NVIDIA CMP 30HX vs NVIDIA GeForce RTX 5090 D Comparison

# Head-to-Head Benchmarks

The benchmark data between these two NVIDIA cards tells a story of two completely different eras. In the shared tests, the NVIDIA GeForce RTX 5090 D dominates every single metric. The Geekbench OpenCL results show the RTX 5090 D scoring 310,674 points against the CMP 30HX’s 65,199 points — a 376.5% advantage. That is not a marginal lead; it is a generational chasm.

The Geekbench Vulkan test is even more lopsided. The RTX 5090 D posts 376,915 points, while the CMP 30HX manages only 62,484 points. That works out to a 503.2% delta in favor of the newer card. For context, this means the RTX 5090 D delivers roughly six times the Vulkan compute performance of the CMP 30HX. The data does not show a single test where the CMP 30HX wins; it loses both head-to-head matchups decisively.

Looking at the broader average benchmark scores reinforces this picture. The RTX 5090 D carries an average benchmark score of 77,712 across all its tested workloads, placing it in the 92nd percentile of all GPUs. The CMP 30HX, with only two benchmark results available, averages 63,842 and sits in the 89th percentile. The gap in average score is 13,870 points, which is substantial, though the percentile difference is narrower than the raw scores might suggest. This indicates that while the CMP 30HX is far behind in absolute performance, it still occupies a respectable position relative to the entire GPU landscape — proof of how much even a modest modern card has advanced over older hardware.

# Where Each One Wins

The RTX 5090 D wins in every measurable category. In compute-heavy workloads like Geekbench OpenCL and Vulkan, it is not merely ahead — it is in a different league. The 376.5% and 503.2% deltas respectively mean that any task relying on general-purpose GPU compute, ray tracing acceleration, or modern graphics APIs will favor the RTX 5090 D overwhelmingly.

The CMP 30HX has no benchmark victories to claim. However, its role as a mining-specific card means its strengths lie outside traditional graphics benchmarks. With no display outputs, it is not designed for gaming or workstation visualization. Its 125 W TDP and 300 W suggested PSU make it a low-power option for dedicated compute tasks, and its PCIe 1.0 x4 interface is sufficient for mining workloads that do not require high-bandwidth host communication. The data shows it is an end-of-life product, so its practical relevance today is limited to legacy mining operations or niche compute scenarios where its low power draw is an advantage.

For the RTX 5090 D, the wins extend beyond raw compute. Its dual-slot design, PCIe 5.0 x16 interface, and modern display outputs (1x HDMI 2.1b, 3x DisplayPort 2.1b) make it suitable for gaming, content creation, and professional visualization. The 32 GB GDDR7 memory with 1.79 TB/s bandwidth is another area where it simply overwhelms the CMP 30HX’s 6 GB GDDR6 with 336 GB/s. Any workload that is memory-bound — large language models, 8K video editing, or high-resolution texture streaming — will see enormous benefits on the RTX 5090 D.

# Architecture Differences

The two cards are built on fundamentally different architectures, and the data reflects this. The RTX 5090 D uses the GB202 chip based on Blackwell 2.0 architecture, manufactured on a 5 nm process at TSMC. It packs 92,200 million transistors onto a 750 mm² die, yielding a transistor density of 122.9 million per mm². The CMP 30HX, by contrast, uses the TU116 chip based on the older Turing architecture, built on a 12 nm process. It has just 6,600 million transistors on a 284 mm² die, with a density of 23.2 million per mm². That is a massive difference — the RTX 5090 D has roughly 14 times the transistor count of the CMP 30HX.

The core configurations are equally divergent. The RTX 5090 D features 21,760 shading units, 680 texture mapping units, 176 ROPs, 170 ray tracing cores, and 680 tensor cores. The CMP 30HX has 1,408 shading units, 88 TMUs, and 48 ROPs — and crucially, it has no ray tracing cores and no tensor cores at all. This absence of dedicated RT and tensor hardware means the CMP 30HX cannot accelerate ray-traced workloads or AI inference tasks that rely on tensor core operations. The RTX 5090 D’s 680 tensor cores and 170 RT cores are a major reason for its benchmark dominance.

Clock speeds also differ significantly. The RTX 5090 D has a base clock of 2017 MHz and a boost clock of 2407 MHz, while the CMP 30HX runs at 1530 MHz base and 1785 MHz boost. The memory clocks tell a similar story: the RTX 5090 D uses 1750 MHz memory (28 Gbps effective), while the CMP 30HX uses 1750 MHz memory but only 14 Gbps effective. The RTX 5090 D’s memory bus is 512 bits wide versus 192 bits, and its bandwidth of 1.79 TB/s is more than five times the CMP 30HX’s 336 GB/s.

# The Verdict

The data is unambiguous: the NVIDIA GeForce RTX 5090 D is the superior product by every measurable metric. Its average benchmark score of 77,712 places it in the 92nd percentile, while the CMP 30HX sits at 89th percentile with an average of 63,842. In head-to-head tests, the RTX 5090 D leads by 376.5% in OpenCL and 503.2% in Vulkan. If your workload requires maximum compute performance, modern graphics API support, ray tracing, or tensor core acceleration, the RTX 5090 D is the only choice from this pair.

Who should pick the CMP 30HX? The data suggests very few scenarios. It is an end-of-life mining card with no display outputs, no RT cores, no tensor cores, and a PCIe 1.0 x4 interface. Its only advantages are a lower 125 W TDP and a 300 W suggested PSU, which could matter in power-constrained or heat-sensitive environments. But even there, its 5.027 TFLOPS FP32 performance is a fraction of the RTX 5090 D’s 104.8 TFLOPS. For any modern application, the RTX 5090 D is the clear winner, and the CMP 30HX should only be considered for legacy mining rigs where its specific design is still functional.

# FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce RTX 5090 D has an average benchmark score of 77,712, compared to the CMP 30HX’s 63,842.

Q: How much faster is the RTX 5090 D in Geekbench Vulkan?

A: The RTX 5090 D scores 376,915 in Geekbench Vulkan, which is 503.2% higher than the CMP 30HX’s 62,484.

Q: Does the CMP 30HX have ray tracing or tensor cores?

A: No, the CMP 30HX has no ray tracing cores and no tensor cores, according to the specification data.

Q: What are the memory sizes of these two GPUs?

A: The RTX 5090 D has 32 GB of GDDR7 memory, while the CMP 30HX has 6 GB of GDDR6 memory.

Q: Which card has a higher percentile ranking among all GPUs?

A: The RTX 5090 D is in the 92nd percentile, while the CMP 30HX is in the 89th percentile.

Q: What is the transistor density difference between these cards?

A: The RTX 5090 D has a transistor density of 122.9 million per mm², while the CMP 30HX has 23.2 million per mm².

# Specification Differences

| Specification | NVIDIA GeForce RTX 5090 D | NVIDIA CMP 30HX |

|---|---|---|

| Architecture | Blackwell 2.0 | Turing |

| Process Node | 5 nm | 12 nm |

| Transistors | 92,200 million | 6,600 million |

| Die Size | 750 mm² | 284 mm² |

| Transistor Density | 122.9M / mm² | 23.2M / mm² |

| Base Clock | 2017 MHz | 1530 MHz |

| Boost Clock | 2407 MHz | 1785 MHz |

| Memory Size | 32 GB | 6 GB |

| Memory Type | GDDR7 | GDDR6 |

| Memory Bus Width | 512 bit | 192 bit |

| Memory Bandwidth | 1.79 TB/s | 336.0 GB/s |

| Shading Units | 21,760 | 1,408 |

| TMUs | 680 | 88 |

| ROPs | 176 | 48 |

| RT Cores | 170 | None |

| Tensor Cores | 680 | None |

| Pixel Rate | 423.6 GPixel/s | 85.68 GPixel/s |

| Texture Rate | 1,636.8 GTexel/s | 157.1 GTexel/s |

| FP32 Performance | 104.8 TFLOPS | 5.027 TFLOPS |

| FP16 Performance | 104.8 TFLOPS (1:1) | 10.05 TFLOPS (2:1) |

| TDP | 575 W | 125 W |

| Power Connectors | 1x 16-pin | 1x 8-pin |

| Suggested PSU | 950 W | 300 W |

| Bus Interface | PCIe 5.0 x16 | PCIe 1.0 x4 |

| Display Outputs | 1x HDMI 2.1b, 3x DisplayPort 2.1b | No outputs |

| DirectX Support | 12 Ultimate (12_2) | 12 (12_1) |

| Length | 304 mm | 229 mm |

| Height | 137 mm | 111 mm |

| Width | 48 mm | 35 mm |

| Production Status | Active | End-of-life |

| Release Date | 2025-01-29 | 2021-02-24 |

| Launch MSRP | 2,299 USD | 799 USD |