NVIDIA CMP 30HX vs NVIDIA GeForce RTX 5090 Mobile Comparison

The NVIDIA CMP 30HX and the NVIDIA GeForce RTX 5090 Mobile represent two very different design philosophies from the same manufacturer. One is a dedicated mining product built on older Turing architecture, while the other is a modern laptop GPU based on Blackwell 2.0. The database records show a dramatic shift in performance capabilities between these two generations, with the RTX 5090 Mobile dominating in every shared benchmark test.

FAQ

Q: How does the RTX 5090 Mobile compare to the CMP 30HX in OpenCL performance?

A: The RTX 5090 Mobile scores 201,834 in Geekbench OpenCL, while the CMP 30HX scores 65,199. This gives the mobile GPU a delta of -67.7% from the perspective of the CMP 30HX, meaning the RTX 5090 Mobile is roughly three times faster in this test.

Q: What is the average benchmark score difference between the two cards?

A: The CMP 30HX has an average benchmark score of 63,842 across its two recorded tests. The RTX 5090 Mobile averages 45,152 across its full suite of ten tests, though this figure is pulled down by several low-scoring legacy DirectX tests.

Q: Which card has more memory and what type does it use?

A: The RTX 5090 Mobile has 24 GB of GDDR7 memory on a 256-bit bus, while the CMP 30HX has 6 GB of GDDR6 on a 192-bit bus. Memory bandwidth jumps from 336.0 GB/s on the CMP 30HX to 896.0 GB/s on the RTX 5090 Mobile.

Q: What is the process node difference between these GPUs?

A: The CMP 30HX uses TSMC's 12 nm process with 6,600 million transistors on a 284 mm² die. The RTX 5090 Mobile uses a 5 nm process with 45,600 million transistors on a 378 mm² die, resulting in a transistor density of 120.6M per mm² versus 23.2M per mm².

Q: Which card has ray tracing and tensor cores?

A: Only the RTX 5090 Mobile includes these features, with 82 ray tracing cores and 328 tensor cores. The CMP 30HX lists no RT cores and no tensor cores, reflecting its mining-focused design.

Q: What are the power requirements for each card?

A: The CMP 30HX has a TDP of 125 W, requires a 1x 8-pin power connector, and suggests a 300 W PSU. The RTX 5090 Mobile has a lower TDP of 95 W, uses no external power connectors, and is designed as an IGP (integrated graphics processor) for laptops.

Architecture Differences

The architectural gap between these two GPUs is substantial. The CMP 30HX is built on the Turing architecture (TU116 chip) using a 12 nm process at TSMC. It packs 6,600 million transistors into a 284 mm² die, yielding a density of 23.2M transistors per mm². The RTX 5090 Mobile, by contrast, uses the Blackwell 2.0 architecture (GB203 chip) on a 5 nm process, fitting 45,600 million transistors into a larger 378 mm² die. This results in a density of 120.6M per mm², which is over five times higher.

The compute resources tell a similar story. The CMP 30HX has 1,408 shading units, 88 texture mapping units, and 48 ROPs. The RTX 5090 Mobile scales this up dramatically with 10,496 shading units, 328 TMUs, and 112 ROPs. The mobile GPU also adds dedicated ray tracing cores (82) and tensor cores (328), neither of which appears in the CMP 30HX specifications. This directly reflects the intended use case: the CMP 30HX was designed for mining workloads without display outputs, while the RTX 5090 Mobile targets full graphics workloads including ray tracing and AI acceleration.

Feature support differs at the API level as well. The CMP 30HX supports DirectX 12 (12_1), while the RTX 5090 Mobile supports DirectX 12 Ultimate (12_2). Both support OpenGL 4.6 and Vulkan 1.4. The bus interface also differs significantly: the CMP 30HX uses PCIe 1.0 x4 (which severely limits data transfer), while the RTX 5090 Mobile uses PCIe 5.0 x16.

The memory subsystem reflects the generational leap. The CMP 30HX uses 6 GB of GDDR6 at 14 Gbps effective on a 192-bit bus, providing 336.0 GB/s of bandwidth. The RTX 5090 Mobile uses 24 GB of GDDR7 at 28 Gbps effective on a 256-bit bus, delivering 896.0 GB/s. This nearly triples the bandwidth while quadrupling capacity.

Clock behavior is noteworthy. The CMP 30HX has a base clock of 1530 MHz and a boost clock of 1785 MHz, which are higher than the RTX 5090 Mobile's 990 MHz base and 1515 MHz boost. However, the RTX 5090 Mobile compensates with far more cores and higher efficiency per clock.

Head-to-Head Benchmarks

The database includes two direct head-to-head comparisons between these GPUs, and both show a decisive victory for the RTX 5090 Mobile.

In Geekbench OpenCL, the CMP 30HX scores 65,199 while the RTX 5090 Mobile scores 201,834. The delta of -67.7% means the CMP 30HX performs at less than one-third the level of the mobile GPU in this compute-oriented test. This is a massive gap that reflects not just more cores but also the memory bandwidth advantage of 896.0 GB/s versus 336.0 GB/s.

Geekbench Vulkan shows a similar pattern. The CMP 30HX scores 62,484, while the RTX 5090 Mobile reaches 198,405. The delta here is -68.5%, indicating an even larger relative deficit for the older card. The Vulkan test benefits from the RTX 5090 Mobile's modern architecture, its ray tracing and tensor core hardware, and its superior memory subsystem.

In terms of win counts, the RTX 5090 Mobile wins 2 out of 2 head-to-head tests, while the CMP 30HX records no wins. The individual benchmark suite for the RTX 5090 Mobile includes additional tests that the CMP 30HX does not share, such as 3DMark Steel Nomad (5,871), Passmark G3D (30,034), and Passmark GPU Compute (13,401). The CMP 30HX's higher percentile ranking of 89 versus the RTX 5090 Mobile's 84 percentile across all GPUs is somewhat misleading in isolation; the CMP 30HX benefits from having only two recorded benchmarks, both of which are compute-focused.

Specification Differences

The two cards differ in nearly every specification field recorded in the database. The CMP 30HX has a base clock of 1530 MHz and boost of 1785 MHz, while the RTX 5090 Mobile runs at 990 MHz base and 1515 MHz boost. The CMP 30HX has a memory clock of 1750 MHz (14 Gbps effective), while the RTX 5090 Mobile also has a 1750 MHz memory clock but with 28 Gbps effective speed.

Shading units: 1,408 versus 10,496. TMUs: 88 versus 328. ROPs: 48 versus 112. The RTX 5090 Mobile adds 82 RT cores and 328 tensor cores; the CMP 30HX has none. Pixel rate is 85.68 GPixel/s for the CMP 30HX versus 169.7 GPixel/s for the RTX 5090 Mobile. Texture rate is 157.1 GTexel/s versus 496.9 GTexel/s. FP32 performance is 5.027 TFLOPS versus 31.80 TFLOPS. FP16 is 10.05 TFLOPS (2:1) versus 31.80 TFLOPS (1:1).

Memory specifications differ across the board: 6 GB GDDR6 versus 24 GB GDDR7, 192-bit versus 256-bit bus, 336.0 GB/s versus 896.0 GB/s bandwidth. TDP is 125 W versus 95 W. The CMP 30HX is dual-slot with a 1x 8-pin connector and 300 W suggested PSU; the RTX 5090 Mobile is IGP with no connectors and no suggested PSU. The CMP has no display outputs; the RTX 5090 Mobile is portable device dependent for outputs. The bus interface is PCIe 1.0 x4 versus PCIe 5.0 x16. DirectX support is 12 (12_1) versus 12 Ultimate (12_2). The CMP 30HX has dimensions of 229 mm length, 111 mm height and 35 mm width; the RTX 5090 Mobile has no dimensions listed. Production status is end-of-life versus active, and the release dates are 2021-02-24 versus 2025-03-26. The CMP 30HX has a launch MSRP of 799 USD; the RTX 5090 Mobile has no launch MSRP field.

The Verdict

The data clearly separates these two products for different buyers. The RTX 3 5090 Mobile is the only choice for anyone who needs modern graphics rendering, ray tracing, AI features, or high-resolution gaming. Its 24 GB of GDDR7 memory provides ample capacity, while 896.0 GB/s bandwidth supports heavy data flows. The 31.80 TFLOPS FP32 compute and 31.80 TFLOPS FP16 performance make it suitable for compute-intensive tasks. The 95 W TDP and IGP design suit mobile deployment.

The CMP 30HX targets a niche use case: it has no display outputs, making it unsuitable for direct rendering output. Its 89th percentile rank in all GPUs is respectable among mining GPUs, and its 125 W TDP is moderate. The 2021 release date and end-of-life status mean this card is a legacy product, and its 799 USD launch MSRP reflects its position in the market position.

For gaming at high resolution with modern features, the RTX 5090 Mobile benefits from DirectX 12 Ultimate support, PCIe 5.0 x16 interface, 82 RT cores, and 328 tensor cores. For raw compute without display requirements, the RTX 5090 Mobile remains the better performer at 201,834 points in OpenCL versus 65,199 for the CMP 30HX.

The performance gap is decisive: 67.7% and 68.5% in the two head-to-head tests. The RTX 5090 Mobile is the definitive choice for nearly every workload.

Where Each One Wins

The RTX 5090 Mobile wins in all recorded benchmarks, including OpenCL and Vulkan. The CMP 30HX has no wins in the head-to-head benchmark suite. The RTX 5090 Mobile also shows strong performance in its own benchmark suite, including Passmark G3D at 30,034 points, Passmark GPU Compute at 13,401, and 5,871 in 3DMark Steel Nomad DX12.

Use cases where the RTX 5090 Mobile excels include scenarios, AI processing, and modern graphics APIs. The RTX 5090 Mobile's Blackwell 2.0 architecture supports ray tracing, and the 82 RT cores with 328 tensor cores. The 24 GB GDDR7 memory at 28 Gbps effective bandwidth supports large datasets, and the 256-bit bus provides high throughput for data transfer.

The CMP 30HX has a more limited role. It was designed for mining workloads with no display outputs, and its 6 GB GDDR6 memory on a 192-bit bus with 336.0 GB/s bandwidth served that purpose. The 1530 MHz base clock and 1785 MHz boost clock are higher than the RTX 5090 Mobile's clocks, though this does not translate into performance advantages in any recorded benchmark.

The RTX 5090 Mobile is the clear winner for anyone who needs a GPU for mobile devices, 24 GB memory capacity, GDDR7 technology, or PCIe 5.0 support. The CMP 30HX represents an earlier generation with different priorities: lower transistor density (23.2M versus 120.6M per mm²), older GDDR6 memory, and a mining-focused design. The RTX 5090 Mobile is also the only one of the two with ray tracing cores, tensor cores, and DirectX 12 Ultimate support, making it suitable for modern gaming applications. The CMP 30HX has no features to recommend it over the RTX 5090 Mobile in any benchmark category.