Intel Arc A380E vs Intel Arc B580 Comparison

The Intel Arc A380E and Intel Arc B580 represent two distinct generations of Intel’s discrete GPU efforts, separated by a significant architectural leap. The A380E is an Alchemist-era entry-level part, while the B580 is a Battlemage-generation mid-range card. The recorded data shows a substantial performance gulf between the two, driven by fundamental differences in chip design, memory configuration, and power delivery. This analysis relies solely on the database entries for both products.

Head-to-Head Benchmarks

Direct head-to-head benchmark comparisons between the two cards are not available in the database, as the A380E has no recorded benchmark scores. However, the B580’s performance metrics and its position among rival GPUs provide a clear reference point for what the newer architecture delivers. The B580’s average benchmark score is 23,021, which places it in the 68th percentile of all GPUs. In contrast, the A380E is listed at the 50th percentile with an average benchmark score of zero, indicating that no standardized test results have been logged for it.

The B580’s individual benchmark results show a wide range of capabilities. In 3DMark Steel Nomad (DX12), it scores 3,068. In Geekbench, it achieves 92,821 in OpenCL and 109,672 in Vulkan. Passmark results vary by API: DirectX 9 yields 183, DirectX 11 yields 128, DirectX 10 and DirectX 12 each yield 76. The Passmark G3D score is 15,748, while GPU Compute is 7,729, and G2D is 709. These figures indicate that the B580 is a well-rounded performer, with its compute and modern API scores being particularly strong.

The B580’s nearest rivals in the database include the AMD Radeon RX 580 2048SP (average score 23,061, delta -0.2%), the NVIDIA GeForce RTX 2080 (22,895, delta +0.6%), the NVIDIA GeForce RTX 3080 (23,172, delta -0.7%), and the NVIDIA P106-100 (23,249, delta -1%). This places the B580 within a tight cluster of older high-end and mid-range cards, effectively matching or slightly trailing them by margins under 1%. The A380E, with no scores, cannot be placed in such a comparison.

Architecture Differences

The two cards are built on different architectures and process nodes. The A380E uses the DG2-128 chip based on Xe-HPG architecture, which belongs to the Alchemist (Arc 3) generation. It is fabricated on a 6 nm process at TSMC, with 7,200 million transistors on a 157 mm² die, yielding a transistor density of 45.9 million per mm². The B580 uses the BMG-G21 chip based on Xe2-HPG architecture, which belongs to the Battlemage (Arc 5) generation. It is fabricated on a 5 nm process at TSMC, with 19,600 million transistors on a 272 mm² die, yielding a transistor density of 72.1 million per mm². The B580’s newer process and larger die allow for more than double the transistor count.

The B580 also carries more execution resources. It has 2,560 shading units, 160 texture mapping units, and 80 ROPs, compared to the A380E’s 1,024 shading units, 64 TMUs, and 32 ROPs. Ray tracing cores are more numerous as well: 20 on the B580 versus 8 on the A380E. Neither card lists tensor cores. The B580’s raw compute figures are substantially higher, with FP32 performance of 13.67 TFLOPS versus 4.096 TFLOPS on the A380E, and FP16 performance of 27.34 TFLOPS versus 8.192 TFLOPS (both at a 2:1 ratio).

Memory configurations differ sharply. The A380E has 6 GB of GDDR6 on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The B580 has 12 GB of GDDR6 on a 192-bit bus, delivering 456.0 GB/s of bandwidth. The B580’s effective memory speed is 19 Gbps, while the A380E runs at 15.5 Gbps effective. Clock speeds also favor the B580: a base and boost of 2670 MHz, compared to the A380E’s fixed 2000 MHz base and boost.

Power and physical specs diverge as well. The A380E has a TDP of 75 W, is single-slot, requires no power connectors, and lists a suggested PSU of 250 W. The B580 has a TDP of 190 W, is dual-slot, uses a single 8-pin power connector, and lists a suggested PSU of 450 W. The A380E is 254 mm long, 127 mm tall, and 20 mm wide, while the B580 is 272 mm long, 115 mm tall, and 45 mm wide. Display outputs differ: the A380E has 4x DisplayPort 2.0, while the B580 has 1x HDMI 2.1a and 3x DisplayPort 2.1. Both use a PCIe 4.0 x8 interface and support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

FAQ

Q: Which card has a higher average benchmark score?

A: The Intel Arc B580 has an average benchmark score of 23,021, while the Intel Arc A380E has no recorded benchmark scores, resulting in an average score of zero.

Q: How does the B580 compare to its nearest rivals?

A: The B580’s average score of 23,021 is 0.2% below the AMD Radeon RX 580 2048SP, 0.6% above the NVIDIA GeForce RTX 2080, 0.7% below the NVIDIA GeForce RTX 3080, and 1% below the NVIDIA P106-100.

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

A: The A380E has 6 GB of GDDR6 memory, while the B580 has 12 GB of GDDR6 memory.

Q: Which card has a higher transistor density?

A: The B580 has a transistor density of 72.1 million per mm², compared to the A380E’s 45.9 million per mm².

Q: Are the APIs supported by both cards the same?

A: Yes, both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: What is the production status of each card?

A: The A380E is listed as end-of-life, while the B580 is listed as active.

The Verdict

The data indicates a clear generational divide. The Intel Arc B580 is an active, current product with a full set of benchmark scores, a 68th percentile ranking, and performance that places it within 1% of several established NVIDIA and AMD cards. The Intel Arc A380E is an end-of-life product with no recorded benchmark results, leaving its performance unquantified in the database. For any task requiring measured performance, the B580 is the only card with supporting data.

The B580’s specifications also point to its superiority in raw throughput. It has 2.5 times the shading units, 2.5 times the texture units, 2.5 times the ROPs, and 2.5 times the ray tracing cores of the A380E. Its FP32 compute is more than three times higher, and its memory bandwidth is nearly 2.5 times higher. These are not incremental differences but substantial architectural steps forward.

The A380E does have some advantages in physical characteristics. It is a single-slot card with no power connectors, a 75 W TDP, and a suggested PSU of 250 W. The B580 requires a dual-slot footprint, a single 8-pin connector, a 190 W TDP, and a 450 W PSU. For constrained systems where power and space are critical, the A380E is the more accommodating option, but it offers no measurable performance evidence.

Specification Differences

The following fields differ between the two cards:

• Chip: DG2-128 (A380E) versus BMG-G21 (B580)
• Architecture: Xe-HPG (A380E) versus Xe2-HPG (B580)
• Generation: Alchemist (Arc 3) (A380E) versus Battlemage (Arc 5) (B580)
• Process Node: 6 nm (A380E) versus 5 nm (B580)
• Transistors: 7,200 million (A380E) versus 19,600 million (B580)
• Die Size: 157 mm² (A380E) versus 272 mm² (B580)
• Transistor Density: 45.9M / mm² (A380E) versus 72.1M / mm² (B580)
• Base Clock: 2000 MHz (A380E) versus 2670 MHz (B580)
• Boost Clock: 2000 MHz (A380E) versus 2670 MHz (B580)
• Memory Clock: 1937 MHz 15.5 Gbps effective (A380E) versus 2375 MHz 19 Gbps effective (B580)
• Memory Size: 6 GB (A380E) versus 12 GB (B580)
• Memory Bus Width: 96 bit (A380E) versus 192 bit (B580)
• Memory Bandwidth: 186.0 GB/s (A380E) versus 456.0 GB/s (B580)
• Shading Units: 1024 (A380E) versus 2560 (B580)
• TMUs: 64 (A380E) versus 160 (B580)
• ROPs: 32 (A380E) versus 80 (B580)
• RT Cores: 8 (A380E) versus 20 (B580)
• Pixel Rate: 64.00 GPixel/s (A380E) versus 213.6 GPixel/s (B580)
• Texture Rate: 128.0 GTexel/s (A380E) versus 427.2 GTexel/s (B580)
• FP32: 4.096 TFLOPS (A380E) versus 13.67 TFLOPS (B580)
• FP16: 8.192 TFLOPS (2:1) (A380E) versus 27.34 TFLOPS (2:1) (B580)
• TDP: 75 W (A380E) versus 190 W (B580)
• Slot Width: Single-slot (A380E) versus Dual-slot (B580)
• Power Connectors: None (A380E) versus 1x 8-pin (B580)
• Suggested PSU: 250 W (A380E) versus 450 W (B580)
• Display Outputs: 4x DisplayPort 2.0 (A380E) versus 1x HDMI 2.1a3x DisplayPort 2.1 (B580)
• Dimensions: 254 mm x 127 mm x 20 mm (A380E) versus 272 mm x 115 mm x 45 mm (B580)
• Production Status: End-of-life (A380E) versus Active (B580)
• Release Date: 2024-03-31 (A380E) versus 2024-12-12 (B580)
• Predecessor: Xe Graphics (A380E) versus Alchemist (B580)
• Successor: Battlemage (A380E) versus null (B580)
• Launch MSRP: null (A380E) versus 249 USD (B580)

Where Each One Wins

The Intel Arc B580 wins in every measurable performance category. Its benchmark scores, while not directly head-to-head with the A380E, establish it as a functional mid-range card with a 68th percentile ranking. Its FP32 throughput of 13.67 TFLOPS, memory bandwidth of 456.0 GB/s, and 12 GB of VRAM make it suitable for modern gaming and compute workloads. The B580’s performance relative to rivals, within 1% of cards like the RTX 2080 and RTX 3080, confirms its capability.

The Intel Arc A380E wins only in physical and power efficiency metrics. It is a single-slot, 75 W card that does not require any power connectors and lists a suggested PSU of 250 W. Its dimensions are shorter (254 mm) and thinner (20 mm) than the B580, although it is taller (127 mm versus 115 mm). This makes it a candidate for compact or low-power systems, but the database contains no benchmark scores to validate its actual performance.

Given the absence of A380E benchmark data, the B580 is the only card with demonstrated performance. The A380E’s end-of-life status and lack of recorded scores mean that any comparison of wins is heavily skewed toward the B580. For users who prioritize measured performance, the B580 is the clear choice. For users with strict power or space constraints, the A380E offers a lower-power footprint, but its performance remains unverified in the database.