Intel Arc Pro A60 vs NVIDIA GeForce RTX 5080 Comparison

The Intel Arc Pro A60 and NVIDIA GeForce RTX 5080 occupy opposite ends of the GPU spectrum: one is a compact, low-power workstation card from Intel’s Alchemist generation, the other a flagship Blackwell consumer GPU with a launch MSRP of 999 USD. Despite both sitting at the 89th percentile of all GPUs, their benchmark profiles diverge sharply. The A60 posts a higher average benchmark score (60326) than the RTX 5080 (59188), yet in the two shared tests, the RTX 5080 wins decisively. This page breaks down the architecture, specifications, and benchmark results to show where each card stands.

FAQ

Q: Which card has more memory and bandwidth?

A: The RTX 5080 has 16 GB of GDDR7 on a 256-bit bus, delivering 960.0 GB/s. The Arc Pro A60 has 12 GB of GDDR6 on a 192-bit bus, delivering 384.0 GB/s.

Q: How do the shading units compare?

A: The RTX 5080 has 10752 shading units, 336 TMUs, and 112 ROPs. The Arc Pro A60 has 2048 shading units, 128 TMUs, and 64 ROPs.

Q: Which card has ray tracing and tensor cores?

A: The RTX 5080 has 84 RT cores and 336 tensor cores. The Arc Pro A60 has 16 RT cores and no tensor cores.

Q: What are the power requirements?

A: The Arc Pro A60 has a 130 W TDP and a suggested PSU of 300 W, with no external power connector. The RTX 5080 has a 360 W TDP, a 750 W suggested PSU, and a single 16-pin connector.

Q: Which card supports PCIe 5.0?

A: The RTX 5080 uses PCIe 5.0 x16, while the Arc Pro A60 uses PCIe 4.0 x16.

Q: How do their display outputs differ?

A: The Arc Pro A60 offers 4x DisplayPort 2.0. The RTX 5080 provides 1x HDMI 2.1b and 3x DisplayPort 2.1b.

Architecture Differences

The two cards are built on fundamentally different architectures and process nodes. The Arc Pro A60 uses Intel’s Xe-HPG architecture (generation Alchemist, Pro Series), fabricated on a 6 nm TSMC process. Its die measures 269 mm² and contains 11,500 million transistors, for a transistor density of 42.8M per mm². The RTX 5080 uses NVIDIA’s Blackwell 2.0 architecture (generation GeForce 50), on a 5 nm TSMC process. Its die is 378 mm² with 45,600 million transistors, giving a density of 120.6M per mm². The RTX 5080’s die is 40% larger in area but packs nearly four times the transistors, reflecting a much denser design.

The compute layout differs substantially. The Arc Pro A60 has 2048 shading units, 128 TMUs, and 64 ROPs, plus 16 RT cores. The RTX 5080 has 10752 shading units, 336 TMUs, 112 ROPs, 84 RT cores, and 336 tensor cores. The RTX 5080 also has a much higher FP32 throughput at 56.28 TFLOPS, compared to the A60’s 8.397 TFLOPS. FP16 performance is also asymmetric: the RTX 5080 achieves 56.28 TFLOPS (1:1), while the A60 reaches 16.79 TFLOPS (2:1). Pixel and texture rates follow the same pattern—the RTX 5080 outputs 293.1 GPixel/s and 879.3 GTexel/s, versus the A60’s 131.2 GPixel/s and 262.4 GTexel/s.

Memory architecture is another major split. The A60 uses 12 GB of GDDR6 with a 192-bit bus and 384.0 GB/s bandwidth. The RTX 5080 uses 16 GB of GDDR7 on a 256-bit bus, doubling bandwidth to 960.0 GB/s. Clock speeds also favor the RTX 5080: base 2295 MHz and boost 2617 MHz, compared to the A60’s 900 MHz base and 2050 MHz boost. The RTX 5080’s memory runs at 1875 MHz (30 Gbps effective), while the A60’s memory runs at 2000 MHz (16 Gbps effective).

Power and physical design are opposite. The A60 is a single-slot card with a 130 W TDP and no power connector, requiring a 300 W PSU. The RTX 5080 is dual-slot, 304 mm long, 137 mm tall, and 40 mm wide, with a 360 W TDP, a 16-pin connector, and a 750 W PSU suggestion. The A60 has no listed dimensions, but its single-slot profile and lower power draw make it a different class of card.

The Verdict

The data positions the Arc Pro A60 as a low-power, compact workstation card with a strong average benchmark score, while the RTX 5080 is a high-performance, power-hungry consumer GPU. The A60’s average benchmark score (60326) is 1.9% higher than the RTX 5080’s (59188), and it sits in the same 89th percentile. But the head-to-head tests tell a different story: the RTX 5080 wins both, with a 76% lead in OpenCL and a 77.8% lead in Vulkan. If you need raw compute or gaming performance, the RTX 5080 is the clear choice. If you need a low-power, single-slot card with four DisplayPort 2.0 outputs and a lower PSU requirement, the Arc Pro A60 is the practical pick. The RTX 5080’s launch MSRP of 999 USD is the only price data available, but the A60 has no listed MSRP.

Specification Differences

| Field | Intel Arc Pro A60 | NVIDIA GeForce RTX 5080 |

|-------|-------------------|-------------------------|

| Chip | DG2-256 | GB203 |

| Architecture | Xe-HPG | Blackwell 2.0 |

| Generation | Alchemist (Pro Series) | GeForce 50 |

| Process Node | 6 nm | 5 nm |

| Transistors | 11,500 million | 45,600 million |

| Die Size | 269 mm² | 378 mm² |

| Transistor Density | 42.8M / mm² | 120.6M / mm² |

| Base Clock | 900 MHz | 2295 MHz |

| Boost Clock | 2050 MHz | 2617 MHz |

| Memory Clock | 2000 MHz / 16 Gbps effective | 1875 MHz / 30 Gbps effective |

| Memory Size | 12 GB | 16 GB |

| Memory Type | GDDR6 | GDDR7 |

| Memory Bus | 192 bit | 256 bit |

| Memory Bandwidth | 384.0 GB/s | 960.0 GB/s |

| Shading Units | 2048 | 10752 |

| TMUs | 128 | 336 |

| ROPs | 64 | 112 |

| RT Cores | 16 | 84 |

| Tensor Cores | null | 336 |

| Pixel Rate | 131.2 GPixel/s | 293.1 GPixel/s |

| Texture Rate | 262.4 GTexel/s | 879.3 GTexel/s |

| FP32 | 8.397 TFLOPS | 56.28 TFLOPS |

| FP16 | 16.79 TFLOPS (2:1) | 56.28 TFLOPS (1:1) |

| TDP | 130 W | 360 W |

| Slot Width | Single-slot | Dual-slot |

| Power Connectors | null | 1x 16-pin |

| Suggested PSU | 300 W | 750 W |

| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |

| Display Outputs | 4x DisplayPort 2.0 | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| Dimensions | null | 304 mm / 12 inches (length), 137 mm / 5.4 inches (height), 40 mm / 1.6 inches (width) |

| Release Date | 2023-06-05 | 2025-01-29 |

| Launch MSRP | null | 999 USD |

Head-to-Head Benchmarks

The only two tests shared by both cards are Geekbench OpenCL and Geekbench Vulkan. In OpenCL, the RTX 5080 scores 264461 against the A60’s 63485, a delta of -76% for the A60. In Vulkan, the RTX 5080 scores 257942 versus the A60’s 57166, a delta of -77.8%. These are lopsided results: the RTX 5080 is roughly 4.2x faster in OpenCL and 4.5x faster in Vulkan, based on the score ratios. The A60’s average benchmark score (60326) is higher than the RTX 5080’s (59188), but that average includes a broader set of tests—the RTX 5080 has 10 benchmark entries, while the A60 has only 2. The head-to-head data shows the RTX 5080 winning both tests, with a combined winsA of 0 and winsB of 2.

Where Each One Wins

The Arc Pro A60 wins in aggregate benchmark score: its average of 60326 is 1.9% higher than the RTX 5080’s 59188, and it is 1.9% above the RTX 5080 in the nearest-rival comparison. It also wins on physical and power characteristics: single-slot design, 130 W TDP, no external power connector, and a 300 W PSU requirement. It has four DisplayPort 2.0 outputs, which is more than the RTX 5080’s three DisplayPort 2.1b plus one HDMI. The A60 also has a lower transistor count and smaller die, which may appeal to users with space or power constraints.

The RTX 5080 wins decisively in raw compute performance. It has 5.3x the shading units, 6.7x the FP32 throughput, and 2.5x the memory bandwidth. It also has 84 RT cores and 336 tensor cores, enabling hardware ray tracing and AI workloads that the A60 lacks. In the head-to-head benchmarks, the RTX 5080 is 76% to 77.8% faster. It also has a higher boost clock (2617 MHz vs 2050 MHz), more memory (16 GB vs 12 GB), and a newer PCIe 5.0 interface. The RTX 5080’s release date is January 2025, nearly two years after the A60’s June 2023 launch.

For a builder who prioritizes compute performance, ray tracing, or modern gaming, the RTX 5080 is the obvious choice. For a workstation that needs a low-profile, single-slot card with multiple DisplayPort outputs and minimal power draw, the Arc Pro A60 is the more practical option. The data does not support a single winner—it depends entirely on the use case.