AMD Radeon Pro WX 9100 vs NVIDIA CMP 30HX Comparison

Launched with a 799 USD MSRP in February 2021, the NVIDIA CMP 30HX is a Turing-based mining card with no display outputs, while the AMD Radeon Pro WX 9100, launched at 1,599 USD in July 2017, is a workstation GPU with six mini-DisplayPort outputs. The benchmark data shows these two end-of-life cards are far closer in raw compute performance than their release dates and target markets suggest, with the CMP 30HX averaging 64,172 points to the WX 9100's 64,002 points.

Head-to-Head Benchmarks

The most decisive victory for the NVIDIA CMP 30HX comes in the Geekbench Vulkan test, where it scores 61,336 against the Radeon Pro WX 9100's 54,711. That is a 12.1% delta in favor of the mining card, a substantial margin that highlights NVIDIA's stronger driver optimization for Vulkan workloads on the Turing architecture. The CMP 30HX also wins the Geekbench OpenCL test, scoring 67,007 versus 65,210 for the WX 9100, a narrower 2.8% advantage. Across the two shared benchmarks, the CMP 30HX wins both, giving it a 2-0 sweep in head-to-head comparisons.

Looking at the broader average benchmark score, the CMP 30HX's 64,172 average places it 0.3% ahead of the WX 9100's 64,002. This is a razor-thin margin, nearly a statistical tie. The nearest rival data reinforces this picture: the AMD Radeon VII sits just 0.3% below the CMP 30HX, while the Radeon RX 7600M trails by 1.1% and the Radeon RX 7800M by 2.9%. From the WX 9100's perspective, it is 0.6% behind the Radeon VII and 0.8% ahead of the RX 7600M, with a 2.6% lead over the RX 7800M. The data shows both cards occupy the same performance tier, with the CMP 30HX holding a slight edge in every direct comparison.

The Vulkan result is particularly telling because it is the only test where the delta exceeds 5%. The CMP 30HX's 12.1% lead in Vulkan is a significant outlier compared to the OpenCL result and the average score differences. This suggests that while the two cards are comparable in general-purpose compute, NVIDIA's Turing architecture has a clear advantage in modern graphics APIs. The WX 9100's best result comes in the Geekbench Metal test, where it scores 72,085, but the CMP 30HX has no Metal benchmark data, so this cannot be compared directly.

The Verdict

Based strictly on benchmark data, the NVIDIA CMP 30HX is the faster card in every head-to-head test. It wins OpenCL by 2.8% and Vulkan by 12.1%, and its average benchmark score of 64,172 is 0.3% higher than the WX 9100's 64,002. For raw compute performance, the CMP 30HX is the pick. However, the WX 9100 has one critical advantage that the benchmarks do not capture: it has display outputs. The CMP 30HX has no outputs at all, making it useless for any visual task. The WX 9100 offers six mini-DisplayPort 1.4a connections, enabling multi-monitor setups and workstation visualization.

The data also shows the WX 9100 has 16 GB of HBM2 memory with 483.8 GB/s bandwidth, compared to the CMP 30HX's 6 GB GDDR6 with 336.0 GB/s. While the CMP 30HX wins in compute benchmarks, the WX 9100's memory capacity is more than double, which matters for large datasets. The WX 9100 also has a higher pixel rate at 96.00 GPixel/s versus 85.68 GPixel/s, and a much higher texture rate at 384.0 GTexel/s versus 157.1 GTexel/s. These specifications suggest the WX 9100 is better suited for geometry-heavy and texture-heavy workloads, even if the CMP 30HX wins in the synthetic benchmarks recorded here.

Architecture Differences

The NVIDIA CMP 30HX is built on the TU116 chip using the Turing architecture, fabricated on a 12 nm process at TSMC. It packs 6,600 million transistors into a 284 mm² die, yielding a transistor density of 23.2M per mm². The AMD Radeon Pro WX 9100 uses the Vega 10 chip with the GCN 5.0 architecture, fabricated on a 14 nm process at GlobalFoundries. It contains 12,500 million transistors on a 495 mm² die, giving a transistor density of 25.3M per mm². The WX 9100 is the physically larger chip with nearly double the transistor count.

Clock speeds differ significantly. The CMP 30HX runs at a base clock of 1530 MHz and a boost clock of 1785 MHz, while the WX 9100 runs at 1200 MHz base and 1500 MHz boost. The higher clocks on the CMP 30HX help it compete despite having far fewer compute units. The CMP 30HX has 1408 shading units, 88 TMUs, and 48 ROPs, while the WX 9100 has 4096 shading units, 256 TMUs, and 64 ROPs. The WX 9100's raw hardware resources are much larger, but the CMP 30HX's higher clocks and architectural efficiency close the gap.

Memory configurations are completely different. The CMP 30HX uses 6 GB of GDDR6 on a 192-bit bus with a memory clock of 1750 MHz effective 14 Gbps, yielding 336.0 GB/s bandwidth. The WX 9100 uses 16 GB of HBM2 on a 2048-bit bus with a memory clock of 945 MHz effective 1890 Mbps, yielding 483.8 GB/s bandwidth. The WX 9100's memory bandwidth is 44% higher, and its memory capacity is nearly triple. The CMP 30HX's FP32 performance is 5.027 TFLOPS, while the WX 9100 achieves 12.29 TFLOPS, a 2.4x advantage. FP16 performance is similarly lopsided: 10.05 TFLOPS versus 24.58 TFLOPS.

Power requirements differ substantially. The CMP 30HX has a 125 W TDP with a 300 W suggested PSU and a single 8-pin power connector, while the WX 9100 has a 230 W TDP with a 550 W suggested PSU and requires both a 6-pin and an 8-pin connector. The bus interfaces also differ: the CMP 30HX uses PCIe 1.0 x4, while the WX 9100 uses PCIe 3.0 x16. The WX 9100 supports Vulkan 1.3, while the CMP 30HX supports Vulkan 1.4. Both support DirectX 12 (12_1) and OpenGL 4.6.

FAQ

Q: Which card is faster in Geekbench OpenCL?

A: The NVIDIA CMP 30HX scores 67,007 points compared to the AMD Radeon Pro WX 9100's 65,210 points, giving the CMP 30HX a 2.8% advantage in OpenCL performance.

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

A: The NVIDIA CMP 30HX wins decisively with a score of 61,336 versus 54,711 for the AMD Radeon Pro WX 9100, a 12.1% lead in favor of the CMP 30HX.

Q: Does the AMD Radeon Pro WX 9100 have a higher average benchmark score?

A: No. The NVIDIA CMP 30HX has an average benchmark score of 64,172, while the AMD Radeon Pro WX 9100 averages 64,002. The CMP 30HX leads by 0.3%.

Q: Can the NVIDIA CMP 30HX be used for multi-monitor setups?

A: No. The NVIDIA CMP 30HX has no display outputs, whereas the AMD Radeon Pro WX 9100 offers six mini-DisplayPort 1.4a outputs.

Q: Which card has more memory bandwidth?

A: The AMD Radeon Pro WX 9100 has 483.8 GB/s bandwidth from its 16 GB HBM2 memory, while the NVIDIA CMP 30HX has 336.0 GB/s from 6 GB GDDR6.

Q: What is the transistor count on each card?

A: The AMD Radeon Pro WX 9100 contains 12,500 million transistors, while the NVIDIA CMP 30HX contains 6,600 million transistors.

Where Each One Wins

The NVIDIA CMP 30HX wins in raw compute benchmarks. It beats the WX 9100 by 2.8% in OpenCL and by 12.1% in Vulkan, and it has a higher average benchmark score overall. Its higher clock speeds (1530 MHz base, 1785 MHz boost) and newer Turing architecture give it the edge in synthetic testing. The CMP 30HX also wins on power efficiency, with a 125 W TDP versus the WX 9100's 230 W TDP, and it requires only a 300 W PSU versus 550 W. The CMP 30HX is the card for applications that rely on OpenCL and Vulkan compute performance, provided display output is not needed.

The AMD Radeon Pro WX 9100 wins in memory capacity and bandwidth. With 16 GB of HBM2 on a 2048-bit bus, it offers 483.8 GB/s bandwidth, which is 44% higher than the CMP 30HX's 336.0 GB/s. Its FP32 throughput of 12.29 TFLOPS is 2.4x higher than the CMP 30HX's 5.027 TFLOPS, and its texture rate of 384.0 GTexel/s is more than double the CMP 30HX's 157.1 GTexel/s. The WX 9100 also wins on display connectivity with six mini-DisplayPort 1.4a outputs. This card is the choice for workloads that need large memory pools, high geometry throughput, or multi-display output.

The data also shows the WX 9100 has a higher pixel rate at 96.00 GPixel/s versus 85.68 GPixel/s for the CMP 30HX. For anyone needing a workstation card with visual output, the WX 9100 is the only option between these two. The CMP 30HX, by design, is a mining-focused card with no outputs, so its wins are purely in compute benchmarks. The Radeon Pro WX 9100's nearest rival data shows it is 0.6% behind the Radeon VII and 0.8% ahead of the RX 7600M, placing it in the same performance band as the CMP 30HX. The CMP 30HX's nearest rival data shows it is 0.3% behind the Radeon VII and 1.1% ahead of the RX 7600M, confirming the two cards are near-peers in average performance.