at 2560x1440, and then to 19 at 3840x2160. This represents a 48.6% reduction from 1080p to 4K, showing severe scaling challenges at higher resolutions.

Q: Does the CPU’s large L3 cache help in this game?

A: The 96 MB shared L3 cache is a significant feature of the Ryzen 5 5500X3D, but the measured results show no evidence of it providing a benefit in Control. The frame rates are consistently low across all settings, indicating that the GPU’s memory bandwidth and compute limits prevent the CPU cache from making a meaningful difference.

Q: What is the combo’s rank among all tested configurations?

A: The combo ranks 2657 out of 2888 tested combinations in Control, placing it in the bottom 8% of all tested systems. This ranking reflects the GPU’s severe performance limitations rather than any deficiency in the CPU.

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

The resolution scaling data paints a clear picture of a GPU-bound system. At Low settings, the average FPS moves from 59 at 1920x1080 to 45 at 2560x1440, a 23.7% drop, and then to 30 at 3840x2160, another 33.3% drop from 1440p. The total reduction from 1080p to 4K at Low settings is 49.2%. This is a classic pattern for a GPU that is running out of fillrate and memory bandwidth as pixel count increases. The CPU’s performance is constant across resolutions, so if the processor were the bottleneck, the FPS would remain flat. Instead, the FPS plummets, confirming the Arc B370 is the sole constraint.

At Medium settings, the scaling is similar but with lower absolute values. The average FPS is 42 at 1080p, 32 at 1440p, and 21 at 4K. The percentage drops are 23.8% from 1080p to 1440p, and 34.4% from 1440p to 4K. The min FPS values also tell a story: at 1080p Medium, the min is 36 FPS, which drops to 27 at 1440p and 18 at 4K. These minimums show that the GPU is not just averaging low numbers; it is also experiencing significant frame time spikes at higher resolutions, likely due to memory bandwidth saturation with system-shared memory.

The High settings curve shows an even steeper decline. From 37 FPS at 1080p to 28 FPS at 1440p (24.3% drop) to 19 FPS at 4K (32.1% drop from 1440p). The Ultra settings are the most punishing, with 32 FPS at 1080p, 25 FPS at 1440p, and 17 FPS at 4K. The difference between Low and Ultra at 1080p is 27 FPS (59 vs 32), which is a 45.8% reduction. This indicates that the GPU’s shading units and TMUs are overwhelmed by the higher quality presets, as expected from a part with only 6.144 TFLOPS of FP32 performance.

The data strongly suggests that the CPU is never the limiting factor at any resolution. If the Ryzen 5 5500X3D were the bottleneck, we would expect to see FPS plateau at 1080p and remain similar at higher resolutions. Instead, the FPS scales inversely with pixel count, which is the signature of a GPU-limited scenario. The Arc B370’s system-shared memory bandwidth, listed as "System Dependent," is likely the primary culprit, as the GPU must compete with the CPU for memory access, further reducing effective throughput.

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

At 1920x1080, the combo delivers its best results. Low settings produce an average of 59 FPS, which is the only playable frame rate in the entire dataset. Medium settings drop to 42 FPS average, with a minimum of 36 FPS and a maximum of 48 FPS. High settings yield 37 FPS average, and Ultra settings fall to 32 FPS average. The spread from Low to Ultra is 27 FPS, showing that the GPU has very little headroom to handle increased visual fidelity. The min FPS at Medium is 36, which is still above the 30 FPS threshold for basic playability, but the lack of min FPS data for Low and High settings means we cannot fully assess frame time consistency at those presets.

At 2560x1440, the performance degrades noticeably. Low settings average 45 FPS, which is a 14 FPS drop from 1080p. Medium settings average 32 FPS, with a min of 27 FPS and a max of 37 FPS. High settings average 28 FPS, and Ultra settings average 25 FPS. The min FPS at Medium is 27, which falls below the 30 FPS playability line, indicating that the GPU is struggling to maintain smooth frame delivery. The gap between Low and Ultra narrows to 20 FPS, suggesting that the GPU is becoming pixel-bound rather than compute-bound at this resolution.

At 3840x2160, the results are consistently poor. Low settings average 30 FPS, which is the only preset to reach the 30 FPS baseline. Medium settings average 21 FPS, with a min of 18 FPS and a max of 25 FPS. High settings average 19 FPS, and Ultra settings average 17 FPS. The min FPS of 18 at Medium is barely above what would be considered a slideshow, and the max FPS of 25 is still below the 30 FPS target. The difference between Low and Ultra at 4K is only 13 FPS, showing that the GPU is completely overwhelmed by the pixel count and cannot differentiate between quality presets effectively.

The data across all resolutions shows a consistent pattern: the Arc B370 is only capable of delivering playable frame rates at 1080p Low, and even then, it barely hits 60 FPS. The 59 FPS average at 1080p Low is the only measurement that approaches a smooth experience. Every other setting and resolution combination falls below 50 FPS, with most falling below 40 FPS. The min FPS values, where available, are always several frames below the average, indicating that frame pacing is inconsistent.

How This Combo Ranks — use comboRankInGame vs other tested combos

The combo ranks 2657 out of 2888 tested combinations in Control, placing it in the bottom 8% of all systems tested. This is a remarkably low ranking, and it is directly attributable to the Intel Arc B370’s performance class. The GPU’s 5th percentile ranking among all GPUs means that 95% of all GPUs tested are faster than this one. The CPU, by contrast, ranks in the 85th percentile among all CPUs, which means it is faster than 85% of all processors. The combination of a high-tier CPU with a bottom-tier GPU creates a severely imbalanced system.

The nearest rival GPUs to the Arc B370, based on the 3DMark Steel Nomad benchmark, are the ATI Mobility Radeon HD 5570 (1186 score, -0.2% delta), the ATI Radeon HD 5770 (1190 score, -0.5% delta), and the AMD Radeon HD 7650M (1192 score, -0.7% delta). These are all parts from the early 2010s, and the fact that the Arc B370 is competitive with them explains why the combo ranks so poorly. The CPU’s nearest rivals, on the other hand, are modern parts like the AMD Ryzen 5 5600F and the Intel Core Ultra 5 225, showing that the processor is in a completely different performance class than the GPU.

The rank of 2657 out of 2888 means that there are only 231 combos that are slower than this one. Given the GPU’s 5th percentile ranking, this is not surprising. The combo’s ranking is essentially dictated by the GPU’s position in the overall distribution. If the same CPU were paired with a mid-range GPU, the combo would likely rank much higher, as the CPU has the headroom to support significantly more graphical performance. The data suggests that the CPU is not the limiting factor in this ranking; it is the GPU that drags the entire system down.

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

Based on the measured FPS data, the only viable settings preset for a playable experience is 1920x1080 with Low settings, which achieves an average of 59 FPS. This is the only configuration that approaches the 60 FPS target, and it is the only one that can be considered smooth. At this setting, the game is playable, though the visual quality will be significantly reduced. The min FPS for this preset is not listed, but the average of 59 suggests that it is likely above 50 FPS for most of the time.

The alternative is 1920x1080 with Medium settings, which averages 42 FPS with a min of 36 FPS. This is a more visually appealing option, but the average frame rate is below the 60 FPS target, and the min FPS of 36 means there will be noticeable stutters during heavy action sequences. For players who prioritize visual quality over frame rate, this is the best compromise, as it is the highest settings preset that maintains an average above 40 FPS.

At 2560x1440, the best option is Low settings at 45 FPS average, but this is a 14 FPS drop from 1080p Low, and the visual fidelity is already low. The Medium preset at 1440p averages 32 FPS with a min of 27 FPS, which falls below the 30 FPS playability threshold. At 4K, no preset is playable, as even Low settings only average 30 FPS, and the min FPS at Medium is 18. The data clearly indicates that this combo should be used at 1080p, and the settings should be set to Low for the best experience. Ultra settings are not recommended at any resolution, as they provide no meaningful visual benefit given the GPU’s inability to render them smoothly.

The recommendation is to use 1920x1080 with Low settings as the primary configuration, and to consider 1920x1080 with Medium settings only if the player is willing to accept sub-60 FPS frame rates for improved visuals. Higher resolutions should be avoided entirely, as the frame rates drop below playable levels. The CPU has ample headroom, so players should not worry about CPU-related stutters; the GPU is the sole constraint, and the settings should be chosen accordingly.