Overpower

How This Combo Ranks

This CPU/GPU pairing sits at rank 127 out of 2,980 tested combinations for Overpower, placing it comfortably inside the top 5% of all tested systems. That rank is a strong indicator of a high-performance pairing, but it is not the absolute top tier — there are 126 other combos that deliver higher average frame rates in this specific action title. The data suggests this is a very capable combination for high-refresh-rate gaming, though not the absolute ceiling of what the benchmark database has recorded.

The individual components also carry strong standing within their own categories. The AMD Ryzen 7 5800X3D ranks in the 80th percentile against all CPUs in the database, while the NVIDIA GeForce RTX 5090 D sits in the 92nd percentile for GPUs. The GPU's percentile is notably higher than the CPU's, which hints that the graphics card is the more dominant partner in this pairing. The CPU's average benchmark score of 27,896 places it in a tight cluster with its nearest rivals — the AMD Ryzen 7 6800U at 27,887 (0% delta), the Intel Core i9-10900K at 27,872 (0.1% ahead), the AMD Ryzen AI 5 PRO 340 at 27,932 (0.1% behind), and the Intel Core i5-12600KF at 27,799 (0.3% ahead). These deltas are all within a fraction of a percent, meaning the 5800X3D is essentially neck-and-neck with these alternatives in raw average benchmark score.

The GPU's benchmark positioning is more spread out. With an average score of 77,712, the RTX 5090 D is 1.1% ahead of the AMD Radeon RX 6650M XT (76,904), but trails the AMD Radeon RX 6850M XT by 1.6% (78,940), the NVIDIA Tesla P100 PCIe 12 GB by 2.1% (79,396), and the NVIDIA Tesla P100 PCIe 16 GB by 2.4% (79,605). The pattern is interesting: the RTX 5090 D is being compared against both mobile and data-center parts, and its scores place it in the middle of that group, not at the top. This suggests that while the card is strong, its raw compute benchmarks do not dominate the field as much as the 92nd percentile might suggest on its own.

GPU Role

The NVIDIA GeForce RTX 5090 D is a massive piece of silicon. Built on the Blackwell 2.0 architecture with a GB202 chip, it contains 92,200 million transistors on a 750 mm² die, fabricated on a 5 nm process from TSMC. The card features 21,760 shading units, 680 TMUs, and 176 ROPs, along with 170 ray tracing cores and 680 tensor cores. The base clock runs at 2,017 MHz, boosting to 2,407 MHz, which drives a pixel rate of 423.6 GPixel/s and a texture rate of 1,636.8 GTexel/s. Floating-point performance is rated at 104.8 TFLOPS for both FP32 and FP16 (1:1).

Memory is where this card separates itself from most of the field. The RTX 5090 D carries 32 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. For Overpower, this memory configuration is almost certainly not a limiting factor at any of the tested settings. The game's data shows no frame rate collapses that would indicate VRAM exhaustion — even at 4K Ultra, the card maintains a playable average. The memory clock is listed at 1,750 MHz with an effective 28 Gbps, which feeds the enormous bus width to produce that 1.79 TB/s figure.

The card's benchmark results are mixed across different test types. In 3DMark Steel Nomad DX12, it scores 14,326. Geekbench OpenCL returns 310,674, while Vulkan scores 376,915 — the Vulkan advantage being notable. Passmark results show a G3D score of 44,065 and GPU compute of 28,396. The DirectX tests are more unusual: DirectX 10 scores 231, DirectX 11 scores 371, DirectX 12 scores 219, and DirectX 9 scores 434. These lower Passmark DirectX scores, compared to the much higher Geekbench and 3DMark numbers, suggest the card's architecture may be more optimized for certain API paths than others. The 12_2 feature level for DirectX 12 Ultimate is supported, which aligns with the game's modern rendering requirements.

The card's physical specifications are equally imposing. It is 304 mm long, 137 mm tall, and 48 mm wide, occupying a dual-slot form factor. It requires a 1x 16-pin power connector and has a suggested PSU rating of 950 W, with a TDP of 575 W. Display outputs include 1x HDMI 2.1b and 3x DisplayPort 2.1b. The card uses a PCIe 5.0 x16 interface.

CPU Role

The AMD Ryzen 7 5800X3D is the older component in this pairing, released in April 2022 and now marked as end-of-life. It is built on the Zen 3 architecture (codenamed Vermeer) using a 7 nm TSMC process, with 8,850 million transistors on a 74 mm² die. The processor has 8 cores and 16 threads, with a base clock of 3.40 GHz and a boost clock of 4.50 GHz. The standout feature is the cache layout: 64 KB of L1 per core, 512 KB of L2 per core, and a massive 96 MB of shared L3 cache. That large L3 cache is the key differentiator for this chip, designed to reduce memory latency in gaming workloads.

Memory support is DDR4 with a dual-channel bus, providing 51.2 GB/s of bandwidth. It supports ECC memory and uses the AMD Socket AM4 platform with PCIe Gen 4 (20 lanes from the CPU). The chip is not multiplier-unlocked, meaning overclocking is restricted by design. Its TDP is 105 W, and it was launched with an MSRP of $449.

Single-threaded performance is a relative weakness for this chip in synthetic tests. It scores 884 in 3DMark single-thread, 241 in Cinebench R15 single-core, 1,452 in Cinebench R23 single-core, 2,017 in Geekbench single-core, and 3,234 in Passmark single-thread. These are respectable but not class-leading numbers. Multi-threaded scores are stronger: 7,299 in 3DMark 16-thread, 7,285 in 3DMark max-threads, 2,413 in Cinebench R15 multicore, 14,317 in Cinebench R23 multicore, 11,765 in Geekbench multicore, and 28,309 in Passmark multithread. The 3DMark 2-thread score of 1,746 and 4-thread score of 3,392 show a reasonable scaling curve, while 8-thread at 5,906 indicates the chip uses its core count effectively.

For Overpower, the CPU's role appears to be sufficient but not the primary driver. The measured FPS data shows that frame rates scale more with resolution changes than with settings changes, which typically indicates GPU-bound scenarios. The 5800X3D's large L3 cache likely helps maintain consistent frame pacing in this action game, but the chip's single-thread scores suggest it is not the bottleneck in most situations.

Measured FPS Breakdown

The measured data covers three resolutions and four settings presets, with 1080p and 1440p showing the most complete picture.

At 1920x1080, the averages are: Low settings 214 FPS, High settings 174 FPS, Medium 178 FPS with a minimum of 152 and maximum of 205, and Ultra 157 FPS. The spread from Low to Ultra is 57 FPS, which is the narrowest of any resolution. Interestingly, Medium beats High by 4 FPS at 1080p, a small but real inversion.

At 2560x1440, the numbers are: Low 206 FPS, High 167 FPS, Medium 172 FPS with a minimum of 146 and maximum of 197, and Ultra 140 FPS. The spread from Low to Ultra widens to 66 FPS. Medium again edges out High by 5 FPS.

At 3840x2160, the results are: Low 195 FPS, High 144 FPS, Medium 155 FPS with a minimum of 132 and maximum of 178, and Ultra 119 FPS. The spread from Low to Ultra is now 76 FPS, the largest of the three resolutions.

The pattern across all resolutions is consistent: Low is always the fastest, Ultra is always the slowest, and Medium sits between High and Low in most cases, though it exceeds High in both 1080p and 1440p. The minimum FPS values are only recorded for Medium settings — 152 at 1080p, 146 at 1440p, and 132 at 4K — which gives a sense of worst-case frame times at that preset.

Resolution Scaling

The data reveals how this combo scales when moving from 1080p to 4K. At Low settings, the drop from 1080p to 1440p is 8 FPS (214 to 206), and from 1440p to 4K is 11 FPS (206 to 195). The total drop from 1080p to 4K at Low is 19 FPS, a reduction of about 8.9%. At High settings, the drops are 7 FPS (174 to 167) and 23 FPS (167 to 144), totaling 30 FPS or about 17.2%. At Medium, the drops are 6 FPS (178 to 172) and 17 FPS (172 to 155), totaling 23 FPS or about 12.9%. At Ultra, the drops are 17 FPS (157 to 140) and 21 FPS (140 to 119), totaling 38 FPS or about 24.2%.

The scaling pattern is informative. At Low settings, the resolution impact is minimal — the GPU is so far ahead that even 4K barely stresses it. At Ultra settings, the resolution impact is the largest, with a 24.2% reduction from 1080p to 4K. This indicates that the GPU becomes more of a limiting factor as settings increase, which is expected. However, the absolute FPS at 4K Ultra (119) is still very high, suggesting the RTX 5090 D has enormous headroom in this title.

The fact that the 1080p to 1440p drops are relatively small (ranging from 6 to 17 FPS) while the 1440p to 4K drops are larger (ranging from 11 to 23 FPS) suggests the GPU starts to feel the pixel count more at 4K. The CPU's role appears minimal in resolution scaling — if the CPU were the bottleneck, we would expect similar FPS across resolutions at the same settings, which is not what the data shows. The FPS decreases consistently with resolution at every preset, pointing to GPU-bound performance throughout.

Settings Recommendations

Based on the measured rows, the Medium preset appears to offer the best balance of performance and visual quality. At every resolution, Medium averages are within striking distance of High (4 FPS higher at 1080p, 5 FPS higher at 1440p, and 11 FPS higher at 4K), while presumably offering better visuals than Low. The minimum FPS data is only available for Medium, which allows for frame time analysis: 152 FPS minimum at 1080p, 146 at 1440p, and 132 at 4K. These minimums are all well above refresh-rate thresholds for standard monitors.

However, the Ultra preset is not a poor choice. At 1080p, Ultra averages 157 FPS, which is only 17 FPS below Medium. At 1440p, Ultra averages 140 FPS, and at 4K, it averages 119 FPS. For users with high-refresh-rate displays, the difference between Medium and Ultra may be worth the visual upgrade. The 4K Ultra result of 119 FPS is still above 100 FPS, which is a strong experience.

Low settings provide the highest frame rates — 214 FPS at 1080p, 206 at 1440p, and 195 at 4K — but the visual quality trade-off is likely too severe for a game of this type. The data does not include minimum FPS for Low, so consistency cannot be verified, but the averages are high enough that even significant dips would likely stay above 144 FPS.

The recommendation hinges on display capability. For 144Hz displays, Medium at 4K (155 FPS average) provides comfortable headroom. For 240Hz displays, Low at 1440p or 1080p would be necessary to approach those refresh rates. The Medium preset's minimum FPS values suggest it can hold frame rates that would satisfy most gaming monitors without the frame rate volatility that can occur at Ultra.

FAQ

*Q: How does this combo rank against other tested systems in Overpower?*

A: This pairing ranks 127th out of 2,980 tested combos, placing it in the top 4.3% of all systems tested for this game.

Q: What is the highest average FPS this combo achieves, and at what settings?

A: The highest average FPS is 214, achieved at 1920x1080 with Low settings. At 4K, the highest average is 195 FPS, also at Low settings.

*Q: Does the CPU or GPU bottleneck this system in Overpower?*

A: The data indicates the GPU is the limiting factor at higher settings. FPS scales consistently with resolution at every preset, which is characteristic of GPU-bound performance. The CPU's benchmark scores, while solid, are not exceptional enough to drive higher frame rates if the GPU were removed.

Q: How does the RTX 5090 D's 32 GB of VRAM factor into this game's performance?

A: The 32 GB of GDDR7 memory on a 512-bit bus (1.79 TB/s bandwidth) is more than sufficient for Overpower. No measured frame rate suggests VRAM exhaustion, even at 4K Ultra with 119 FPS average.

Q: What is the FPS difference between 1080p and 4K at Ultra settings?

A: At Ultra, the average drops from 157 FPS at 1080p to 119 FPS at 4K, a reduction of 38 FPS or about 24.2%. This is the largest percentage drop across all settings presets.

Q: Is the Medium preset consistently better than High?

A: Yes, in the measured data, Medium averages higher FPS than High at every resolution: 178 vs 174 at 1080p, 172 vs 167 at 1440p, and 155 vs 144 at 4K. The advantage ranges from 4 to 11 FPS.