Intel Arc A530M vs NVIDIA CMP 90HX Comparison

Head-to-Head Benchmarks

The database contains a single direct head-to-head comparison between the NVIDIA CMP 90HX and the Intel Arc A530M, using the Geekbench OpenCL workload. This is a compute-oriented test that stresses raw shading throughput, memory bandwidth, and driver-level execution efficiency. The result is decisive: the NVIDIA CMP 90HX scores 69,000 points, while the Intel Arc A530M scores 49,735 points. The delta is 38.7 percent in favor of the NVIDIA part. This is not a marginal victory; it is a substantial gap that places the two products in entirely different performance tiers.

Context from the nearest rivals in the database reinforces this separation. The NVIDIA CMP 90HX sits at the 90th percentile of all GPUs tested. Its closest competitors are the Intel Arc A770 at 68,809 points (a 0.3 percent deficit for the CMP 90HX), the AMD Radeon Instinct MI25 at 68,562 points (0.6 percent behind), the AMD Radeon Pro WX 8200 at 69,870 points (1.2 percent ahead of the CMP 90HX), and the NVIDIA Quadro P6000 at 69,986 points (1.4 percent ahead). In other words, the CMP 90HX is effectively trading blows with flagship-class workstation and enthusiast cards from the previous generation. The Arc A530M, by contrast, sits at the 85th percentile and its nearest rivals are the AMD Radeon RX 5600M at 46,601 points (0 percent delta), the AMD Radeon RX 6550M at 46,702 points (0.2 percent ahead of the Arc), the NVIDIA RTX A2000 at 46,043 points (1.2 percent behind the Arc), and the NVIDIA RTX 5880 Ada Generation at 45,972 points (1.4 percent behind the Arc). The Arc A530M is competitive within its own mobile-class peer group, but that peer group is far below the performance class occupied by the CMP 90HX.

The raw score difference translates into practical implications. In OpenCL workloads, the CMP 90HX delivers roughly 39 percent more compute throughput than the Arc A530M. That is the kind of margin that changes workload feasibility: tasks that take one hour on the Arc would take roughly 43 minutes on the CMP, assuming linear scaling. Conversely, the Arc A530M has no benchmark win in the recorded data. The head-to-head tally is 1 win for the CMP 90HX and 0 for the Arc A530M. There is no Vulkan result recorded for the CMP 90HX in the database, so cross-API comparisons cannot be made with measured data. The Arc A530M does have a Geekbench Vulkan score of 43,492, but without an equivalent CMP 90HX Vulkan result, that number stands alone.

The Verdict

The data points to a straightforward recommendation for most users. If the workload is OpenCL-heavy and maximum compute throughput is the priority, the NVIDIA CMP 90HX is the clear choice. Its 69,000-point score places it in the 90th percentile of all GPUs, and it outperforms the Arc A530M by 38.7 percent in the only shared benchmark. The CMP 90HX also has a substantial hardware resource advantage: 6,400 shading units versus 1,536, 200 texture mapping units versus 96, 80 ROPs versus 48, and 760.3 GB/s of memory bandwidth versus 224.0 GB/s. These are not subtle differences; they are structural advantages that manifest directly in compute-heavy tasks.

However, the CMP 90HX is not a general-purpose graphics card. It has no display outputs at all, meaning it cannot drive a monitor. It is also an end-of-life product, whereas the Arc A530M is still in active production. The Arc A530M, despite its lower compute score, is a portable device dependent GPU with display outputs that depend on the host system. It draws 65 W compared to the CMP 90HX's 320 W, and it uses an IGP slot width, which means it is designed for integration into laptops or compact systems. The CMP 90HX is a dual-slot card requiring 2x 8-pin power connectors and a 700 W suggested power supply.

So the verdict is not simply "the CMP 90HX is better." It is: the CMP 90HX is better at raw compute, and that is the only dimension measured in the shared benchmark. The Arc A530M is the only one of the two that can be used in a portable system with display output, and it is the only one that is still an active product. Users who need compute density in a mining or server context and have no need for display output should choose the CMP 90HX. Users who need a mobile GPU with modest compute performance and active production status should choose the Arc A530M. Neither product is a substitute for the other.

Where Each One Wins

The NVIDIA CMP 90HX wins the only recorded head-to-head benchmark, and it wins by a wide margin. Its 69,000 OpenCL score is 38.7 percent higher than the Arc A530M's 49,735. The CMP 90HX also wins on every major hardware specification that affects compute performance: shading units (6,400 versus 1,536), texture units (200 versus 96), ROPs (80 versus 48), ray tracing cores (50 versus 12), FP32 throughput (21.89 TFLOPS versus 3.994 TFLOPS), texture rate (342.0 GTexel/s versus 124.8 GTexel/s), pixel rate (136.8 GPixel/s versus 62.40 GPixel/s), and memory bandwidth (760.3 GB/s versus 224.0 GB/s). The CMP 90HX also has more memory capacity at 10 GB versus 8 GB, and it uses faster GDDR6X memory compared to the Arc's GDDR6. Its boost clock is 1710 MHz versus 1300 MHz, and its base clock is 1500 MHz versus 900 MHz.

The Intel Arc A530M wins on efficiency and form factor, though these are not directly benchmarked. Its 65 W TDP is far lower than the CMP 90HX's 320 W. It uses a 6 nm process from TSMC, while the CMP 90HX uses an 8 nm process from Samsung. The Arc A530M is an IGP-class part with no separate slot width, meaning it is designed for mobile integration. It has display outputs that are portable device dependent, which means it can drive a screen in a laptop or compact system. The CMP 90HX has no display outputs at all. The Arc A530M also has FP16 throughput of 7.987 TFLOPS with a 2:1 ratio, which is more than double its FP32 rate, whereas the CMP 90HX has 1:1 FP16 and FP32 at 21.89 TFLOPS each. For workloads that can use FP16 acceleration, the Arc's architecture may be more efficient per watt, but no direct FP16 benchmark is recorded.

The Arc A530M also has a Vulkan score of 43,492 in the database, which indicates it can handle Vulkan workloads at a moderate level. The CMP 90HX has no recorded Vulkan score, so no comparison is possible there. The Arc A530M is an active product with a release date in 2023, while the CMP 90HX is end-of-life with a release date in 2021. For anyone concerned about long-term availability and driver support for current software, the Arc A530M is the safer choice.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The NVIDIA CMP 90HX scores 69,000 in Geekbench OpenCL, while the Intel Arc A530M scores 49,735. The CMP 90HX leads by 38.7 percent.

Q: Can the NVIDIA CMP 90HX be used to drive a display?

A: No. The CMP 90HX has no display outputs. The Intel Arc A530M has display outputs that are portable device dependent.

Q: How do these GPUs compare in memory bandwidth?

A: The NVIDIA CMP 90HX has 760.3 GB/s of bandwidth with 10 GB of GDDR6X on a 320-bit bus. The Intel Arc A530M has 224.0 GB/s with 8 GB of GDDR6 on a 128-bit bus.

Q: Which product is still in production?

A: The Intel Arc A530M is listed as active in production. The NVIDIA CMP 90HX is end-of-life.

Q: What is the power consumption difference?

A: The NVIDIA CMP 90HX has a TDP of 320 W and requires a 700 W suggested power supply with 2x 8-pin connectors. The Intel Arc A530M has a TDP of 65 W and is an IGP-class part with no separate power connector listing.

Q: Does the Intel Arc A530M have any benchmark result that the CMP 90HX lacks?

A: Yes. The Arc A530M has a Geekbench Vulkan score of 43,492. The CMP 90HX has no recorded Vulkan benchmark in the database.

Architecture Differences

The two GPUs come from different manufacturers, different foundries, and different architectural generations. The NVIDIA CMP 90HX uses the GA102 chip built on Ampere architecture, fabricated by Samsung on an 8 nm process. It contains 28,300 million transistors on a 628 mm² die, yielding a transistor density of 45.1 million transistors per square millimeter. The Intel Arc A530M uses the DG2-256 chip built on Xe-HPG architecture, fabricated by TSMC on a 6 nm process. It contains 11,500 million transistors on a 269 mm² die, yielding a transistor density of 42.8 million transistors per square millimeter. The CMP 90HX is the larger chip by die area and transistor count, but the Arc A530M uses a slightly more advanced process node.

The compute resources differ sharply. The CMP 90HX has 6,400 shading units, 200 texture mapping units, 80 ROPs, 50 ray tracing cores, and 200 tensor cores. The Arc A530M has 1,536 shading units, 96 texture mapping units, 48 ROPs, and 12 ray tracing cores, with no tensor core count listed. The FP32 throughput of the CMP 90HX is 21.89 TFLOPS, while the Arc A530M manages 3.994 TFLOPS. The FP16 numbers are interesting: the CMP 90HX achieves 21.89 TFLOPS with a 1:1 ratio, meaning FP16 and FP32 are equal. The Arc A530M achieves 7.987 TFLOPS with a 2:1 ratio, meaning FP16 is double its FP32 rate. For mixed-precision workloads, the Arc's FP16 advantage relative to its own FP32 is notable, but in absolute terms the CMP 90HX still has far higher FP16 throughput.

Memory architecture is another major divider. The CMP 90HX uses 10 GB of GDDR6X on a 320-bit bus with 760.3 GB/s of bandwidth and an effective memory speed of 19 Gbps. The Arc A530M uses 8 GB of GDDR6 on a 128-bit bus with 224.0 GB/s of bandwidth and an effective speed of 14 Gbps. The CMP 90HX has more than three times the memory bandwidth, which is critical for bandwidth-sensitive compute workloads.

Clock speeds also favor the CMP 1500 MHz base clock. The base clocks are 1500 MHz and 1710 MHz boost for the CMP 90HX, versus 900 MHz base clock and 1300 MHz boost for the Arc A530M. The Arc A530M also has a 112 MHz height of 112 mm and a 285 mm length, with width not listed. The CMP 90HX is 285 mm long, 112 mm high, and dual-slot. The Arc A530M has no listed dimensions and uses an IGP slot width.

The bus interfaces differ as well. The CMP 90HX uses PCIe 1.0 x4, which is an unusual and potentially limiting interface for a modern GPU. The Arc A530M uses PCIe 4.0 x8, which is more appropriate for a current-generation part. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The CMP 90HX is from the Mining GPUs generation with a release date in 2021, while the Arc A530M is from the Alchemist (Arc 5 Mobile) generation with a release date in 2023. The production status confirms the generational split: the CMP 90HX is end-of-life, and the Arc A530M is active. These architectural differences explain the benchmark gap. The CMP 90HX was built for maximum throughput in mining and compute tasks, with no display hardware, a wide memory bus, and enormous shader counts. The Arc A530M was built for mobile efficiency, with a smaller chip, lower power envelope, and integrated design philosophy.