Intel Arc A380M vs Intel Arc B770 Comparison

FAQ

Q: What are the core architecture differences between the Intel Arc A380M and Intel Arc B770?

A: The Arc A380M is built on the Xe-HPG architecture (Alchemist generation, Arc 3 Mobile) using the DG2-128 chip, fabricated on a 6 nm process at TSMC. The Arc B770 uses the Xe2-HPG architecture (Battlemage generation, Arc 7) with the BMG-G31 chip, fabricated on a 5 nm process at TSMC. The B770 die is 368 mm² compared to 157 mm² for the A380M.

Q: How do the memory subsystems compare?

A: The A380M has 6 GB of GDDR6 on a 96-bit bus, delivering 186.0 GB/s bandwidth. The B770 has 16 GB of GDDR6 on a 256-bit bus, delivering 512.0 GB/s bandwidth. The B770 offers nearly 2.7 times the memory capacity and roughly 2.75 times the bandwidth.

Q: What are the compute unit counts for each GPU?

A: The A380M features 1024 shading units, 64 texture mapping units, 32 render output units, and 8 ray tracing cores. The B770 features 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores, which are 4 times the shading units, 4 times the TMUs, 4 times the ROPs, and 4 times the ray tracing cores.

Q: What is the rated power consumption for each card?

A: The A380M has a TDP of 35 W and uses an MXM module form factor. The B770 has a TDP of 225 W, requires a dual-slot cooling solution, uses 1x 6-pin plus 1x 8-pin power connectors, and has a suggested power supply of 550 W.

Q: What is the difference in clock speeds?

A: The A380M has a base clock of 1550 MHz and a boost clock of 2000 MHz. The B770 has a base clock of 2100 MHz and a boost clock of 2400 MHz. The B770 runs 550 MHz higher at base and 400 MHz higher at boost.

Q: What are the FP32 and FP16 performance figures?

A: The A380M delivers 4.096 TFLOPS FP32 and 8.192 TFLOPS FP16 (2:1). The B770 delivers 19.66 TFLOPS FP32 and 39.32 TFLOPS FP16 (2:1). The B770 provides roughly 4.8 times the FP32 throughput and 4.8 times the FP16 throughput.

Architecture Differences

The two GPUs represent successive Intel architectures. The A380M uses the Xe-HPG architecture from the Alchemist generation, specifically the Arc 3 Mobile tier, built on the DG2-128 chip. The B770 uses the Xe2-HPG architecture from the Battlemage generation, specifically the Arc 7 tier, built on the BMG-G31 chip. The B770 lists Alchemist as its predecessor, confirming the direct generational progression.

The manufacturing process differs significantly. Both use TSMC as the foundry, but the A380M is on a 6 nm node while the B770 is on a 5 nm node. The A380M packs 7,200 million transistors into a 157 mm² die, yielding a transistor density of 45.9 million transistors per mm². The B770's transistor count is not recorded in the database, but its die size is 368 mm², which is 2.34 times the physical area of the A380M die.

The compute arrangement scales by a factor of four across the board. The A380M has 1024 shading units, 64 TMUs, 32 ROPs, and 8 ray tracing cores. The B770 has 4096 shading units, 256 TMUs, 128 ROPs, and 32 ray tracing cores. This uniform 4x scaling suggests a straightforward doubling of the underlying execution resources per cluster, though the Xe2-HPG architecture introduces changes beyond raw counts that are not detailed in the database.

Clock speeds also differ substantially. The A380M runs at 1550 MHz base and 2000 MHz boost. The B770 runs at 2100 MHz base and 2400 MHz boost. The B770's higher clocks, combined with its larger die and newer process, position it as a much faster part on paper.

Memory architecture diverges sharply. The A380M uses a 96-bit bus with 6 GB GDDR6, while the B770 uses a 256-bit bus with 16 GB GDDR6. Memory clock is listed as 1937 MHz with 15.5 Gbps effective for the A380M, versus 2000 MHz with 16 Gbps effective for the B770. The B770's wider bus is the dominant factor in its bandwidth advantage.

The form factor and interface differ completely. The A380M is an MXM module (MXM-A 3.1 interface) with display outputs marked as "Portable Device Dependent," indicating it is intended for laptops or other mobile devices. The B770 is a dual-slot desktop card with a PCIe 4.0 x16 interface and fixed display outputs: 1x HDMI 2.1a and 3x DisplayPort 2.1.

Both cards support the same API levels: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither lists tensor cores in the database. The A380M is marked as Active in production status, while the B770's production status is not recorded. The A380M was released on 2023-01-23, and the B770's release date is listed as 2025-12-31.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark entries for these two GPUs. Both have an average benchmark score of 0 and no nearest rivals listed. The percentile versus all GPUs is 50 for each, which indicates the median position in the current database distribution, but with no actual benchmark data attached to either entry, this percentile is not derived from comparative testing results.

The absence of benchmark data means direct performance comparisons must rely on the recorded specification differences. The FP32 compute throughput tells a clear story: the B770 delivers 19.66 TFLOPS versus 4.096 TFLOPS for the A380M, a 4.8 times advantage. FP16 throughput scales similarly, with 39.32 TFLOPS versus 8.192 TFLOPS.

Pixel and texture rates reinforce this gap. The B770 achieves 307.2 GPixel/s and 614.4 GTexel/s. The A380M achieves 64.00 GPixel/s and 128.0 GTexel/s. The B770 leads by 4.8 times in pixel rate and 4.8 times in texture rate, matching the compute scaling.

Memory bandwidth is another decisive metric. The B770's 512.0 GB/s is 2.75 times the A380M's 186.0 GB/s. This bandwidth difference is smaller than the compute difference, but it still positions the B770 as the stronger card for memory-intensive workloads.

The clock advantage is modest but consistent. The B770 boosts to 2400 MHz versus 2000 MHz for the A380M, a 20% higher boost clock. Combined with the 4x resource scaling, the B770's per-clock advantage is not a major factor in the overall performance gap.

The B770 also carries a larger memory pool at 16 GB versus 6 GB. This 10 GB difference affects capacity-limited scenarios such as large texture sets, high-resolution rendering, and multi-tasking with GPU memory. The 96-bit bus on the A380M is a structural constraint that the B770 avoids entirely.

Specification Differences

| Specification | Intel Arc A380M | Intel Arc B770 |

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

| Architecture | Xe-HPG | Xe2-HPG |

| Generation | Alchemist (Arc 3 Mobile) | Battlemage (Arc 7) |

| Chip | DG2-128 | BMG-G31 |

| Process Node | 6 nm | 5 nm |

| Transistors | 7,200 million | unknown |

| Die Size | 157 mm² | 368 mm² |

| Transistor Density | 45.9M / mm² | null |

| Base Clock | 1550 MHz | 2100 MHz |

| Boost Clock | 2000 MHz | 2400 MHz |

| Memory Clock | 1937 MHz, 15.5 Gbps effective | 2000 MHz, 16 Gbps effective |

| Memory Size | 6 GB | 16 GB |

| Memory Type | GDDR6 | GDDR6 |

| Memory Bus Width | 96 bit | 256 bit |

| Memory Bandwidth | 186.0 GB/s | 512.0 GB/s |

| Shading Units | 1024 | 4096 |

| TMUs | 64 | 256 |

| ROPs | 32 | 128 |

| Ray Tracing Cores | 8 | 32 |

| Pixel Rate | 64.00 GPixel/s | 307.2 GPixel/s |

| Texture Rate | 128.0 GTexel/s | 614.4 GTexel/s |

| FP32 | 4.096 TFLOPS | 19.66 TFLOPS |

| FP16 | 8.192 TFLOPS (2:1) | 39.32 TFLOPS (2:1) |

| TDP | 35 W | 225 W |

| Slot Width | MXM Module | Dual-slot |

| Power Connectors | null | 1x 6-pin + 1x 8-pin |

| Suggested PSU | null | 550 W |

| Bus Interface | MXM-A (3.1) | PCIe 4.0 x16 |

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1a, 3x DisplayPort 2.1 |

| Production Status | Active | null |

| Release Date | 2023-01-23 | 2025-12-31 |

| Predecessor | null | Alchemist |

Where Each One Wins

The A380M wins in power efficiency and form factor flexibility. Its 35 W TDP is a fraction of the B770's 225 W, and its MXM module design allows integration into portable devices where the dual-slot B770 cannot fit. The A380M requires no external power connectors and no minimum power supply rating, making it suitable for systems with limited power delivery. Its smaller die size of 157 mm² and lower transistor count of 7,200 million also point to lower manufacturing cost per unit, though pricing data is not recorded.

The B770 wins in every raw performance category recorded in the database. It delivers 4.8 times the FP32 throughput, 4.8 times the FP16 throughput, 4.8 times the pixel rate, and 4.8 times the texture rate. It has 2.75 times the memory bandwidth and 2.67 times the memory capacity. Its 16 GB frame buffer is more than double the A380M's 6 GB, which matters for modern game assets and compute workloads that exceed 6 GB.

The B770 also wins on clock speed, with a 2100 MHz base and 2400 MHz boost versus 1550 MHz and 2000 MHz for the A380M. The B770's 32 ray tracing cores versus 8 for the A380M gives it a 4x advantage in ray tracing workloads, assuming the Xe2-HPG architecture maintains similar per-core efficiency. The B770's 256-bit memory bus is also a structural advantage that avoids the bandwidth bottleneck of the A380M's 96-bit bus.

For desktop use cases, the B770 is the clear choice based on the recorded data. Its PCIe 4.0 x16 interface, fixed display outputs with HDMI 2.1a and DisplayPort 2.1, and dual-slot cooling solution are designed for traditional desktop systems. The B770's 368 mm² die and 5 nm process node indicate a high-end part intended for demanding workloads.

For mobile or embedded use cases, the A380M is the only viable option between the two. Its MXM-A 3.1 interface and portable-device-dependent display outputs are designed for laptops, compact systems, and other form factors where the B770's PCIe slot and dual-slot cooler cannot be accommodated. The A380M's 35 W TDP also makes it feasible in thermally constrained chassis.

The database shows no benchmark wins for either GPU, so the use-case split relies entirely on the specification differences. The B770 dominates compute, memory, and connectivity. The A380M dominates power consumption and physical footprint. Neither card has a recorded launch MSRP, so no cost comparison is possible from the available data.

The B770's release date of 2025-12-31 places it later than the A380M's 2023-01-23 release. The B770's predecessor is listed as Alchemist, which is the A380M's generation, confirming the direct lineage. The A380M remains active in production, while the B770's production status is not recorded.