Control

Intel Core Ultra 7 265 paired with the Intel Arc B370 delivers a playable but constrained experience in Control, with the combo ranking in the bottom 8% of tested configurations. The data shows a clear pattern: this system is only viable at lower resolutions and settings, with 1080p Low being the only preset that approaches a smooth 60 FPS average. At higher resolutions, the GPU becomes the dominant bottleneck, while the 20-core CPU provides ample headroom that goes largely unused.

FAQ

Q: What average frame rate does this combo achieve at 1920x1080 on Low settings?

A: The measured average FPS at 1080p Low is 59, which is the highest average across all tested settings and resolutions. This is the only configuration in the dataset that comes close to a 60 FPS target.

Q: How does the system perform at 3840x2160 resolution?

A: At 4K, the average frame rates range from 17 FPS on Ultra to 30 FPS on Low. Even the best 4K result, 30 FPS on Low, falls below the threshold for smooth gameplay, and the High preset drops to 19 FPS average.

Q: What is the difference between Medium and High settings at 2560x1440?

A: At 1440p, Medium settings produce an average of 32 FPS with a minimum of 27 and maximum of 37, while High settings yield an average of 28 FPS. The gap between these presets is 4 FPS, indicating that the GPU is already heavily loaded at this resolution.

Q: Does the Intel Arc B370 have dedicated VRAM?

A: No. The GPU uses system shared memory for both its memory size, type, and bus width. The memory bandwidth is listed as system dependent, meaning performance scales with the host system's memory configuration.

Q: How does the CPU's benchmark score compare to its nearest rivals?

A: The Intel Core Ultra 7 265 has an average benchmark score of 64640, which is 0.3% higher than the Intel Core Ultra 7 265F (64438) and 0.7% higher than the AMD EPYC 7343 (64202). It trails the AMD EPYC 4464P by 0.3% (64823) and the AMD EPYC 9124 by 0.7% (65104).

Q: What is the combo's overall rank among all tested systems in Control?

A: The combo ranks 2657 out of 2888 tested combinations, placing it in the bottom portion of the database for this game. This rank reflects the GPU's limited performance relative to other configurations.

Settings Recommendations

The measured data strongly indicates that this hardware combination should be run at 1920x1080 with Low settings for the best playable experience. The 1080p Low preset delivers an average of 59 FPS, which is nearly double the 32 FPS average seen at 1080p Ultra. The jump from Low to Medium at 1080p costs 17 FPS on average (from 59 to 42), and the further step to High costs another 5 FPS (from 42 to 37). Given that the GPU's percentile ranking is 5 out of 100, prioritizing frame rate over visual fidelity is the correct approach.

At 2560x1440, even the Low preset only reaches 45 FPS average, which is playable but not smooth by modern standards. The Medium preset at 1440p drops to 32 FPS average, and High falls to 28 FPS. Based on the data, 1440p should only be considered if the user accepts sub-50 FPS averages, and even then, Low is the only viable setting. The 3840x2160 results are consistently below 30 FPS across all presets, making 4K effectively unplayable for this game with this combo, regardless of settings adjustments.

The recommendation hierarchy from the data is clear: 1080p Low is the primary target, with 1080p Medium as a fallback if visual quality is prioritized over frame rate. The 1440p Low preset can serve as a compromise for users with higher-resolution displays, but the 45 FPS average indicates noticeable frame pacing issues will occur. The data shows no scenario where Ultra settings are advisable, as the 1080p Ultra average of 32 FPS is identical to the 1440p Medium result, and the visual gains do not compensate for the halving of frame rate compared to Low.

GPU Role

The Intel Arc B370 is the limiting factor in this configuration. Its 3DMark Steel Nomad DX12 score of 1184 places it at the 5th percentile of all GPUs, and its nearest rivals are all older or lower-tier parts: the ATI Mobility Radeon HD 5570 scores 1186 (0.2% higher), the ATI Radeon HD 5770 scores 1190 (0.5% higher), and the AMD Radeon HD 7650M scores 1192 (0.7% higher). The only GPU in the nearest rival list that the Arc B370 beats is the AMD FirePro M2000, which scores 1168 (1.4% lower). This positioning explains the measured FPS results across all resolutions.

The GPU's memory configuration is system shared, meaning there is no dedicated VRAM. The memory clock is listed as system shared, and the bandwidth is system dependent. This architecture relies on the host system's DDR5 memory, which for the Intel Core Ultra 7 265 is dual-channel with a bandwidth of 102.4 GB/s. The lack of dedicated VRAM likely contributes to the steep performance drops when moving from 1080p to 1440p and from 1440p to 4K, as the shared memory bandwidth becomes a bottleneck.

The shading unit count is 1280, with 40 texture mapping units and 20 render output units. The GPU has 10 ray tracing cores, which is notable for a part in this performance class, but the measured FPS data does not indicate that ray tracing is enabled or that it contributes to playable performance. The pixel rate is 48.00 GPixel/s and the texture rate is 96.00 GTexel/s, with a FP32 performance of 6.144 TFLOPS. These specifications align with the observed performance: the card can handle 1080p Low at 59 FPS but struggles as resolution and settings increase.

The GPU's boost clock is 2400 MHz with a base clock of 300 MHz, and the TDP is 25 W. The slot width is listed as IGP, indicating this is an integrated graphics solution rather than a discrete card. This form factor explains the lack of power connectors and the system shared memory. The 3DMark score of 1184 is the only GPU benchmark in the dataset, and it correlates well with the Control results: the card is positioned at the very bottom of the performance spectrum, and the FPS data reflects that.

Measured FPS Breakdown

At 1920x1080, the data shows a clear progression across settings. Low settings produce an average of 59 FPS, which is the only measured result above 50 FPS in the entire dataset. Medium settings drop to 42 FPS average, with a minimum of 36 and maximum of 48. High settings yield 37 FPS average, and Ultra settings produce 32 FPS average. The spread from Low to Ultra is 27 FPS, which is the largest settings-based delta at any resolution, indicating that the GPU is not fully utilized even at 1080p Low but becomes increasingly saturated as settings rise.

At 2560x1440, the results scale down proportionally. Low settings give 45 FPS average, Medium drops to 32 FPS average (minimum 27, maximum 37), High yields 28 FPS average, and Ultra produces 25 FPS average. The delta from Low to Ultra is 20 FPS, slightly less than at 1080p, suggesting that the GPU is hitting a hard performance ceiling around 45 FPS at this resolution. The Medium preset at 1440p (32 FPS) matches the Ultra preset at 1080p (32 FPS), which suggests that resolution scaling is roughly linear for this GPU in Control.

At 3840x2160, the performance collapses. Low settings give 30 FPS average, Medium drops to 21 FPS average (minimum 18, maximum 25), High yields 19 FPS average, and Ultra produces 17 FPS average. The delta from Low to Ultra is only 13 FPS, which is the smallest of any resolution. This compression indicates that the GPU is severely bottlenecked at 4K, and even the lowest settings cannot overcome the pixel throughput limitation. The 4K Low result of 30 FPS is exactly half the 1080p Low result of 59 FPS, which is a direct reflection of the fourfold increase in pixel count.

The Medium presets across all three resolutions provide the most complete data, with minimum and maximum FPS values recorded. At 1080p Medium, the range is 36 to 48 FPS. At 1440p Medium, the range is 27 to 37 FPS. At 4K Medium, the range is 18 to 25 FPS. These ranges show that frame pacing is relatively stable, with the spread between minimum and maximum being 12 FPS at 1080p, 10 FPS at 1440p, and 7 FPS at 4K. The narrowing range at higher resolutions suggests that the GPU becomes more consistent as it operates near its performance limit, albeit at unplayable frame rates.

How This Combo Ranks

The combination of the Intel Core Ultra 7 265 and Intel Arc B370 ranks 2657 out of 2888 tested combos in Control. This places the combo in the bottom 8% of all configurations, which aligns with the GPU's 5th percentile ranking. The CPU's 93rd percentile ranking (relative to all CPUs) does not compensate for the GPU's limitations, as the game's frame rate is clearly constrained by the graphics subsystem across all measured settings.

The GPU's nearest rivals in the benchmark database all score within 1.4% of the Arc B370's 1184 3DMark Steel Nomad score. This tight clustering means that any of these GPUs would likely produce similar results in Control, and the combo rank would not change significantly with a swap to a comparable part. The CPU's nearest rivals, which include the AMD EPYC 4464P (64823 average score, 0.3% higher), the Intel Core Ultra 7 265F (64438, 0.3% lower), the AMD EPYC 7343 (64202, 0.7% lower), and the AMD EPYC 9124 (65104, 0.7% higher), are all server or workstation parts that would not meaningfully alter the gaming performance in this scenario.

The rank of 2657 out of 2888 means that over 90% of tested combos achieve higher frame rates in Control. This is consistent with the measured FPS data, which shows only one preset (1080p Low at 59 FPS) reaching a playable threshold. The gap between this combo and higher-ranked systems is not a matter of tuning or settings optimization; it is a fundamental hardware limitation driven by the GPU's position in the performance hierarchy. The CPU's 20 cores and 5.30 GHz boost clock cannot overcome the fact that the Arc B370 produces 6.144 TFLOPS of FP32 performance, which is insufficient for modern action-adventure titles at higher quality presets.

CPU Role

The Intel Core Ultra 7 265 provides far more processing capability than this GPU configuration can utilize in Control. The CPU has 20 cores and 20 threads, with a base clock of 2.40 GHz and a boost clock of 5.30 GHz. Its 30 MB of shared L3 cache and 3 MB L2 per core support high multi-threaded throughput, as evidenced by a Cinebench R23 multi-core score of 42216 and a single-core score of 5960. The PassMark multi-thread score of 49682 and single-thread score of 4689 further confirm that the CPU is not a bottleneck in this game.

The CPU's benchmark scores place it at the 93rd percentile of all CPUs, and its average benchmark score of 64640 is within 0.7% of its nearest rivals. The Cinebench R20 scores of 6268 multi-core and 884 single-core, along with the Cinebench R15 scores of 4255 multi-core and 600 single-core, all indicate strong computational performance. The PassMark data compression score of 522983, data encryption score of 40456, and extended instructions score of 41478 show that the CPU excels at varied workloads, but these capabilities have minimal impact on Control's frame rate.

The 3 nm process node and Arrow Lake-S architecture contribute to a 65 W TDP, which is modest for a 20-core processor. The memory support for DDR5 with a dual-channel bus and 102.4 GB/s bandwidth matches the GPU's system shared memory requirement, but this bandwidth is shared between CPU and GPU operations. The integrated graphics solution means that the Arc B370 uses the same memory pool as the CPU, which can create contention under load, though the measured FPS data does not show abnormal frame drops that would indicate severe memory conflicts.

In Control, the CPU's role is to feed the GPU with draw calls and physics data. The 20 cores and 20 threads are more than sufficient for this task, as even the single-thread performance (R23 single-core of 5960) exceeds what most games require. The data shows that the frame rate scales almost entirely with GPU settings and resolution, not with CPU workload. The 59 FPS at 1080p Low versus 17 FPS at 4K Ultra is a GPU-driven difference, and the CPU's performance metrics remain constant across these scenarios. If a more powerful GPU were paired with this CPU, the combo rank would likely improve substantially, but that is outside the scope of the measured data.