NVIDIA CMP 90HX vs NVIDIA GeForce RTX 4080 SUPER Comparison

FAQ

Q: How much faster is the RTX 4080 SUPER in the only direct benchmark comparison?

A: The RTX 4080 SUPER scores 219,065 in Geekbench OpenCL, while the CMP 90HX scores 69,000. That is a 68.5% higher score for the RTX 4080 SUPER, making it the decisive winner in the head-to-head data.

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

A: The CMP 90HX sits at the 90th percentile, while the RTX 4080 SUPER sits at the 86th percentile. Despite this, the RTX 4080 SUPER wins the benchmark comparison by a wide margin, which reflects how different benchmark suites and score distributions can shift percentile positions.

Q: What are the closest rivals to each card according to the database?

A: The CMP 90HX's nearest rival is the Intel Arc A770, with a delta of +0.3%. The RTX 4080 SUPER's closest rival is the NVIDIA GeForce RTX 4080, with a delta of -0.1%. These deltas indicate nearly identical average scores for each card versus its nearest competitor.

Q: Does either card have a display output?

A: The CMP 90HX has no display outputs at all, as it is a mining-specific product. The RTX 4080 SUPER has 1x HDMI 2.1 and 3x DisplayPort 1.4a outputs, making it suitable for standard desktop use.

Q: What is the memory configuration difference between the two cards?

A: The CMP 90HX has 10 GB of GDDR6X on a 320-bit bus, yielding 760.3 GB/s bandwidth. The RTX 4080 SUPER has 16 GB of GDDR6X on a 256-bit bus, yielding 736.3 GB/s bandwidth. So the CMP 90HX has slightly higher bandwidth, but the RTX 4080 SUPER has 60% more capacity.

Q: Are both cards considered end-of-life products?

A: Yes, both are marked as end-of-life in the database. The CMP 90HX was released in July 2021, while the RTX 4080 SUPER was released in January 2024.

Where Each One Wins

The data shows a single head-to-head benchmark, and the RTX 4080 SUPER wins it outright. In Geekbench OpenCL, the RTX 4080 SUPER delivers 219,065 points versus 69,000 for the CMP 90HX, a 68.5% advantage. That is not a close contest; it is a generational leap in raw compute performance.

However, the CMP 90HX has its own niche. Its 90th percentile ranking versus the RTX 4080 SUPER's 86th percentile suggests that, within the broader database of all GPUs, the CMP 90HX sits higher relative to its peers. This is likely because the CMP 90HX was a specialized mining card with a focused workload profile, and its single benchmark score of 69,000 places it above many other products. Still, the RTX 4080 SUPER's absolute score is far higher.

For gaming, the RTX 4080 SUPER is the clear choice due to its display outputs, 16 GB memory, and modern architecture. For compute tasks that favor OpenCL, the RTX 4080 SUPER also dominates. The CMP 90HX, with no outputs and a mining-oriented design, has no practical use case in a standard PC build unless the workload specifically matches its profile. The benchmark results indicate that, in every measurable head-to-head metric, the RTX 4080 SUPER wins.

Architecture Differences

The two cards are built on entirely different architectures from different generations. The CMP 90HX uses the GA102 chip on the Ampere architecture, fabricated on Samsung's 8 nm process. It packs 28,300 million transistors on a 628 mm² die, giving a transistor density of 45.1 million per mm². The RTX 4080 SUPER uses the AD103 chip on the Ada Lovelace architecture, fabricated on TSMC's 5 nm process. It contains 45,900 million transistors on a 379 mm² die, resulting in a much higher density of 121.1 million per mm².

This process and density difference is fundamental. The RTX 4080 SUPER fits more transistors into a smaller die, which drives its higher clock speeds and efficiency. The CMP 90HX's larger, older die cannot match that density.

The CMP 90HX has 6,400 shading units, 200 TMUs, and 80 ROPs. It also has 50 RT cores and 200 tensor cores. The RTX 4080 SUPER has 10,240 shading units, 320 TMUs, and 112 ROPs, along with 80 RT cores and 320 tensor cores. Every compute unit count is higher on the RTX 4080 SUPER.

The CMP 90HX is a dual-slot card with no display outputs, while the RTX 4080 SUPER is a triple-slot card with a full set of display outputs. The CMP 90HX uses a PCIe 1.0 x4 interface, which is extremely limited for data transfer, while the RTX 4080 SUPER uses PCIe 4.0 x16. This makes the RTX 4080 SUPER vastly better suited for any workload that requires host communication.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical.

Specification Differences

The clock speeds differ significantly. The CMP 90HX has a base clock of 1500 MHz and a boost clock of 1710 MHz. The RTX 4080 SUPER has a base clock of 2295 MHz and a boost clock of 2550 MHz. Memory clocks also differ: the CMP 90HX runs at 1188 MHz (19 Gbps effective), while the RTX 4080 SUPER runs at 1438 MHz (23 Gbps effective).

Memory capacity favors the RTX 4080 SUPER: 16 GB versus 10 GB. The bus width favors the CMP 90HX: 320-bit versus 256-bit. Bandwidth is close, with the CMP 90HX at 760.3 GB/s and the RTX 4080 SUPER at 736.3 GB/s, a small edge for the older card.

Pixel rate and texture rate strongly favor the RTX 4080 SUPER. It achieves 285.6 GPixel/s and 816.0 GTexel/s, versus 136.8 GPixel/s and 342.0 GTexel/s for the CMP 90HX. FP32 performance is 52.22 TFLOPS for the RTX 4080 SUPER versus 21.89 TFLOPS for the CMP 90HX, and FP16 is identical at 1:1 ratios for both.

Both cards have a TDP of 320 W and a suggested PSU of 700 W. The CMP 90HX uses two 8-pin connectors, while the RTX 4080 SUPER uses a single 16-pin connector. Dimensions differ: the CMP 90HX is 285 mm long and 112 mm tall, while the RTX 4080 SUPER is 310 mm long, 140 mm tall, and 61 mm wide.

The CMP 90HX has no launch MSRP listed. The RTX 4080 SUPER had a launch MSRP of 999 USD.

Head-to-Head Benchmarks

The database contains one direct head-to-head comparison: Geekbench OpenCL. The RTX 4080 SUPER scores 219,065, while the CMP 90HX scores 69,000. The delta is -68.5% from the perspective of the CMP 90HX, meaning it trails by over two-thirds. This is the largest single benchmark gap in the comparison.

To contextualize the CMP 90HX's score, its nearest rivals are the Intel Arc A770 at 68,809 (delta +0.3%), the AMD Radeon Instinct MI25 at 68,562 (delta +0.6%), the AMD Radeon Pro WX 8200 at 69,870 (delta -1.2%), and the NVIDIA Quadro P6000 at 69,986 (delta -1.4%). The CMP 90HX is essentially tied with these cards, all within a 1.4% range. Its 69,000 score is not an outlier; it sits squarely in that performance band.

For the RTX 4080 SUPER, its nearest rivals are the NVIDIA GeForce RTX 4080 at 54,247 (delta -0.1%), the AMD Radeon Pro W5700X at 54,828 (delta -1.1%), the AMD Radeon RX 6750 GRE 12 GB at 55,698 (delta -2.7%), and the AMD Radeon 8060S at 55,757 (delta -2.8%). These deltas are all negative, meaning the RTX 4080 SUPER scores slightly lower than those rivals in the average benchmark metric, yet its OpenCL score is far higher than its own average. This discrepancy suggests the RTX 4080 SUPER performs exceptionally well in OpenCL specifically, while its average across all tests is lower.

The RTX 4080 SUPER also has additional benchmark data: 3DMark Steel Nomad DX12 at 6,600, Geekbench Vulkan at 260,075, Passmark DirectX 10 at 193, DirectX 11 at 301, DirectX 12 at 134, DirectX 9 at 381, G2D at 1,270, G3D at 34,245, and GPU Compute at 19,822. The CMP 90HX has no comparable data for those tests, so the RTX 4080 SUPER's versatility is evident from the breadth of its recorded results.

The Verdict

The data is unambiguous: the RTX 4080 SUPER wins the only head-to-head benchmark by 68.5%, and it does so across a much wider range of tests. Its 16 GB memory, 52.22 TFLOPS FP32 performance, and full display outputs make it a practical choice for gaming and general compute. The CMP 90HX, with its 10 GB memory, 21.89 TFLOPS, and no display outputs, is a mining-era artifact with no modern desktop use case.

Who should pick the CMP 90HX? Only someone with a specific workload that matches its exact profile, such as a dedicated mining operation that no longer needs outputs or modern features. Its 90th percentile ranking shows it is not a weak card in the broader database, but its single benchmark score of 69,000 is dwarfed by the RTX 4080 SUPER.

Who should pick the RTX 4080 SUPER? Anyone building a gaming PC or a workstation that needs strong OpenCL performance, high memory capacity, and standard connectivity. Its 86th percentile ranking is slightly lower than the CMP 90HX, but that is a statistical artifact of averaging across many benchmarks; the direct comparison shows it is far faster. The RTX 4080 SUPER is the practical recommendation from every angle the recorded data supports.