Intel Arc Pro A60 vs NVIDIA CMP 40HX Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA CMP 40HX records an average benchmark score of 85,637, placing it in the 93rd percentile of all GPUs. The Intel Arc Pro A60 averages 60,326, which puts it in the 88th percentile.

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

A: In Geekbench OpenCL, the NVIDIA CMP 40HX scores 93,395 versus 63,485 for the Intel Arc Pro A60. This gives NVIDIA a 47.1% lead in that workload.

Q: What is the difference in memory capacity and bandwidth?

A: The Intel Arc Pro A60 has 12 GB of GDDR6 memory on a 192-bit bus, delivering 384.0 GB/s bandwidth. The NVIDIA CMP 40HX has 8 GB of GDDR6 on a 256-bit bus, with higher bandwidth at 448.0 GB/s.

Q: Which card has a higher boost clock?

A: The Intel Arc Pro A60 boosts to 2050 MHz, while the NVIDIA CMP 40HX boosts to 1650 MHz. The Arc Pro A60 also has a lower base clock at 900 MHz versus 1470 MHz for the CMP 40HX.

Q: What are the power requirements for each card?

A: The NVIDIA CMP 40HX has a TDP of 185 W and requires a 450 W power supply. The Intel Arc Pro A60 has a lower TDP of 130 W and needs only a 300 W power supply.

Q: Do both cards support the same graphics APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. However, the NVIDIA card has no display outputs, while the Intel card offers 4x DisplayPort 2.0.

Where Each One Wins

The recorded data splits cleanly between the two cards across the two benchmark tests. The NVIDIA CMP 40HX wins both head-to-head tests: Geekbench OpenCL and Geekbench Vulkan. That gives it a 2-0 advantage in the direct comparison.

The Intel Arc Pro A60 does not win any of the recorded benchmarks against the CMP 40HX. However, its wins appear in other areas of the specification sheet. It has a larger memory pool, 12 GB versus 8 GB, which matters for workloads that need to hold larger datasets locally. It also has a higher boost clock, 2050 MHz versus 1650 MHz, and a higher pixel rate at 131.2 GPixel/s versus 105.6 GPixel/s. Its texture rate of 262.4 GTexel/s also exceeds the NVIDIA card's 237.6 GTexel/s.

For raw compute throughput, the Intel card reports higher FP32 performance at 8.397 TFLOPS versus 7.603 TFLOPS for the NVIDIA card. The FP16 numbers follow the same pattern: 16.79 TFLOPS for Intel versus 15.21 TFLOPS for NVIDIA.

The NVIDIA card wins on memory bandwidth (448.0 GB/s versus 384.0 GB/s), transistor count (10,800 million versus 11,500 million, though Intel has more), and the actual benchmark scores. The CMP 40HX also holds a higher percentile ranking at 93 versus 88.

Architecture Differences

The NVIDIA CMP 40HX uses the TU106 chip based on the Turing architecture, fabricated on a 12 nm process at TSMC. It contains 10,800 million transistors on a die size of 445 mm², giving a transistor density of 24.3M per mm².

The Intel Arc Pro A60 uses the DG2-256 chip based on the Xe-HPG architecture, built on a 6 nm process, also at TSMC. It packs 11,500 million transistors into a smaller die of 269 mm², achieving a much higher transistor density of 42.8M per mm². This means Intel packs roughly 6.5% more transistors into a die that is about 40% smaller.

The NVIDIA card has 2304 shading units, 144 texture mapping units, and 64 ROPs. It includes 36 ray tracing cores and 288 tensor cores. The Intel card has 2048 shading units, 128 TMUs, and 64 ROPs, with 16 ray tracing cores and no tensor cores listed.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The NVIDIA card supports a PCIe 1.0 x4 bus interface, which is unusual, while the Intel card uses PCIe 4.0 x16.

The NVIDIA CMP 40HX has no display outputs, reflecting its mining-oriented design. The Intel Arc Pro A60 provides 4x DisplayPort 2.0 outputs, making it suitable for professional display use.

Specification Differences

The two cards differ across nearly every major specification field. Starting with the chip: NVIDIA uses TU106 (Turing architecture), Intel uses DG2-256 (Xe-HPG architecture). Process nodes are 12 nm for NVIDIA and 6 nm for Intel. Die sizes are 445 mm² versus 269 mm², with corresponding transistor densities of 24.3M per mm² versus 42.8M per mm².

Clock speeds differ significantly. The NVIDIA card has a base clock of 1470 MHz and a boost of 1650 MHz. The Intel card has a base of 900 MHz and a boost of 2050 MHz. Memory clocks also differ: 1750 MHz (14 Gbps effective) for NVIDIA versus 2000 MHz (16 Gbps effective) for Intel.

Memory configuration: NVIDIA has 8 GB GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth. Intel has 12 GB GDDR6 on a 192-bit bus with 384.0 GB/s bandwidth.

Compute resources: NVIDIA has 2304 shading units, 144 TMUs, 64 ROPs, 36 RT cores, and 288 tensor cores. Intel has 2048 shading units, 128 TMUs, 64 ROPs, 16 RT cores, and no tensor cores listed.

Performance rates: NVIDIA delivers 105.6 GPixel/s pixel rate and 237.6 GTexel/s texture rate, with 7.603 TFLOPS FP32 and 15.21 TFLOPS FP16. Intel delivers 131.2 GPixel/s pixel rate and 262.4 GTexel/s texture rate, with 8.397 TFLOPS FP32 and 16.79 TFLOPS FP16.

Power and physical: NVIDIA has a 185 W TDP, dual-slot width, 1x 8-pin power connector, and requires a 450 W PSU. Intel has a 130 W TDP, single-slot width, no power connector listed, and requires a 300 W PSU. The NVIDIA card measures 229 mm in length, 111 mm in height, and 35 mm in width. Intel lists no dimensions.

Bus interface: NVIDIA uses PCIe 1.0 x4, Intel uses PCIe 4.0 x16. Display outputs: none for NVIDIA, 4x DisplayPort 2.0 for Intel.

Production status: NVIDIA is end-of-life, released on 2021-02-24 with a launch MSRP of 699 USD. Intel is active, released on 2023-06-05 with no launch MSRP listed.

Head-to-Head Benchmarks

The head-to-head results are decisive in favor of the NVIDIA CMP 40HX. In Geekbench OpenCL, the NVIDIA card scores 93,395 against 63,485 for the Intel Arc Pro A60. This is a 47.1% advantage, the largest margin in either test.

In Geekbench Vulkan, the NVIDIA card scores 77,879 against 57,166 for Intel, a 36.2% lead. Both tests show the NVIDIA card delivering substantially higher compute throughput in these specific workloads.

The average benchmark scores reinforce this pattern. The NVIDIA card averages 85,637, which is 42% higher than the Intel card's 60,326 average. The percentile rankings reflect this gap: 93rd percentile for NVIDIA versus 88th percentile for Intel.

When compared to its nearest rivals, the NVIDIA card sits close to the AMD Radeon PRO W7600 (87,108 average, 1.7% higher) and the NVIDIA Quadro GP100 (87,445 average, 2.1% higher). It is ahead of the AMD Radeon PRO W6600 by 4.4% and the AMD Radeon Pro Vega 64X by 5.8%.

The Intel Arc Pro A60's nearest rival list shows a different competitive landscape. It matches the NVIDIA GeForce RTX 4090 (60,347 average, 0% difference) and the AMD Radeon Pro Vega 48 (60,140 average, 0.3% lower). It trails the AMD Radeon Pro W6600M by 2.5% and beats the AMD Radeon PRO V710 by 2.8%.

The data shows the NVIDIA card competing in a higher performance tier, while the Intel card sits in a lower tier despite its more modern architecture and higher boost clock.

The Verdict

The benchmark data points to a clear performance hierarchy. The NVIDIA CMP 40HX wins both recorded benchmarks by substantial margins: 47.1% in OpenCL and 36.2% in Vulkan. Its average benchmark score of 85,637 places it 42% above the Intel Arc Pro A60's 60,326. Anyone prioritizing raw compute throughput in these workloads should choose the NVIDIA card.

The Intel Arc Pro A60, however, offers advantages that the benchmarks do not capture. It has 12 GB of memory versus 8 GB, a higher boost clock (2050 MHz versus 1650 MHz), and higher pixel and texture rates. It also draws less power (130 W TDP versus 185 W) and comes in a single-slot form factor. For workloads that depend on memory capacity rather than raw bandwidth, or that benefit from higher clock speeds, the Intel card may be the better fit.

The architecture differences explain part of the performance gap. The NVIDIA card uses a larger, older 12 nm process with more shading units and tensor cores. The Intel card uses a denser 6 nm process with fewer shading units but higher clock speeds. The result is that Intel achieves higher theoretical FP32 throughput (8.397 TFLOPS versus 7.603 TFLOPS), yet the NVIDIA card still wins the actual benchmark tests.

The production status also matters. The NVIDIA CMP 40HX is end-of-life, released in 2021 with a launch MSRP of 699 USD. The Intel Arc Pro A60 is active, released in 2023. For new system builds, the Intel card offers current availability and display outputs, while the NVIDIA card has no display outputs at all.

The verdict depends on the use case. For compute benchmarks and high-bandwidth memory access, the NVIDIA CMP 40HX is the stronger choice. For professional workstation use requiring display outputs, lower power draw, and a larger memory pool, the Intel Arc Pro A60 presents a more balanced package despite its lower benchmark scores.