To The Core

The AMD Ryzen 5 4600G — a 6-core, 12-thread Zen 2 desktop chip on Socket AM4 — paired with the Intel Arc B570 (Battlemage, Xe2-HPG) produces a measured combination that is fully playable in “To The Core” across the tested resolution and settings matrix. The launch MSRP for the CPU is $154 and for the GPU is 219 USD, though this analysis focuses purely on the measured frame-rate data rather than market positioning. With 12 configured FPS data points spanning 1080p, 1440p, and 4K at four quality levels, the pattern that emerges is consistent scaling: the GPU becomes the dominant bottleneck once pixel count rises, while the CPU provides ample headroom for this idle-style title. The data is directly measured, and every figure below comes from the measured FPS table plus the documented hardware specifications.

Resolution Scaling

Starting at 1920x1080, the measured average frame rates are 225 FPS on Low, 191 on Medium, 165 on High, and 133 on Ultra. Going from Low to Ultra at 1080p costs 92 FPS, which is a 41% reduction in output. At this setting the game is trivially smooth, with every preset — even Ultra — clearing the 120 FPS mark by a comfortable margin. The Medium preset records a minimum of 162 FPS, meaning even the worst delivered frames in that row stay above the refresh rate of a basic 144 Hz monitor. While Low is the only preset before the vertical axis, the High setting’s 165 and Ultra’s 133 are both useful metrics to understand how the renderer’s pixel fill and shading pressure scale.

Moving to 2560x1440, the same pattern holds with a modest increase in load. Low averages 152, Medium 128 (recorded min 109, max 147), High 111, and Ultra 89. The drop from 1080p High to 1440p High is 165 to 111, a 33% reduction in average FPS. From 1440p High to 4K High, the average falls to 64, which is a further 42% reduction. That steeper second jump highlights the 4K category’s increased pressure: the resolution scale from 1440p to 3840x2160 multiplies the pixel count by 2.25, and the measured scaling reflects a heavier than proportional cost.

At 3840x2160, the measured averages are 87 on Low, 73 on Medium (min 62, max 84), 64 on High, and 51 on Ultra. The scaling ratio from 1080p Low to 4K Low is 225 to 87, which is a 61% drop. High follows the same shape: 165 to 64, also a 61% drop. Ultra 133 is 51 at 4K, again a 61% drop. That uniformity — roughly 61% across all three presets — indicates that the resolution scaling is proportionate to pixel count and not blocked by CPU threading. The only fully recorded frames are on Medium, where 4K the minimum is 62 FPS. This provides the sole proof in this data set that a 60 FPS floor is achievable at 4K on this hardware.

Looking at the whole picture, the pattern is clear: across 1080p, 1440p, and 4K, the FPS values at each setting level hold a similar order — Low > Medium > High > Ultra. The relative gap between presets doesn’t widen as resolution rises, except that the final step from 1440p to 4K is heavier than the step from 1080p to 1440p. The near-constant 61% drop implies that the rendering pipeline spends most of its time on fragment processing — something the Arc B570 does slower as pixel area quadruples from 1080p. A built-in budget exists at 4K for 60 FPS, but it will force the choice of Medium settings and a careful choice of 32%/resolution.

GPU Role

The Intel Arc B570 is built on TSMC 5 nm with the Xe2-HPG architecture, using 22 shading units (2304 ®) and 18 raytracing cores. Its 10 GB GDDR6 memory runs on a 160-bit bus at 2375 MHz with 380.0 GB/s of bandwidth, which sits at 2500 MHz base and 2500 MHz boost. The GPU is rated at 150 W TDP and connects via PCIe 4.0 x8. Recorded in the database, its PassMark G3D score is 14195, and its average benchmark spread sits at 20556, placing it in the 65th percentile of all tested GPUs.

In “To The Core”, the GPU’s measured role is decisive. At 1080p Low, the 225 FPS result is near the upper loop limit of a 144 Hz monitor, and it is not the highest in the queue. On Ultra, the 133 FPS — a 41% cut — is evidence that this title votes for pixel and shielding throughput rather than raw VRAM transfer. With 10 GB of GDDR6, VRAM capacity is not the limiting factor at 4K Ultra, because the measured FPS 51 is still relatively high, meaning memory usage didn’t fall off a wall, but the GPU’s compute did.

The video card also shows its strength in its two nearby rivals: the NVIDIA GeForce RTX 3070 Mobile (deltaPct 0.1), the Intel Arc A750 (deltaPct -0.1), the NVIDIA Quadro M4000M (deltaPct 0.4), and the AMD Radeon R9 M390X (deltaPct -0.5). None of these are more than half a percentage below or above the B570’s average score, placing the Arc B570 in a tight cluster. In a normal gaming scene, those rivals would presumably render Frappy details differently; but in this specific title, the measured row of data does not draw any special distinction between those tenth-of-a-percent differences. What matters more is that the 65th percentile GPU supports the 4K Medium with a 62 FPS minimum — a clear sign that it sits low-profile.

The only proviso is memory clock at 1176 MHz effective, which yields 380.0 GB/s. That range is enough, because the game’s average FPS at 4K Medium 73, but at 4K Ultra it dips to 51 — once the texture and shading effects load more, the bandwidth allocation seems adequate but the shading units saturate. For an idle game, the GPU of this class is performant; it faces square pixels, not the physics or compute of a real-time 3D title, but the shader unit count (2304) and texture fill rate (360.0 GTexel/s) carry this workload consistently up to High.

Settings Recommendations

For a player looking at the data, the setting that lives best depends on the target refresh rate. At 1080p, all four presets are well above 100 FPS. The Medium setting is the most balanced: average 191 with a minimum of 162, which is perfect for a 144Hz display, and it keeps a comfortable 63% headroom above the 102 FPS baseline of the next-lowest preset. If the goal is pure frame rate, Low gives 225, but that is far beyond what any standard 60-1920×1080 monitor needs. Unfortunately, no min FPS is provided for High or Ultra on that resolution — no source data for — so Medium is the only preset where the min was recorded, which makes it the most confident choice.

At 1440p, the data is equally clear: Low at 144, Medium at 186 (min 109), High at 111, Ultra at 89. Medium stands out: only it has a reported min, 109, which is 72% of its average. To sustain a 144 Hz output, only Low can reach that target; as a result, the 1440p High preset at 111 FPS is more suitable for a 100-144 Hz display, and Ultra’s 89 offers excellent visual quality at 60 Hz. If the monitor is 4K, the spectrum is tighter: Medium (73 avg, min 62) is the the most stable 60-Hz-compatible option, while High at 64 avg has no min data and will come near but maybe not hold - while Ultra is clearly 51 — sub-60, so it’s not recommended for that refresh.

The service recommendations: Use 1080p Ultra if you own a 1080p screen and don’t care about absurd refresh; it sits at 133 with no min FPS observed. For a 1440p monitor at high quality, Medium (min 109) or High (111) both keep above 100 on average; propagate the fact that High is safer for visual fidelity. On 4K, choose Medium. It has the lowest avg 73 of all 4K presets, but the recorded min of 62 proves the frame time stays at 55-60, while High and Ultra are either marginally above or measurably below. 4K Ultra is the hardest and only true test case; it does deliver 51 avg but misses a stable 60.

How This Combo Ranks

The AMD Ryzen 5 4600G and Intel Arc B570 combination is ranked 2992 out of 3725 tested combos for “To The Core”. On the surface, that places a 874th-percentile result in the low 80% of all tested pairs — not a headline job. However, the reason for that rank is the sheer size of the tested combos, and the pointer that this particular game is an idle game where the hardware buckets are dense at the top. The game’s measured frame rates are restricted by the monitor’s vertical refresh; at 1080p Ultra with 133, most of the top combo list would deliver, likely, more units, but the rank isn’t a verdict on how good the system is.

Look at the CPU in this context. It sits at the 71st percentile of all CPUs, with an average score of 17507. The nearest neighbors (AMD Ryzen 3 PRO 5355GE, 0.1% gap; AMD Ryzen 3 8300GE, 0.6% gap; Intel Core 7 150U, 0.6% gap; Intel Core i3-14100T, 0.8% gap) all have near-identical averages. This is a CPU that trades blows at the geometric mean of its peers, yet the combo’s rank is below the combine average. So the rank itself is not an assessment of the CPU; it’s a statement that this CPU does not provide unique performance.

On the GPU side, the Arc B570 (65th percentile) is also docking in a group where the previous-mentioned RTX 3070 Mobile, Arc A750, Quadro M4000M, and R9 M390X are all within one tenth of a percent. The short lives of the theoretical difference — a 0.1% delta isn’t observable. Yet in practice, the game’s bottleneck is the GPU as resolution climbs, and this is the clearest statement of why the combo’s 2992 rank is lower than the top scores: the actual FPS at 4