NVIDIA CMP 30HX vs NVIDIA Quadro RTX 6000 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA Quadro RTX 6000 records an average benchmark score of 101872, while the NVIDIA CMP 30HX scores 63842. That places the Quadro RTX 6000 in the 94th percentile of all GPUs in the database, versus the 89th percentile for the CMP 30HX.

Q: How do the two cards compare in OpenCL performance?

A: The Quadro RTX 6000 scores 74179 in Geekbench OpenCL, which is 13.8% ahead of the CMP 30HX's 65199. Both cards show their strongest relative results in this API, but the gap is moderate compared to Vulkan.

Q: What is the biggest performance difference between these two cards?

A: In Geekbench Vulkan, the Quadro RTX 6000 scores 129564 against 62484 for the CMP 30HX, a delta of 107.4%. This is the largest single-test margin in the head-to-head data, more than doubling the CMP 30HX's result.

Q: Do both cards use the same architecture?

A: Yes, both are built on NVIDIA's Turing architecture and use a 12 nm process from TSMC. However, they use different chips: the Quadro RTX 6000 uses TU102, while the CMP 30HX uses TU116.

Q: Which card has more memory and bandwidth?

A: The Quadro RTX 6000 offers 24 GB of GDDR6 on a 384-bit bus, yielding 672.0 GB/s of bandwidth. The CMP 30HX has 6 GB of GDDR6 on a 192-bit bus, delivering 336.0 GB/s, exactly half the bandwidth.

Q: What are the nearest rivals for each card in the database?

A: For the Quadro RTX 6000, the closest competitors are the AMD Radeon Pro Vega II Duo (4.6% higher average score) and the AMD Radeon Pro W6600X (5.1% higher). For the CMP 30HX, the AMD Radeon RX 9060 XT LP is essentially tied at 0.0% delta, with the AMD Radeon RX 7600M just 0.1% behind.

Architecture Differences

Both cards share the Turing architecture and the same 12 nm TSMC process node, but the silicon underneath is very different in scale. The Quadro RTX 6000 uses the TU102 chip with 18,600 million transistors on a 754 mm² die, giving a transistor density of 24.7M per mm². The CMP 30HX uses the TU116 chip with 6,600 million transistors on a 284 mm² die, a density of 23.2M per mm². The TU102 is nearly three times the transistor count and more than 2.5 times the die area, which explains the magnitude of the performance gap.

The most consequential architectural difference is the presence of specialized hardware. The Quadro RTX 6000 includes 72 RT cores and 576 tensor cores, enabling hardware-accelerated ray tracing and AI workloads. The CMP 30HX has no RT cores and no tensor cores at all. This is a fundamental split: the Quadro is a full-featured workstation GPU, while the CMP 30HX is a stripped-down mining-oriented card with those blocks removed entirely.

Compute resources also differ sharply. The Quadro RTX 6000 packs 4608 shading units, 288 texture mapping units, and 96 ROPs. The CMP 30HX has 1408 shading units, 88 TMUs, and 48 ROPs. That is roughly one-third of the shading units and TMUs, and half the ROPs. The raw FP32 throughput follows: 16.31 TFLOPS for the Quadro versus 5.027 TFLOPS for the CMP 30HX. FP16 performance scales similarly at 32.62 TFLOPS versus 10.05 TFLOPS, both at a 2:1 ratio.

Memory architecture is another major divider. The Quadro RTX 6000 uses a 384-bit bus with 24 GB of GDDR6, while the CMP 30HX uses a 192-bit bus with 6 GB. Bandwidth is exactly double on the Quadro (672.0 GB/s versus 336.0 GB/s). Clock speeds are surprisingly close: the Quadro runs at 1440 MHz base and 1770 MHz boost, while the CMP 30HX runs slightly higher at 1530 MHz base and 1785 MHz boost. The CMP 30HX's higher clocks do not compensate for its much smaller core configuration.

The bus interface also differs. The Quadro RTX 6000 uses PCIe 3.0 x16, while the CMP 30HX uses PCIe 1.0 x4, a far narrower and older interface. Display outputs tell the story of intended use: the Quadro has 4x DisplayPort 1.4a and 1x USB Type-C, while the CMP 30HX has no outputs at all. API support also differs, with the Quadro supporting DirectX 12 Ultimate (12_2) versus DirectX 12 (12_1) for the CMP 30HX, though both support OpenGL 4.6 and Vulkan 1.4.

The Verdict

The data points to a clear conclusion: the NVIDIA Quadro RTX 6000 is the superior card in every recorded benchmark, with an average score 59.6% higher than the CMP 30HX (101872 versus 63842). It wins both head-to-head tests, including a 107.4% margin in Vulkan. For any workload that relies on compute performance, RT cores, tensor cores, or display output, the Quadro RTX 6000 is the only viable choice between these two.

The CMP 30HX does have one meaningful advantage: power efficiency. Its TDP of 125 W is less than half of the Quadro's 260 W, and it requires only a 300 W suggested PSU versus 600 W. It also has a shorter physical length (229 mm versus 267 mm). But the CMP 30HX lacks display outputs entirely, making it useless for any conventional desktop or workstation use case. Its position in the 89th percentile with an average score of 63842 puts it in the range of the AMD Radeon RX 9060 XT LP and RX 7600M, but that is a different performance class than the Quadro's 94th percentile.

Who should pick which? If the task involves rendering, graphics, AI inference, or any workload requiring a display, the Quadro RTX 6000 is the answer, and the 24 GB memory capacity alone justifies it over the 6 GB CMP 30HX. If the only goal is raw compute in a headless environment with minimal power draw and the workload does not need RT or tensor cores, the CMP 30HX could serve, but the benchmark data shows it delivers far less performance per dollar of capability. The recorded numbers do not favor the CMP 30HX in any compute scenario.

Specification Differences

The two cards differ across nearly every specification field in the database:

• Chip: TU102 (Quadro RTX 6000) versus TU116 (CMP 30HX)
• Transistors: 18,600 million versus 6,600 million
• Die size: 754 mm² versus 284 mm²
• Transistor density: 24.7M / mm² versus 23.2M / mm²
• Base clock: 1440 MHz versus 1530 MHz
• Boost clock: 1770 MHz versus 1785 MHz
• Memory size: 24 GB versus 6 GB
• Memory bus width: 384 bit versus 192 bit
• Memory bandwidth: 672.0 GB/s versus 336.0 GB/s
• Shading units: 4608 versus 1408
• TMUs: 288 versus 88
• ROPs: 96 versus 48
• RT cores: 72 versus none
• Tensor cores: 576 versus none
• Pixel rate: 169.9 GPixel/s versus 85.68 GPixel/s
• Texture rate: 509.8 GTexel/s versus 157.1 GTexel/s
• FP32: 16.31 TFLOPS versus 5.027 TFLOPS
• FP16: 32.62 TFLOPS versus 10.05 TFLOPS
• TDP: 260 W versus 125 W
• Power connectors: 1x 6-pin + 1x 8-pin versus 1x 8-pin
• Suggested PSU: 600 W versus 300 W
• Bus interface: PCIe 3.0 x16 versus PCIe 1.0 x4
• Display outputs: 4x DisplayPort 1.4a + 1x USB Type-C versus none
• DirectX support: 12 Ultimate (12_2) versus 12 (12_1)
• Length: 267 mm versus 229 mm
• Width: not specified versus 35 mm
• Release date: 2018-08-12 versus 2021-02-24
• Launch MSRP: 6,299 USD versus 799 USD

Head-to-Head Benchmarks

The head-to-head data contains only two benchmark entries, but they tell a consistent story. In Geekbench OpenCL, the Quadro RTX 6000 scores 74179 against 65199 for the CMP 30HX, a 13.8% advantage. This is the closer of the two tests, suggesting that raw compute throughput per shader is relatively similar, and the Quadro wins primarily through sheer scale: 4608 shading units versus 1408, 288 TMUs versus 88, and 384-bit memory versus 192-bit.

The Vulkan result is far more lopsided. The Quadro RTX 6000 scores 129564, more than double the CMP 30HX's 62484, for a 107.4% delta. This dramatic widening suggests that Vulkan workloads leverage features the CMP 30HX lacks, likely the RT cores and tensor cores, or the larger memory bandwidth available on the 384-bit bus. The CMP 30HX's higher boost clock (1785 MHz versus 1770 MHz) does not rescue it here; the architectural deficits are too large.

Looking at the broader benchmark context, the Quadro RTX 6000's average score of 101872 places it 4.5% above the AMD Radeon RX 7900M and 4.9% above the AMD Radeon Pro VII, while sitting 4.6% below the AMD Radeon Pro Vega II Duo. The CMP 30HX's average of 63842 is essentially tied with the AMD Radeon RX 9060 XT LP (0.0% delta) and the AMD Radeon RX 7600M (0.1% ahead). The gap between the two cards in the database, 101872 versus 63842, is roughly the same order of magnitude as the gap between the CMP 30HX and the lowest-tier AMD workstation cards.

Where Each One Wins

The Quadro RTX 6000 wins every recorded benchmark, so the analysis of "where each one wins" is primarily about workload types implied by the specification differences rather than benchmark victories.

The Quadro RTX 6000 wins in any scenario involving graphics output. It has 4x DisplayPort 1.4a and 1x USB Type-C, while the CMP 30HX has no outputs at all. It also wins in ray tracing and AI workloads thanks to its 72 RT cores and 576 tensor cores, which the CMP 30HX completely lacks. The 24 GB memory capacity versus 6 GB makes the Quadro the clear choice for large datasets, high-resolution textures, or multi-application workflows. Its DirectX 12 Ultimate support (12_2) also gives it access to newer rendering features that the CMP 30HX's DirectX 12 (12_1) cannot use.

The CMP 30HX wins in power-constrained environments. Its 125 W TDP is less than half of the Quadro's 260 W, and its 300 W suggested PSU requirement is half of the Quadro's 600 W. It is also physically shorter at 229 mm versus 267 mm, which could matter in compact chassis. The CMP 30HX's higher base clock (1530 MHz versus 1440 MHz) and boost clock (1785 MHz versus 1770 MHz) mean it extracts more performance per watt from its smaller TU116 die, but the absolute performance is still far below the Quadro in every test.

For a headless compute node where power draw is the primary concern and the workload does not require RT cores, tensor cores, or display output, the CMP 30HX is the more efficient option per watt. But the data does not support any scenario where the CMP 30HX outperforms the Quadro RTX 6000 in raw speed. The Quadro's 13.8% OpenCL lead and 107.4% Vulkan lead are decisive, and its 94th percentile ranking versus 89th puts it in a different performance tier entirely.