Fortnite

The AMD Ryzen 7 3700X paired with the Intel Arc B570 in Fortnite presents a compelling case study in system balance, where the data reveals a configuration that shifts from GPU-limited at 4K to increasingly CPU-bound as resolution drops. This analysis examines the measured frame rates across three resolutions and four quality presets, interpreting what the numbers imply about the limiting component, the combo’s standing among 2,740 tested configurations, and the specific roles played by the CPU’s eight Zen 2 cores and the GPU’s 10 GB of GDDR6 memory.

Resolution Scaling

The measured FPS data shows a dramatic scaling pattern as resolution increases, and the shape of that curve tells a clear story about where this system’s bottleneck lies. At 1080p with Low settings, the combo delivers 176 FPS average, which drops to 112 FPS at 1440p and then to 59 FPS at 4K. That represents a 36% reduction from 1080p to 1440p, and a further 47% reduction from 1440p to 4K. The scaling is not linear—it steepens at higher resolutions, which is a classic sign that the GPU is becoming the dominant constraint at 4K.

Looking at High settings, the pattern repeats but with lower absolute numbers: 96 FPS at 1080p, 61 FPS at 1440p, and 32 FPS at 4K. The percentage drops are 36% from 1080p to 1440p and 48% from 1440p to 4K, nearly identical to the Low preset’s scaling. This consistency across presets suggests that the resolution scaling behavior is driven by the GPU’s raw throughput rather than by any particular setting. The Intel Arc B570, with its 11.52 TFLOPS FP32 performance and 380.0 GB/s memory bandwidth, appears to be the limiting factor at higher pixel counts.

At 4K Ultra, the average FPS falls to 21, which is a 67% drop from 4K Low’s 59 FPS. That is a massive penalty for enabling Ultra settings, and it indicates that the GPU is severely overburdened at this combination. The 4K Medium row shows 42 FPS average with a minimum of 35 and maximum of 48, providing a narrow range that suggests consistent GPU-bound behavior rather than CPU-induced frame pacing issues.

The question the data raises is whether the Ryzen 7 3700X ever becomes the bottleneck. At 1080p Low, 176 FPS is a high frame rate, but the CPU’s 8 cores and 16 threads with a boost clock of 4.40 GHz should be capable of sustaining that in a title like Fortnite. However, the scaling from 1080p to 1440p is nearly identical across presets, which implies the GPU is still the primary limiter even at 1080p High. The absence of minimum FPS data for most rows makes it difficult to assess frame time consistency, but the Medium preset’s min/max values at 1080p (106 to 143) show a spread of 37 FPS, which is proportionally similar to the spread at 1440p (67 to 91, a 24 FPS gap) and 4K (35 to 48, a 13 FPS gap). This narrowing spread at higher resolutions further supports the GPU-bound interpretation.

How This Combo Ranks

The combo ranks 2,178th out of 2,740 tested combinations in Fortnite, placing it in the 79.5th percentile of all configurations. That means roughly four out of five combos tested faster, which positions this pairing as decidedly mid-pack for this game. The rank is based on measured data, not synthetic estimates, so it reflects real-world performance in Fortnite specifically.

Given the CPU’s overall percentile of 84 among all CPUs and the GPU’s percentile of 65 among all GPUs, the combo rank of 2178 out of 2740 suggests that Fortnite’s performance is not purely a function of component strength. The CPU is stronger relative to its peers than the GPU is, yet the combo still lands below the 80th percentile. This implies that Fortnite’s engine may not scale perfectly with the Ryzen 7 3700X’s multithreading capabilities, or that the GPU’s mid-tier standing drags the overall result down more than the CPU can compensate.

Comparing to the CPU’s nearest rivals, the Ryzen 7 3700X scores within 0.2% of the Intel Core i5-13450HX and within 0.5% of the Intel Core Ultra 7 165H in average benchmark score, yet those CPUs would likely pair with different GPUs in practice. The GPU’s nearest rivals include the Intel Arc A750, which is just 0.1% slower on average, and the NVIDIA GeForce RTX 3070 Mobile, which is 0.1% faster. The Arc B570 sits in a tight performance cluster around the 20,500 average score mark, meaning that swapping to any of these near-identical GPUs would not meaningfully change the combo’s rank in Fortnite.

The data suggests that this combo’s ranking is more constrained by the GPU tier than by the CPU. The CPU’s 84th percentile standing would typically suggest a higher combo rank, but Fortnite’s measured results place it at the 79.5th percentile of combos, indicating that the game’s performance ceiling is set by the GPU’s capabilities at most settings and resolutions tested.

GPU Role

The Intel Arc B570’s specifications directly explain several of the measured FPS patterns. The GPU has 10 GB of GDDR6 memory on a 160-bit bus, delivering 380.0 GB/s of bandwidth. In Fortnite, which is not typically a VRAM-heavy title at 1080p, 10 GB is adequate, but at 4K Ultra the combination of higher resolution textures and more demanding effects could push memory usage higher. The fact that 4K Ultra produces only 21 FPS average suggests that memory bandwidth or capacity is not the primary issue—rather, the GPU’s raw compute of 11.52 TFLOPS FP32 is insufficient for that workload.

The GPU’s clock behavior is notable: base and boost clocks are both listed at 2500 MHz, which is unusual because most GPUs have a higher boost than base. This could indicate that the card runs at a fixed clock under load, which would make performance more predictable but also means there is no headroom to dynamically increase clocks in lighter scenes. The memory clock of 2375 MHz (19 Gbps effective) combined with the 160-bit bus yields that 380.0 GB/s figure, which is moderate for a 2025 GPU.

The Arc B570’s architecture is Xe2-HPG on a 5 nm process, with 19,600 million transistors on a 272 mm² die. The 2304 shading units and 144 texture mapping units provide the computational backbone for rasterization. In Fortnite, the measured FPS at 1080p Low (176) versus 1080p Ultra (64) shows a 64% performance drop when enabling all effects, which aligns with the GPU’s limited shader and ROP count (80 ROPs). The pixel rate of 200.0 GPixel/s and texture rate of 360.0 GTexel/s are mid-range figures, and the data reflects that—at 4K, even Low settings only achieve 59 FPS, which is below the 60 FPS threshold many players consider acceptable.

The GPU’s 65th percentile standing among all GPUs, with an average benchmark score of 20,556, puts it in the same performance class as the Intel Arc A750 and NVIDIA RTX 3070 Mobile. The passmark G3D score of 14,195 and 3DMark Steel Nomad DX12 score of 2,649 provide additional context, but the Fortnite measurements show that this GPU’s real-world gaming performance is solid at 1080p and 1440p with moderate settings, but falls short at 4K Ultra.

Measured FPS Breakdown

The measured FPS data provides a complete matrix across three resolutions and four settings. At 1920x1080, the Low preset delivers 176 FPS average, which is the highest figure recorded for this combo. Medium drops to 124 FPS average, with a minimum of 106 and maximum of 143. High further reduces to 96 FPS, and Ultra bottoms out at 64 FPS. The step from Low to Medium is a 30% reduction, from Medium to High is 23%, and from High to Ultra is 33%. This shows that each preset tier costs roughly a quarter to a third of the previous performance, with Ultra being the most punishing.

At 2560x1440, the pattern is similar but with lower absolute numbers. Low achieves 112 FPS, Medium 79 FPS (67 min, 91 max), High 61 FPS, and Ultra 40 FPS. The percentage reductions from Low to Medium (29%), Medium to High (23%), and High to Ultra (34%) are remarkably consistent with the 1080p pattern, suggesting that the performance cost of each preset tier is resolution-independent. This consistency implies that the GPU’s workload scales uniformly with both resolution and settings, and that no single setting is causing an anomalous bottleneck.

At 3840x2160, the numbers become challenging for smooth gameplay. Low yields 59 FPS, Medium 42 FPS (35 min, 48 max), High 32 FPS, and Ultra 21 FPS. Only Low settings approach playability at 4K, and even then, 59 FPS is borderline. The Medium preset’s minimum of 35 FPS would likely cause noticeable stutters in fast-paced Battle Royale scenarios. The Ultra preset at 21 FPS is effectively unplayable for competitive play, though it might be acceptable for cinematic viewing or screenshots.

The Medium preset is the only one with min/max data across all three resolutions, providing a glimpse into frame time variability. At 1080p, the range is 106 to 143 (a 35% spread relative to the average), at 1440p it is 67 to 91 (30% spread), and at 4K it is 35 to 48 (31% spread). The consistency of this spread percentage suggests that the GPU’s frame pacing is stable regardless of resolution, and that the bottleneck remains the same across the board.

Settings Recommendations

Based on the measured rows, the optimal settings depend on the target resolution and whether the user prioritizes frame rate or visual fidelity. At 1920x1080, the Low preset’s 176 FPS is ideal for competitive play where high refresh rates matter. The Medium preset at 124 FPS still provides a smooth experience for most players, with a 30% visual upgrade over Low at the cost of 30% performance. High at 96 FPS is a reasonable middle ground, but Ultra at 64 FPS may dip below the 60 FPS threshold in intense moments, making it less suitable for fast-paced gameplay.

At 2560x1440, Low’s 112 FPS is the clear recommendation for players with 120Hz or higher monitors. Medium at 79 FPS is acceptable for those who want better visuals and can tolerate slightly lower frame rates. High at 61 FPS is marginal—it stays above 60 FPS on average, but the lack of min/max data means there is a risk of drops below that threshold. Ultra at 40 FPS is not recommended for anything other than casual play or screenshots.

At 3840x2160, the data strongly suggests that only Low settings provide a playable experience, with 59 FPS average. Medium at 42 FPS is below the 60 FPS target and would feel sluggish in a Battle Royale. High and Ultra at 32 and 21 FPS respectively are not viable for actual gameplay. For 4K users, the recommendation would be to either stick with Low settings or consider lowering the resolution to 1440p for better frame rates.

The overall best experience per the measured rows appears to be 1080p Medium or 1440p Low, depending on monitor resolution. The 1080p Medium row shows 124 FPS average with a minimum of 106, which ensures that even in the most demanding scenes, the frame rate stays above 100 FPS. The 1440p Low row at 112 FPS is also strong, and the lack of min/max data does not necessarily indicate instability—it simply means those values were not recorded.

FAQ

Q: Can this combo run Fortnite at 4K with High settings?

A: The measured data shows 32 FPS average at 4K High, which is well below the 60 FPS target for smooth gameplay. This setting is not recommended for competitive play, though it could be used for less demanding scenarios.

Q: What is the highest FPS this combo achieves in Fortnite?

A: The highest recorded average FPS is 176 at 1920x1080 with Low settings. The next highest is 124 FPS at 1080p Medium, followed by 112 FPS at 1440p Low.

Q: Is the Ryzen 7 3700X or the Intel Arc B570 the limiting factor?

A: The resolution scaling data indicates the GPU is the primary limiter. The FPS drops consistently as resolution increases, and the percentage drop from 1080p to 1440p is nearly identical across all presets, which is characteristic of GPU-bound performance.

Q: How does this combo rank compared to other tested configurations?

A: This combo ranks 2,178th out of 2,740 tested combos in Fortnite, placing it in the 79.5th percentile. This is lower than the CPU’s 84th percentile standing, suggesting the GPU holds the combo back.

Q: What settings provide the smoothest experience at 1440p?

A: Low settings at 1440p deliver 112 FPS average, which is the best recorded at that resolution. Medium at 79 FPS is also viable, but the 33 FPS drop may be noticeable for players accustomed to high frame rates.

Q: Does the GPU’s 10 GB VRAM cause any issues in Fortnite?

A: The measured data shows no direct evidence of VRAM-related stuttering or crashes. However, the 4K Ultra preset at 21 FPS suggests that the GPU’s compute power, not its 10 GB capacity, is the limiting factor at that setting.

CPU Role

The AMD Ryzen 7 3700X brings 8 cores and 16 threads based on the Zen 2 architecture, manufactured on a 7 nm process by TSMC. The base clock of 3.60 GHz and boost clock of 4.40 GHz provide a solid foundation for gaming, and the 32 MB of L3 cache helps with game data locality. The CPU’s 65 W TDP is modest, and its socket is AMD Socket AM4, meaning it can be paired with a wide range of motherboards.

In Fortnite, the CPU’s role appears secondary to the GPU at most tested settings. At 1080p Low, the 176 FPS average suggests the CPU can feed the GPU with enough frames, but the identical scaling pattern across resolutions implies that even at 1080p, the GPU is not being fully saturated. The CPU’s passmark multithread score of 22,458 and single-thread score of 2,656 indicate balanced performance, but Fortnite is known to favor single-core performance, and the Ryzen 7 3700X’s single-thread score places it in the upper mid-range.

The CPU’s nearest rivals in average benchmark score include the AMD Ryzen AI 7 350 (0.1% faster) and Intel Core i5-13450HX (0.2% faster), which suggests that the 3700X is still competitive in synthetic tests. However, in Fortnite, the combo’s rank of 2,178 out of 2,740 indicates that the CPU’s strengths are not fully realized. The L3 cache of 32 MB is shared across all cores, which helps in multi-threaded workloads but may not provide a significant advantage in a game that typically uses fewer threads.

The CPU’s memory bandwidth of 51.2 GB/s from dual-channel DDR4 is adequate for Fortnite, and the PCIe Gen 4 support with 24 lanes ensures the GPU has sufficient bandwidth. The data suggests that upgrading the CPU alone would not dramatically improve Fortnite performance, since the GPU is the bottleneck at most resolutions. Conversely, upgrading the GPU while keeping this CPU would likely yield larger FPS gains, particularly at 1440p and 4K.

The measured FPS at 1080p Medium (124 FPS average) with a minimum of 106 suggests that the CPU can sustain higher frame rates without significant drops, but the lack of min/max data for other presets makes it impossible to determine if any CPU-induced stutters occur at 176 FPS Low. The CPU’s 84th percentile ranking among all CPUs indicates it is a capable processor, but Fortnite’s engine may not leverage all 16 threads effectively, leaving the 8-core, 16-thread configuration partially underutilized in this specific title.