Intel Arc A380M vs Intel Arc Pro B60 Dual Comparison

Head-to-Head Benchmarks

The recorded database contains no direct head-to-head benchmark entries for the Intel Arc A380M versus the Intel Arc Pro B60 Dual. Both entries list empty benchmark arrays, and the wins tally shows zero for each side. This absence of measured performance data means the comparison must rely entirely on the architectural and specification fields captured in the database.

The Intel Arc A380M belongs to the Alchemist generation under the Arc 3 Mobile branding, built on the DG2-128 chip using the Xe-HPG architecture. It operates on a 6 nm process from TSMC, with a transistor count of 7,200 million on a 157 mm² die. Its compute configuration includes 1,024 shading units, 64 texture mapping units, 32 raster output units, and 8 ray tracing cores. The FP32 throughput is recorded at 4.096 TFLOPS, with FP16 at 8.192 TFLOPS using a 2:1 ratio. Pixel rate reaches 64.00 GPixel/s and texture rate hits 128.0 GTexel/s.

The Intel Arc Pro B60 Dual sits in the Battlemage generation under the Pro Series branding, using the BMG-G21 chip with the Xe2-HPG architecture. It is manufactured on a 5 nm TSMC process and carries 19,600 million transistors across a 272 mm² die. The shader complex scales to 2,560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. FP32 performance is listed at 12.29 TFLOPS, while FP16 reaches 24.58 TFLOPS at a 2:1 ratio. Pixel rate is 192.0 GPixel/s and texture rate is 384.0 GTexel/s.

Without benchmark scores, the raw compute ratios derived from database fields indicate the Pro B60 Dual delivers approximately 3 times the FP32 throughput of the A380M, and roughly 3 times the pixel and texture throughput. These are structural ratios from the listed specifications, not measured results.

Where Each One Wins

The Intel Arc A380M holds advantages in several database fields that favor compact, low-power deployments. Its TDP is recorded at 35 W, compared to 400 W for the Pro B60 Dual. The A380M uses an MXM Module slot width with an MXM-A (3.1) bus interface, making it a mobile-oriented form factor. Its display outputs are listed as portable device dependent, which reflects its design for embedded or laptop use. The A380M was released on 2023-01-23, more than two years before the Pro B60 Dual's 2025-09-04 release date, and it uses 6 GB of GDDR6 memory on a 96-bit bus with 186.0 GB/s bandwidth.

The Intel Arc Pro B60 Dual wins on every raw performance-oriented field in the database. Its 24 GB GDDR6 memory on a 192-bit bus provides 456.0 GB/s of bandwidth, which is roughly 2.45 times the A380M's bandwidth. Memory capacity is 4 times larger. The Pro B60 Dual's base clock of 2000 MHz and boost clock of 2400 MHz exceed the A380M's 1550 MHz base and 2000 MHz boost. Memory clock is recorded at 2375 MHz with 19 Gbps effective transfer, versus 1937 MHz and 15.5 Gbps effective for the A380M.

The Pro B60 Dual also carries a 16-pin power connector and a suggested PSU of 800 W, while the A380M lists no power connector or suggested PSU. The Pro B60 Dual's physical dimensions are 300 mm in length, 110 mm in height, and 40 mm in width, presented as a dual-slot card. It uses a PCIe 5.0 x8 bus interface and provides 4x mini-DisplayPort 2.1 outputs. The process node advantage goes to the Pro B60 Dual at 5 nm versus 6 nm, and its transistor density of 72.1M per mm² exceeds the A380M's 45.9M per mm².

The Verdict

The data indicates two distinctly positioned products. The Intel Arc A380M is a 35 W MXM module targeting portable or embedded systems where power draw and physical footprint are constrained. Its 6 GB memory and 186.0 GB/s bandwidth suit lighter workloads, and its 4.096 TFLOPS FP32 places it in a lower compute tier. The Pro B60 Dual is a 400 W dual-slot workstation card with a 1,199 USD launch MSRP, 24 GB memory, 456.0 GB/s bandwidth, and 12.29 TFLOPS FP32. The performance class separation is substantial: the Pro B60 Dual shows roughly 3x the FP32, pixel rate, and texture rate of the A380M, along with 4x the memory capacity and 2.45x the bandwidth.

Users constrained by power budgets or mobile form factors would align with the A380M's specifications. Users requiring high memory capacity, high bandwidth, or professional display outputs would align with the Pro B60 Dual. The absence of measured benchmarks in the database means verdicts rest on specification deltas rather than empirical testing. Both cards share the same DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 API support, so software compatibility is consistent at the API level.

FAQ

Q: Which card has higher FP32 compute performance?

A: The Intel Arc Pro B60 Dual records 12.29 TFLOPS FP32, while the Intel Arc A380M records 4.096 TFLOPS. The Pro B60 Dual provides approximately 3 times the FP32 throughput.

Q: What are the memory differences between the two cards?

A: The A380M uses 6 GB of GDDR6 on a 96-bit bus with 186.0 GB/s bandwidth. The Pro B60 Dual uses 24 GB of GDDR6 on a 192-bit bus with 456.0 GB/s bandwidth. The Pro B60 Dual has 4 times the memory capacity and roughly 2.45 times the bandwidth.

Q: Do both cards support the same graphics APIs?

A: Yes. Both list DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 in the database.

Q: How do the power requirements differ?

A: The A380M has a TDP of 35 W and lists no power connector or suggested PSU. The Pro B60 Dual has a TDP of 400 W, uses a 16-pin power connector, and lists a suggested PSU of 800 W.

Q: What form factors do the two cards use?

A: The A380M is an MXM Module with an MXM-A (3.1) bus interface. The Pro B60 Dual is a dual-slot card measuring 300 mm by 110 mm by 40 mm, using a PCIe 5.0 x8 bus interface.

Q: Which card has more ray tracing cores?

A: The Pro B60 Dual has 20 ray tracing cores. The A380M has 8 ray tracing cores.

Architecture Differences

The two cards belong to different Intel GPU generations. The A380M uses the DG2-128 chip with the Xe-HPG architecture, while the Pro B60 Dual uses the BMG-G21 chip with the Xe2-HPG architecture. The A380M is listed under the Alchemist generation with Arc 3 Mobile branding, and the Pro B60 Dual is listed under the Battlemage generation with Pro Series branding.

Process technology differs: the A380M is built on TSMC's 6 nm node, and the Pro B60 Dual uses TSMC's 5 nm node. Transistor counts scale accordingly, with the Pro B60 Dual at 19,600 million versus the A380M's 7,200 million. Die size also differs, with the Pro B60 Dual at 272 mm² versus 157 mm² for the A380M. Transistor density improves from 45.9M per mm² on the A380M to 72.1M per mm² on the Pro B60 Dual.

Compute resource allocation differs across every measured unit. The Pro B60 Dual has 2,560 shading units versus 1,024 on the A380M, 160 TMUs versus 64, 80 ROPs versus 32, and 20 ray tracing cores versus 8. The FP16 to FP32 ratio remains 2:1 on both cards, but the absolute figures scale with the shader count. The Pro B60 Dual records 24.58 TFLOPS FP16 and 12.29 TFLOPS FP32, while the A380M records 8.192 TFLOPS FP16 and 4.096 TFLOPS FP32.

Clock behavior also differs. The A380M runs at 1550 MHz base and 2000 MHz boost, while the Pro B60 Dual runs at 2000 MHz base and 2400 MHz boost. Neither card lists a game clock in the database. The memory clock for the A380M is 1937 MHz with 15.5 Gbps effective transfer, and the Pro B60 Dual runs at 2375 MHz with 19 Gbps effective transfer.

Interface architecture separates the two products. The A380M uses MXM-A (3.1) as its bus interface, reflecting a mobile module standard. The Pro B60 Dual uses PCIe 5.0 x8, a desktop-oriented interface with higher bandwidth potential. Display output configuration likewise differs: the A380M's outputs are listed as portable device dependent, while the Pro B60 Dual provides 4x mini-DisplayPort 2.1.

Specification Differences

The database records the following fields where the two cards differ:

• Chip: DG2-128 for the A380M, BMG-G21 for the Pro B60 Dual
• Architecture: Xe-HPG for the A380M, Xe2-HPG for the Pro B60 Dual
• Generation: Alchemist (Arc 3 Mobile) for the A380M, Battlemage (Pro Series) for the Pro B60 Dual
• Process node: 6 nm for the A380M, 5 nm for the Pro B60 Dual
• Transistors: 7,200 million for the A380M, 19,600 million for the Pro B60 Dual
• Die size: 157 mm² for the A380M, 272 mm² for the Pro B60 Dual
• Transistor density: 45.9M per mm² for the A380M, 72.1M per mm² for the Pro B60 Dual
• Base clock: 1550 MHz for the A380M, 2000 MHz for the Pro B60 Dual
• Boost clock: 2000 MHz for the A380M, 2400 MHz for the Pro B60 Dual
• Memory clock: 1937 MHz with 15.5 Gbps effective for the A380M, 2375 MHz with 19 Gbps effective for the Pro B60 Dual
• Memory size: 6 GB for the A380M, 24 GB for the Pro B60 Dual
• Memory bus width: 96 bit for the A380M, 192 bit for the Pro B60 Dual
• Memory bandwidth: 186.0 GB/s for the A380M, 456.0 GB/s for the Pro B60 Dual
• Shading units: 1,024 for the A380M, 2,560 for the Pro B60 Dual
• TMUs: 64 for the A380M, 160 for the Pro B60 Dual
• ROPs: 32 for the A380M, 80 for the Pro B60 Dual
• Ray tracing cores: 8 for the A380M, 20 for the Pro B60 Dual
• Pixel rate: 64.00 GPixel/s for the A380M, 192.0 GPixel/s for the Pro B60 Dual
• Texture rate: 128.0 GTexel/s for the A380M, 384.0 GTexel/s for the Pro B60 Dual
• FP32: 4.096 TFLOPS for the A380M, 12.29 TFLOPS for the Pro B60 Dual
• FP16: 8.192 TFLOPS for the A380M, 24.58 TFLOPS for the Pro B60 Dual
• TDP: 35 W for the A380M, 400 W for the Pro B60 Dual
• Slot width: MXM Module for the A380M, Dual-slot for the Pro B60 Dual
• Power connectors: none listed for the A380M, 1x 16-pin for the Pro B60 Dual
• Suggested PSU: none listed for the A380M, 800 W for the Pro B60 Dual
• Bus interface: MXM-A (3.1) for the A380M, PCIe 5.0 x8 for the Pro B60 Dual
• Display outputs: Portable device dependent for the A380M, 4x mini-DisplayPort 2.1 for the Pro B60 Dual
• Dimensions: none listed for the A380M, 300 mm by 110 mm by 40 mm for the Pro B60 Dual
• Release date: 2023-01-23 for the A380M, 2025-09-04 for the Pro B60 Dual
• Launch MSRP: not listed for the A380M, 1,199 USD for the Pro B60 Dual

Both cards share the same manufacturer, API set, memory type (GDDR6), and production status (Active). Neither lists a game clock, tensor cores, predecessor, successor, or a series name in the database.