Pc Building Simulator

Resolution Scaling

The measured frame rates for Pc Building Simulator with the AMD Ryzen 5 5500X3D and Intel Arc B390 show a clear, predictable pattern: average FPS collapses as resolution increases, revealing that the graphics hardware is the overwhelming constraint in this configuration. At High settings, the average FPS moves from 62 at 1920x1080 to 39 at 2560x1440 and then falls sharply to 21 at 3840x2160. This progression from 62 to 21 across the three common resolutions is not a subtle degradation; it is a drastic loss of playability at the top end. The same pattern repeats at Low settings, where 1080p delivers 125 average FPS, 1440p drops to 79, and 4K lands at 41. The scaling ratio between the 1080p and 4K results is roughly three to one, which points strongly at a fill-rate or shader-limited processor rather than a CPU-limited scenario.

The game engine, being a Simulation, typically has modest texture and lighting processing, but the resolution-wise collapse in FPS means the data is consistent with a GPU that has a very low pixel rate. With 24 ROPs, 48 TMUs, and a boost clock of 2500 MHz the graphics core is delivering 60 GPixel/s, yet even this theoretical rate is far below what would be needed to maintain fluid frames up to 4K. Since the CPU in this system is a 6-core, 12-thread Zen 3 part with a high single-thread score of 2941, it would not be expected to throttle at any measured resolution – the FPS trends to do not behave as if the CPU becomes the governor at lower settings. The data shows that doubling the spatial area (from 1080p to 1440p) incurs a roughly 37% reduction in average FPS, and moving to 4K (four times the 1080p area) causes a 66% drop – a scaling curve consistent with a device that is raster-bound.

Interestingly, the scaling pattern holds even when settings are lowered. At 4K, Low produces 41 FPS and Ultra produces 20 FPS, but the drop from Ultra to Low is only about two-fold. At 1440p, Low achieves 79 FPS while Ultra achieves 38, so the game turns out a strong relationship between pixel count and performance. A modern integrated graphics solution – the Arc B390 is an IGP with system shared memory – is forced to move data across a shared memory bus, which amplifies the resolution scaling. In summary, the measured data categorically describe a scenario where the FPS depends almost linearly on the number of pixels processed, isolating the efficiently weak GPU as the limiting factor across settings.

GPU Role

The Intel Arc B390’s specifications clarify why the resolution scaling is severe. Its base clock is 300 MHz, trimming to a boost of 2500 MHz, but the card is fundamentally a mobile-class integrated part in the Panther Lake generation using Xe3-LPG. Its 80 W TDP and lack of discrete power connectors indicate it belongs to the low-power segment. The GPU sits at a 9th percentile benchmark rank, with a 3DMark Steel Nomad DX12 score of 1482. That places it only 1.3% ahead of the NVIDIA GeForce GT 520MX and 2.7% above the GeForce GT 710, establishing its niche as a chip intended for entry-level productivity or casual business tasks, not gaming at high fidelity.

The memory subsystem is explicitly "System Shared" – both in size and bus width. There is no dedicated VRAM, working as a shared memory configuration that steals system bandwidth. This compromises the memory bandwidth, listed as "System Dependent," creating a severe constraint for any game that requires frequent texture reads, even a simulation. In this test, the GPU’s boost clock of 2500 MHz is a reliable number but cannot offset the lack of dedicated memory bandwidth. The 7.680 TFLOPS FP32 compute and 120.0 GTexel/s texture rate are the raw output numbers, but the 24 ROPs limit pixel throughput to 60.00 GPixel/s, which identifies exactly why 4K frames hover at around 21-29 FPS except for the minimum of 41 FPS at Low.

The data from the combination shows that raising the settings multiplier directly inflates the cost. For 1080p, the Ultra setting delivers 59 FPS, while Low delivers 125, a clear three-and-a-half-fold difference across the quality scale. The extreme divergence of the GPU’s 9th percentile versus the CPU’s 85th percentile confirms that graphical limitations are the central factor – even an efficient game like the simulation cannot overcome a capable component with 1536 shading units and 12 RT cores.

Settings Recommendations

Converting the measured data to a synthetic recommendation, the best experience for this combination comes strictly from Low settings at 1080p. At 1920x1080 with Low quality, the average FPS is 125, which is far smooth and responsive, while the Medium preset provides 88 FPS with a comfort margin for most desktops. High and Ultra at 1080p degrade to 62 and 59 FPS respectively, which are less fluid, so they present no tangible benefit over Medium in games that does not demand exhaustive shadows. The gameplay of Pc Building Simulator relies on menu tweaks and component swaps rather than twitch response, so 88 FPS at Medium is still a clear choice if the user has a 90 Hz display; for 60 Hz displays, Medium sets a stable 56 FPS on 1440p.

At 2560x1440, the options become limited: Low achieves 79 FPS, while Medium handles 56 FPS and High just 39 FPS. There is no plausible use of Ultra at this resolution if smoothness is desired. For 4K (3840x2160), the data is disappointing: High averages 21 FPS, Ultra only 20 FPS, Medium 29, and Low 41. A 4K monitor is simply not a viable target. The user must either drop to 1080p Medium (88 FPS) or sustain 2D to allow for summary; the recommendation of 1080p Medium and 1080p Low offers the only of captured numbers down to 125 FPS and 88 FPS respectively. The clear verdict: target 1080p Medium or Low to keep the simulator fully usable, because every other resolution/settings combination returns sub-60 average FPS.

Measured FPS Breakdown

The observed frame-rate data is provided for all three resolutions and four quality preset (Low, Medium, High, Ultra). Here are the critical values exactly as measured:

At 1920x1080:

• Low: 125 FPS average
• Medium: 88 FPS average (minimum 75, maximum 102)
• High: 62 FPS average
• Ultra: 59 FPS average

At 2560x1440:

• Low: 79 FPS average
• Medium: 56 FPS average (minimum 47, maximum 64)
• High: 39 FPS average
• Ultra: 38 FPS average

At 3840x2160:

• Low: 41 FPS average
• Medium: 29 FPS average (minimum 25, maximum 34)
• High: 21 FPS average
• Ultra: 20 FPS average

Notably, the minimum and maximum FPS are only recorded for Medium settings across all three resolutions. The frame spike range at 1080p Medium is 75 to 102, at 1440p 47 to 64, and at 4K 25 to 34. This partial data indicates with typical variation in a simulation that occasional small stutters can happen. Ultra is indistinguishable from High at 1080p (59 vs 62 FPS) and at 4K (20 vs 21 FPS) in terms of measurable performance, but the quality jump provides no measurable benefit at these low framerates. The overall pattern in the absolute numbers is unmistakable: the Arc B390 yields usable frames only at 1080p with Low or Medium, and the 125 FPS on Low dwarfs every other setup.

CPU Role

The AMD Ryzen 5 5500X3D is a 6-core, 12-thread CPU based on Zen 3 architecture, with a process node of 7 nm and a base clock of 3.00 GHz boosting to 4.00 GHz. The 96 MB L3 cache shared across the six cores is its distinguishing feature, along with an L2 cache of 512 KB per core. A glance at its recorded benchmarks shows a multithreading score of 20363, a single-thread score of 2941, and a physics score of 2282, while its percentile rank across all CPUs is 85. Consequently, the CPU is in the top tier among its immediate peers, sitting close (0.2% difference) to the Ryzen 5 5600F in average benchmark scores.

Yet – in this measured game, the CPU has almost no weight on the final FPS readings. Across every resolution and setting the frame rate augmentation is dominated by the GPU, as the differences track with pixel count rather than with CPU workload. The CPU’s strong multithread performance (20363) and the 1 60 GB/s memory bandwidth is more than enough for a simulation, so no core-starving is appears in the data. The CPU’s single-threaded marked low only decreases game performance; given that 1080p Ultra still reaches only 59 FPS, it becomes clear that the GPU is used to its absolute limit. Since the average benchmark score of the 5500X3D is 37018, the observed FPS decode could not be smaller as a fraction of the CPU’s capability.

One might question whether the game has specific acceleration for the L3 cache, but the data does not indicate. The combo rank of 3310 out of 3723 confirms that this processor-GPU pairing is close to the bottom (about 11th percentile of all tested pairs). In practice, users seeking higher frame rates should rather replace the graphics solution; the CPU would almost certainly sustain far higher FPS in a hypothetical point if it were paired with a discrete card. The conclusion is that the 5500X3D chips are fully capable, but their role in this game is solely to feed a weak GPU that frames drop, making the CPU the inactive but ready partner.

FAQ

Q: Is a 4K 60 FPS target possible with this combination?

A: No. The measured data shows only 20 FPS at 4K Ultra and 41 FPS at 4K Low. Even the Medium preset reaches only 29 FPS. There is no 4K setting that approaches playable smoothness.

Q: What is the fastest setting to pair with this CPU/GPU at 1080p?

A: Low quality is the fastest setting as measured, with an average FPS of 125 at 1920x1080. If higher detail is required, Medium delivers 88 FPS, still a very usable frame rate.

Q: How much faster is the Ryzen 5 5500X3D than its closest rival?

A: According to the nearest rivals list, the Ryzen 5 5500X3D has an average benchmark score of 37018, which is 0.2% above the AMD Ryzen 5 5600F (36945) and 0.2% above the Intel Core Ultra 5 225 (36938). The Intel Core i9-12900T is 0.3% higher.

Q: Does the game rely more on CPU or GPU?

A: The scaling analysis shows the GPU is the clear limitation. As resolution increases across points from 1080P to 4K, FPS plummets (e.g., High settings from 62 to 39 to 21). The CPU’s 85th percentile rank and 12 threads are inconsistent with such an abrupt loss, so GPU-bound.

Q: What are the lowest and highest recorded minimum frames?

A: The dataset includes minimum FPS only for Medium quality. The lowest minimum is 25 FPS at 4K, and the highest is 102 FPS at 1080p. No minimums are provided for Low, High, or Ultra filters.

Q: That is the GPU percentile vs all GPUs?

A: The Intel Arc B390 ranks at the 9th percentile of all GPUs, on par with the GeForce GT 710. This effectively puts the card in the lower – 1% of companions.

GPU Role revisited – Drilling into the Bottleneck

For a deeper look, the GPU’s specifications and benchmark data synthesize its role. The B390 generates a score of 1482 in the 3DMark Steel Nomad DX12 test, which is 1.3% better than the NVIDIA GT 520MX but 2.7% lower than the GT 710 (the latter being its closest lower rival). The attaches a dedicated 3D score of 1536 shading units along with 48 texel textures, yet the lack of dedicated VRAM, as “System Shared,” reduces its localized bandwidth to “System Dependent”, meaning that the system must share the 512 GB/s-capable memory of the CPU. The base clock of 300 MHz and boost of 2500 MHz appear strong for an integrated processor, but the Pixel rate of only 60.00 GPixel/s and texture out (120.0 GTexel/s) do not compensate for the bandwidth vacuum.

The impact is compounded by the model’s power envelope: at 80 W TDP and no display connectors (since it is an IGP port). These numbers are consistently very low even by integration standards. In the Pc Building Simulator measurements, the raw FPS counts range from 20 to 125 depending on resolution, meaning the GPU can only hgain by reflecting power reductions (Low vs Ultra). At 1080p the Low/ Ultra difference represents a 66 FPS jump upward, while at 4K that difference is just 21 FPS (11% to 41 FPS). The GPU is so weak that even lowering quality does not restore smooth 4K play. The insight is that for this game, the CPU rolls above the baseline, and the only real upgrade path is the GPU itself.