Intel Arc A380E vs Intel Arc Pro B390 Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark results between the Intel Arc A380E and the Intel Arc Pro B390. Both entries show zero benchmark scores and zero wins in the comparison matrix. This absence of measured performance data means the comparison must rely on the architectural and specification differences captured in the database.

The Intel Arc A380E posts a theoretical FP32 throughput of 4.096 TFLOPS, while the Intel Arc Pro B390 delivers 7.680 TFLOPS. That is a 87.5% higher raw compute figure for the B390. In FP16, the gap persists: the A380E manages 8.192 TFLOPS (2:1) against the B390's 15.36 TFLOPS (2:1). The B390 doubles the shader count as well, with 1536 shading units versus 1024 on the A380E.

The A380E counters in fixed-function throughput. Its pixel rate of 64.00 GPixel/s is 6.7% higher than the B390's 60.00 GPixel/s, and its texture rate of 128.0 GTexel/s is 6.7% higher than 120.0 GTexel/s. The A380E also carries more texture mapping units (64 versus 48) and more render output units (32 versus 24). These numbers indicate the A380E retains an edge in fill-rate-bound work, despite its lower raw compute.

Memory separates the two completely. The A380E uses 6 GB of dedicated GDDR6 on a 96-bit bus, yielding 186.0 GB/s of bandwidth. The B390 uses system shared memory, with bandwidth listed as system dependent. For workloads that rely on consistent memory throughput, the A380E's dedicated pool provides a fixed, measurable advantage. The B390's shared memory performance cannot be quantified from the database, as it depends entirely on the host platform.

Clock behavior also differs sharply. The A380E runs at a flat 2000 MHz for both base and boost, while the B390 idles at 300 MHz base and boosts to 2500 MHz. The B390's boost clock is 25% higher than the A380E's fixed clock, which contributes to its FP32 advantage. However, the B390's low base clock suggests a power-adaptive design typical of an integrated GPU, where sustained clocks depend on thermal and power headroom.

The Verdict

From the recorded data, the Intel Arc Pro B390 is the stronger compute device. Its FP32 figure of 7.680 TFLOPS is 87.5% above the A380E's 4.096 TFLOPS, and its FP16 output of 15.36 TFLOPS is similarly dominant. The B390 also has 50% more shading units (1536 versus 1024) and 50% more ray tracing cores (12 versus 8). For any workload that scales with shader count or ray tracing throughput, the B390 holds the advantage.

The Intel Arc A380E remains relevant for memory-bound tasks. Its 6 GB GDDR6 pool with 186.0 GB/s bandwidth is a concrete, dedicated resource. The B390's system shared memory has no fixed bandwidth figure in the database, which introduces platform dependency. The A380E also wins in pixel fill rate (64.00 GPixel/s versus 60.00 GPixel/s) and texture fill rate (128.0 GTexel/s versus 120.0 GTexel/s).

The production status favors the B390. It is listed as active, while the A380E is end-of-life. The B390's release date is also later, and its successor field is empty, whereas the A380E has Battlemage listed as its successor. This suggests the B390 represents the current direction of Intel's graphics lineup.

Neither GPU has a launch MSRP in the database, and neither has measured benchmark scores. The verdict rests entirely on architectural and specification data.

Architecture Differences

The two GPUs come from different generations and process nodes. The Intel Arc A380E uses the DG2-128 chip built on Xe-HPG architecture, belonging 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. The transistor density works out to 45.9M per mm².

The Intel Arc Pro B390 uses the Panther Lake chip built on Xe3-LPG architecture, belonging to the Arc Graphics-WM (Panther Lake) generation. It is fabricated on a 3 nm process at Intel. The database lists transistor count and die size as unknown, so no density figure can be calculated.

The A380E is a discrete, single-slot card with no power connectors and a 75 W TDP. It uses a PCIe 4.0 x8 bus interface and offers 4x DisplayPort 2.0 outputs. Dimensions are recorded at 254 mm length, 127 mm height, and 20 mm width. The suggested PSU is 250 W.

The B390 is an integrated GPU (IGP) with an 80 W TDP. It has no power connectors, uses the IGP bus interface, and its display outputs are listed as portable device dependent. No dimensions are recorded. No suggested PSU is listed, which is consistent with an integrated part that draws from the host system's power delivery.

The A380E has its predecessor listed as Xe Graphics and its successor as Battlemage. The B390 lists HD Graphics-WM as its predecessor and has no successor recorded.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have 4.6 OpenGL support. Neither lists tensor cores in the database.

FAQ

Q: Which GPU has higher raw compute throughput?

A: The Intel Arc Pro B390. Its FP32 output is 7.680 TFLOPS, which is 87.5% higher than the A380E's 4.096 TFLOPS. The B390 also leads in FP16 with 15.36 TFLOPS versus 8.192 TFLOPS.

Q: Does the A380E have any performance advantage?

A: Yes, in fill rate. The A380E delivers 64.00 GPixel/s versus 60.00 GPixel/s on the B390, and 128.0 GTexel/s versus 120.0 GTexel/s. It also has more TMUs (64 versus 48) and more ROPs (32 versus 24).

Q: How do the memory configurations differ?

A: The A380E has 6 GB of GDDR6 on a 96-bit bus with 186.0 GB/s bandwidth. The B390 uses system shared memory with system dependent bandwidth. No fixed memory size or bus width is recorded for the B390.

Q: What are the clock speed differences?

A: The A380E runs at a constant 2000 MHz for both base and boost. The B390 has a 300 MHz base clock and a 2500 MHz boost clock, which is 25% higher than the A380E's fixed clock.

Q: Which GPU has more shader and ray tracing cores?

A: The B390. It has 1536 shading units versus 1024 on the A380E, and 12 ray tracing cores versus 8. Both are 50% increases.

Q: What is the production status of each GPU?

A: The A380E is end-of-life, while the B390 is active. The A380E's successor is listed as Battlemage, and the B390 has no successor recorded.

Where Each One Wins

The Intel Arc Pro B390 wins in compute-heavy scenarios. Its FP32 throughput of 7.680 TFLOPS is 87.5% above the A380E, and its FP16 throughput of 15.36 TFLOPS is 87.5% above as well. The 1536 shading units and 12 ray tracing cores give it a structural advantage for shader-intensive rendering and ray traced workloads. The 2500 MHz boost clock, 25% above the A380E's fixed 2000 MHz, supports this compute lead.

The Intel Arc A380E wins in fixed-function throughput. Its 64.00 GPixel/s pixel rate beats the B390's 60.00 GPixel/s, and its 128.0 GTexel/s texture rate beats the B390's 120.0 GTexel/s. The higher TMU and ROP counts (64 and 32 versus 48 and 24) reinforce this. The dedicated 6 GB GDDR6 pool with 186.0 GB/s bandwidth is a clear advantage for workloads that need predictable memory performance, since the B390's shared memory bandwidth is system dependent and unquantified.

The B390 is an integrated GPU with an 80 W TDP and no power connectors, making it suitable for compact or portable systems where a discrete card cannot fit. The A380E is a discrete single-slot card with a 75 W TDP, no power connectors, and a 250 W suggested PSU, requiring an available PCIe 4.0 x8 slot and four DisplayPort 2.0 connections for output.

The production status further separates them. The A380E is end-of-life with a successor already named, while the B390 is active. The B390's 3 nm Intel process node is more advanced than the A380E's 6 nm TSMC node, though the database does not record transistor count or die size for the B390, so density comparisons are not possible.

Specification Differences

| Field | Intel Arc A380E | Intel Arc Pro B390 |

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

| Chip | DG2-128 | Panther Lake |

| Architecture | Xe-HPG | Xe3-LPG |

| Generation | Alchemist (Arc 3) | Arc Graphics-WM (Panther Lake) |

| Process node | 6 nm (TSMC) | 3 nm (Intel) |

| Transistors | 7,200 million | unknown |

| Die size | 157 mm² | unknown |

| Transistor density | 45.9M / mm² | null |

| Base clock | 2000 MHz | 300 MHz |

| Boost clock | 2000 MHz | 2500 MHz |

| Memory | 6 GB GDDR6, 96-bit, 186.0 GB/s | System Shared, system dependent |

| Shading units | 1024 | 1536 |

| TMUs | 64 | 48 |

| ROPs | 32 | 24 |

| Ray tracing cores | 8 | 12 |

| Pixel rate | 64.00 GPixel/s | 60.00 GPixel/s |

| Texture rate | 128.0 GTexel/s | 120.0 GTexel/s |

| FP32 | 4.096 TFLOPS | 7.680 TFLOPS |

| FP16 | 8.192 TFLOPS (2:1) | 15.36 TFLOPS (2:1) |

| TDP | 75 W | 80 W |

| Slot width | Single-slot | IGP |

| Power connectors | None | None |

| Suggested PSU | 250 W | null |

| Bus interface | PCIe 4.0 x8 | IGP |

| Display outputs | 4x DisplayPort 2.0 | Portable Device Dependent |

| Dimensions | 254 mm x 127 mm x 20 mm | null |

| Production status | End-of-life | Active |

| Release date | 2024-03-31 | 2026-01-26 |

| Predecessor | Xe Graphics | HD Graphics-WM |

| Successor | Battlemage | null |

| Launch MSRP | null | null |

Both GPUs share DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 support. Neither lists tensor cores, game clocks, or codenames in the database.