AMD Radeon RX 6900 XT vs NVIDIA CMP 30HX Comparison

NVIDIA CMP 30HX vs. AMD Radeon RX 6900 XT: a head-to-head comparison of two end-of-life GPUs from the mining and gaming segments. The data reveals a clear performance hierarchy, but the architectural story is more nuanced than a simple scoreboard.

Head-to-Head Benchmarks

The recorded data shows a decisive sweep for the AMD Radeon RX 6900 XT in the two shared benchmark tests. In Geekbench OpenCL, the RX 6900 XT scores 187,673 against the CMP 30HX’s 65,199. That is a delta of -65.3% from the perspective of the NVIDIA card, meaning the AMD part is roughly 2.9 times faster in this compute-oriented test. The gap is narrower but still substantial in Geekbench Vulkan: 148,525 for the AMD card versus 62,484 for the NVIDIA card, a -57.9% delta. The CMP 30HX wins zero head-to-head tests, while the RX 6900 XT wins both.

These are not marginal differences. The RX 6900 XT’s OpenCL score is nearly triple the CMP 30HX’s, and its Vulkan score is about 2.4 times higher. Looking at the broader database, the CMP 30HX’s average benchmark score is 63,842, placing it at the 89th percentile of all GPUs. Its nearest rival is the AMD Radeon RX 9060 XT LP at 63,830 (a 0% delta), showing it sits in a solid mid-to-high tier. However, the RX 6900 XT, despite its lower 86th percentile, has an average score of 50,951. This percentile discrepancy is curious: the CMP 30HX has a higher percentile but a lower peak score. The explanation lies in the benchmark mix. The RX 6900 XT’s average is dragged down by the low Passmark DirectX scores (113 to 279), while its Geekbench and Passmark G3D results (26,732) are far stronger. The CMP 30HX, with only two recorded tests, benefits from a narrower, more consistent sample.

The RX 6900 XT’s nearest rivals in the database include the NVIDIA CMP 50HX (avg score 51,790, a -1.6% delta) and the AMD Radeon RX Vega 64 (50,001, a +1.9% delta). The CMP 30HX, by contrast, is bracketed by the RX 9060 XT LP and the Radeon Pro WX 9100 (64,212, a -0.6% delta). This placement reinforces that the RX 6900 XT, despite its lower percentile, is a higher-performance part in absolute terms, especially in modern APIs.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. The NVIDIA CMP 30HX is built on the Turing architecture, using the TU116 chip manufactured on TSMC’s 12 nm process. It packs 6,600 million transistors into a 284 mm² die, yielding a transistor density of 23.2 million per square millimeter. The AMD Radeon RX 6900 XT uses the RDNA 2.0 architecture with the Navi 21 chip on a 7 nm process. It contains 26,800 million transistors on a 520 mm² die, for a density of 51.5 million per square millimeter. The node advantage is clear: AMD fits 4 times the transistors on a die that is less than twice the size.

The shader configuration differs dramatically. The CMP 30HX has 1,408 shading units, 88 texture mapping units (TMUs), and 48 render output units (ROPs). The RX 6900 XT has 5,120 shading units, 320 TMUs, and 128 ROPs. The AMD card also includes 80 ray tracing cores, a feature entirely absent from the CMP 30HX, which has no RT cores and no tensor cores. This is a direct result of their intended markets: the CMP 30HX is a mining card with no display outputs, while the RX 6900 XT is a full-featured gaming GPU with 1x HDMI 2.1, 2x DisplayPort 1.4a, and 1x USB Type-C outputs.

Memory subsystems also diverge. The CMP 30HX offers 6 GB of GDDR6 on a 192-bit bus, providing 336.0 GB/s of bandwidth. The RX 6900 XT doubles the capacity to 16 GB on a 256-bit bus, with 512.0 GB/s of bandwidth. Clock speeds favor AMD as well: the RX 6900 XT’s base clock is 1825 MHz with a boost of 2250 MHz and a game clock of 2015 MHz, versus the CMP 30HX’s 1530 MHz base and 1785 MHz boost. The memory clock is also higher on the AMD card at 2000 MHz (16 Gbps effective) versus 1750 MHz (14 Gbps effective).

FAQ

Q: Which GPU is faster in compute workloads?

A: The AMD Radeon RX 6900 XT is significantly faster. In Geekbench OpenCL, it scores 187,673 versus 65,199 for the CMP 30HX, a -65.3% delta in favor of AMD.

Q: Does the CMP 30HX have any ray tracing capability?

A: No. The CMP 30HX has no ray tracing cores, while the RX 6900 XT includes 80 RT cores. The CMP 30HX is designed purely for mining, not gaming or ray-traced workloads.

Q: What is the memory size difference?

A: The RX 6900 XT has 16 GB of GDDR6, while the CMP 30HX has 6 GB. The AMD card also has a wider 256-bit bus compared to the 192-bit bus on the NVIDIA card.

Q: Which GPU has a higher average benchmark score?

A: The CMP 30HX has a higher average score of 63,842, placing it at the 89th percentile. The RX 6900 XT has an average of 50,951 at the 86th percentile, but this is skewed by its low Passmark DirectX scores.

Q: Can the CMP 30HX be used for display output?

A: No, the CMP 30HX has no display outputs. The RX 6900 XT has 1x HDMI 2.1, 2x DisplayPort 1.4a, and 1x USB Type-C.

Q: What is the process node for each GPU?

A: The CMP 30HX uses a 12 nm process at TSMC, while the RX 6900 XT uses a 7 nm process at the same foundry.

Specification Differences

The following fields differ between the two GPUs:

• Architecture: Turing (NVIDIA) vs. RDNA 2.0 (AMD)
• Process Node: 12 nm vs. 7 nm
• Transistors: 6,600 million vs. 26,800 million
• Die Size: 284 mm² vs. 520 mm²
• Transistor Density: 23.2M / mm² vs. 51.5M / mm²
• Base Clock: 1530 MHz vs. 1825 MHz
• Boost Clock: 1785 MHz vs. 2250 MHz
• Game Clock: None vs. 2015 MHz
• Memory Clock: 1750 MHz (14 Gbps effective) vs. 2000 MHz (16 Gbps effective)
• Memory Size: 6 GB vs. 16 GB
• Memory Bus Width: 192 bit vs. 256 bit
• Memory Bandwidth: 336.0 GB/s vs. 512.0 GB/s
• Shading Units: 1,408 vs. 5,120
• TMUs: 88 vs. 320
• ROPs: 48 vs. 128
• RT Cores: None vs. 80
• Pixel Rate: 85.68 GPixel/s vs. 288.0 GPixel/s
• Texture Rate: 157.1 GTexel/s vs. 720.0 GTexel/s
• FP32: 5.027 TFLOPS vs. 23.04 TFLOPS
• FP16: 10.05 TFLOPS vs. 46.08 TFLOPS
• TDP: 125 W vs. 300 W
• Slot Width: Dual-slot vs. Triple-slot
• Power Connectors: 1x 8-pin vs. 2x 8-pin
• Suggested PSU: 300 W vs. 700 W
• Bus Interface: PCIe 1.0 x4 vs. PCIe 4.0 x16
• Display Outputs: No outputs vs. 1x HDMI 2.1, 2x DisplayPort 1.4a, 1x USB Type-C
• DirectX Support: 12 (12_1) vs. 12 Ultimate (12_2)
• Dimensions: 229 mm x 111 mm x 35 mm vs. 267 mm x 120 mm x 50 mm
• Release Date: 2021-02-24 vs. 2020-10-27
• Launch MSRP: 799 USD vs. 999 USD

The Verdict

The data is unambiguous: the AMD Radeon RX 6900 XT is the superior GPU in raw performance. It wins both head-to-head benchmarks by wide margins, has 4 times the shading units, 3.6 times the memory bandwidth, and includes ray tracing cores that the CMP 30HX lacks entirely. The RX 6900 XT’s FP32 compute of 23.04 TFLOPS is 4.6 times the CMP 30HX’s 5.027 TFLOPS. For anyone needing compute performance, display output, or modern API support, the RX 6900 XT is the only choice.

However, the CMP 30HX is not without its own logic. It has a higher average benchmark score (63,842 vs. 50,951) and a higher percentile ranking (89th vs. 86th), driven by consistent performance across its limited test set. Its TDP is 125 W versus 300 W, and it requires a 300 W PSU versus 700 W, making it far more power-efficient for its intended mining workload. The CMP 30HX also has a lower launch MSRP of 799 USD versus 999 USD. But these advantages are irrelevant for gaming or general-purpose use, as the card cannot output video and lacks the architectural features needed for modern titles.

Where Each One Wins

The AMD Radeon RX 6900 XT wins in every performance metric that matters for end users: compute, memory, clocks, and API support. It is the clear winner for gaming, content creation, and any workload requiring DirectX 12 Ultimate features. Its 16 GB memory and 512.0 GB/s bandwidth handle large textures and high resolutions, and its 80 RT cores enable ray-traced effects. The Passmark G3D score of 26,732 confirms its strength in rasterized 3D rendering.

The NVIDIA CMP 30HX wins only in niche, mining-specific scenarios. Its lower TDP of 125 W and 1x 8-pin connector make it easier to power in dense mining rigs. Its PCIe 1.0 x4 interface is sufficient for mining, which does not rely on high-bandwidth host communication. The higher average score and percentile, while counterintuitive, reflect a more stable benchmark profile across its two recorded tests. For a pure mining operation where display output and gaming features are irrelevant, the CMP 30HX offers a lower power draw and lower launch price. For everything else, the RX 6900 XT is the definitive choice.