Intel Arc Pro A60M vs Intel Arc Pro B390 Comparison

FAQ

Q: What are the core architectural generations of these two Intel Arc Pro GPUs?

A: The Intel Arc Pro A60M is built on the Xe-HPG architecture, part of the Alchemist (Pro-Series Mobile) generation, using the DG2-256 chip. The Intel Arc Pro B390 is built on the Xe3-LPG architecture, part of the Arc Graphics-WM (Panther Lake) generation, using the Panther Lake chip.

Q: How do the process nodes and foundries differ between the two?

A: The Arc Pro A60M is manufactured on a 6 nm process at TSMC, with a die size of 269 mm² and 11,500 million transistors. The Arc Pro B390 is manufactured on a 3 nm process at Intel, with both die size and transistor count listed as unknown in the database.

Q: What are the memory configurations for each GPU?

A: The Arc Pro A60M uses 8 GB of GDDR6 memory on a 128-bit bus, delivering 256.0 GB/s of bandwidth with a memory clock of 2000 MHz (16 Gbps effective). The Arc Pro B390 uses system shared memory, with the type, bus width, and bandwidth all listed as system dependent.

Q: Which GPU has a higher FP32 compute throughput?

A: The Arc Pro B390 has a higher FP32 throughput at 7.680 TFLOPS, compared to the Arc Pro A60M at 5.325 TFLOPS. The B390 also leads in FP16 at 15.36 TFLOPS versus 10.65 TFLOPS for the A60M.

Q: What are the power consumption figures for these two GPUs?

A: The Arc Pro A60M has a TDP of 95 W. The Arc Pro B390 has a TDP of 80 W and requires no power connectors, as it is an integrated graphics processor (IGP) with a bus interface of IGP.

Q: Do both GPUs support the same graphics APIs?

A: Yes, both the Arc Pro A60M and the Arc Pro B390 support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both are listed as Active in production status.

Where Each One Wins

The recorded data shows a clear split between the two GPUs based on workload characteristics. The Intel Arc Pro A60M wins in memory bandwidth and fixed-function throughput metrics tied to its dedicated memory subsystem. It delivers 256.0 GB/s of bandwidth versus the system dependent bandwidth of the Arc Pro B390. The A60M also has higher pixel rate at 83.20 GPixel/s and texture rate at 166.4 GTexel/s.

The Intel Arc Pro B390 wins in raw shader compute and clock speed. Its boost clock of 2500 MHz substantially exceeds the 1300 MHz boost of the A60M. The B390 also delivers higher FP32 throughput (7.680 TFLOPS) and higher FP16 throughput (15.36 TFLOPS), indicating stronger general compute performance for shader-heavy workloads.

The A60M holds an advantage in memory capacity with 8 GB of dedicated GDDR6, which suits workloads requiring large local data residency. The B390 relies on system shared memory, making its effective memory performance dependent on the host platform.

The B390 wins on power efficiency per TDP: it delivers 7.680 TFLOPS within an 80 W TDP, while the A60M delivers 5.325 TFLOPS within a 95 W TDP. The B390 also has a smaller process node (3 nm versus 6 nm), indicating a more advanced manufacturing process.

Architecture Differences

The Intel Arc Pro A60M uses the DG2-256 chip based on the Xe-HPG architecture. This is a discrete mobile GPU with a 128-bit memory bus and dedicated GDDR6 memory. It has 2048 shading units, 128 texture mapping units, 64 ROPs, and 16 ray tracing cores. The chip contains 11,500 million transistors on a 269 mm² die, produced on a 6 nm TSMC process. The transistor density is 42.8M per mm².

The Intel Arc Pro B390 uses the Panther Lake chip based on the Xe3-LPG architecture. This is an integrated GPU with system shared memory and an IGP bus interface. It has 1536 shading units, 48 texture mapping units, 24 ROPs, and 12 ray tracing cores. The chip is produced on a 3 nm Intel process, with transistor count and die size unknown in the database.

The architectural difference is significant: Xe-HPG is a discrete GPU design optimized for dedicated memory bandwidth, while Xe3-LPG is an integrated design that shares system memory. The B390 has fewer shading units (1536 versus 2048) but compensates with a much higher boost clock (2500 MHz versus 1300 MHz). The B390 also has fewer TMUs (48 versus 128) and fewer ROPs (24 versus 64), which explains its lower pixel and texture rates despite the higher compute throughput.

The A60M has a base clock of 900 MHz and a boost clock of 1300 MHz. The B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The B390's power connectors are listed as "None", consistent with an integrated design. The A60M uses PCIe 4.0 x16 as its bus interface, while the B390 uses IGP.

Specification Differences

The database records the following specification differences between the two GPUs:

| Specification | Intel Arc Pro A60M | Intel Arc Pro B390 |

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

| Architecture | Xe-HPG | Xe3-LPG |

| Generation | Alchemist (Pro-Series Mobile) | Arc Graphics-WM (Panther Lake) |

| Process Node | 6 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 11,500 million | Unknown |

| Die Size | 269 mm² | Unknown |

| Transistor Density | 42.8M / mm² | Not listed |

| Base Clock | 900 MHz | 300 MHz |

| Boost Clock | 1300 MHz | 2500 MHz |

| Memory Clock | 2000 MHz (16 Gbps effective) | System Shared |

| Memory Size | 8 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 128 bit | System Shared |

| Memory Bandwidth | 256.0 GB/s | System Dependent |

| Shading Units | 2048 | 1536 |

| TMUs | 128 | 48 |

| ROPs | 64 | 24 |

| Ray Tracing Cores | 16 | 12 |

| Pixel Rate | 83.20 GPixel/s | 60.00 GPixel/s |

| Texture Rate | 166.4 GTexel/s | 120.0 GTexel/s |

| FP32 | 5.325 TFLOPS | 7.680 TFLOPS |

| FP16 | 10.65 TFLOPS (2:1) | 15.36 TFLOPS (2:1) |

| TDP | 95 W | 80 W |

| Power Connectors | Not listed | None |

| Bus Interface | PCIe 4.0 x16 | IGP |

| Release Date | 2023-06-05 | 2026-01-26 |

| Predecessor | Not listed | HD Graphics-WM |

Both GPUs share the same manufacturer (Intel), the same API support (DirectX 12 Ultimate, OpenGL 4.6, Vulkan 1.4), the same slot width (IGP), the same display outputs (Portable Device Dependent), and the same production status (Active). Neither has a launch MSRP recorded.

Head-to-Head Benchmarks

The database shows no direct head-to-head benchmark entries for these two GPUs, and both have zero recorded benchmark scores with an average benchmark score of 0. The percentile versus all GPUs is 50 for both. However, the specification data allows for direct comparison of computed throughput rates.

The most significant win for the Arc Pro B390 is in FP32 compute: 7.680 TFLOPS versus 5.325 TFLOPS, a margin of 2.355 TFLOPS. This represents a 44% advantage over the A60M in raw shader compute. The FP16 comparison shows a similar pattern: 15.36 TFLOPS versus 10.65 TFLOPS, a 4.71 TFLOPS difference.

The boost clock difference is substantial: 2500 MHz for the B390 versus 1300 MHz for the A60M, which is a 1200 MHz advantage. This higher clock enables the B390 to achieve higher compute throughput despite having 512 fewer shading units.

The Arc Pro A60M wins in pixel rate: 83.20 GPixel/s versus 60.00 GPixel/s, a difference of 23.20 GPixel/s. The texture rate comparison shows 166.4 GTexel/s for the A60M versus 120.0 GTexel/s for the B390, a 46.4 GTexel/s advantage.

The memory bandwidth comparison is not directly numeric because the B390 uses system shared memory with system dependent bandwidth. The A60M's dedicated 256.0 GB/s provides a fixed bandwidth figure, while the B390's effective bandwidth depends on the host system.

The TDP comparison favors the B390: 80 W versus 95 W, a 15 W difference. Combined with the higher FP32 throughput, the B390 achieves 0.096 TFLOPS per watt versus 0.056 TFLOPS per watt for the A60M.

The release date difference indicates the B390 is newer by roughly two and a half years, with the A60M released on 2023-06-05 and the B390 on 2026-01-26.

The Verdict

The data points to two distinct use cases. The Intel Arc Pro A60M is better suited for workloads that depend on fixed-function throughput and dedicated memory. Its higher pixel rate (83.20 GPixel/s) and texture rate (166.4 GTexel/s) indicate stronger performance for rasterization-heavy tasks. The 8 GB of dedicated GDDR6 memory with 256.0 GB/s of bandwidth provides a predictable memory subsystem that does not compete with the host system for bandwidth.

The Intel Arc Pro B390 is better suited for compute-heavy workloads that scale with shader throughput. Its FP32 throughput of 7.680 TFLOPS is 44% higher than the A60M, and its FP16 throughput of 15.36 TFLOPS is 44% higher as well. The 2500 MHz boost clock gives it a significant advantage in clock-limited scenarios. The 80 W TDP also indicates lower power draw, and the 3 nm process node suggests a more advanced manufacturing technology.

The A60M has more shading units (2048 versus 1536), more TMUs (128 versus 48), more ROPs (64 versus 24), and more ray tracing cores (16 versus 12). The B390 compensates with a much higher clock speed and a newer architecture generation (Xe3-LPG versus Xe-HPG).

For users prioritizing memory bandwidth, fixed-function throughput, and dedicated VRAM capacity, the Arc Pro A60M is the appropriate selection. For users prioritizing raw compute throughput, clock speed, and power efficiency, the Arc Pro B390 is the appropriate selection.

Both GPUs are active in production, and neither has recorded benchmark scores in the database, so real-world performance comparisons beyond specifications are not available. The percentile versus all GPUs is identical at 50 for both, indicating no measured performance separation in the database. The choice between them should be guided by the specific workload requirements: memory-bound rasterization for the A60M, or compute-bound shader workloads for the B390.