Intel Arc A380E vs Intel Arc Pro B70 Comparison

Head-to-Head Benchmarks

The recorded database contains no direct benchmark runs for either the Intel Arc A380E or the Intel Arc Pro B70. Both entries show an average benchmark score of zero and a percentile rank of 50 against all GPUs, which places them at the median of the distribution solely based on their specification data. Without measured performance numbers, the comparison rests entirely on the computed capabilities derived from each chip's architecture.

The raw compute throughput tells a clear story. The Arc A380E delivers 4.096 TFLOPS of FP32 performance, while the Arc Pro B70 reaches 22.94 TFLOPS. That represents a 5.6x gap in raw shader output. In FP16 workloads, the A380E achieves 8.192 TFLOPS with a 2:1 ratio, whereas the B70 reaches 45.88 TFLOPS under the same 2:1 ratio, again a 5.6x difference. These figures scale directly from the shading unit counts: 1024 on the A380E versus 4096 on the B70.

Texture throughput shows a similar pattern. The A380E renders 128.0 GTexel/s, and the B70 reaches 716.8 GTexel/s, a 5.6x advantage. Pixel fill rates diverge even more sharply. The A380E manages 64.00 GPixel/s, while the B70 hits 358.4 GPixel/s, a 5.6x margin as well. The consistency of these ratios across every throughput metric indicates the B70's architecture scales the same shader, texture, and raster units by a factor of four, then adds higher clock speeds on top.

Memory bandwidth amplifies the difference further. The A380E has 186.0 GB/s of bandwidth from a 96-bit bus with 6 GB of GDDR6. The B70 offers 608.0 GB/s from a 256-bit bus with 32 GB of GDDR6. That is a 3.27x bandwidth advantage for the B70, slightly less than the compute gap but still substantial. The memory clock also differs: the A380E runs at 1937 MHz (15.5 Gbps effective), while the B70 runs at 2375 MHz (19 Gbps effective).

The clock speeds favor the B70 as well. The A380E has a base and boost of 2000 MHz, with no separation between the two. The B70 has a base of 2280 MHz and a boost of 2800 MHz, meaning its boost exceeds its base by 520 MHz. Even at base clocks, the B70 runs 14% faster than the A380E's fixed frequency. At boost, the B70 runs 40% faster. This clock advantage compounds with the wider execution resources.

Neither card has any recorded wins in the head-to-head benchmark section. The winsA and winsB fields both read zero. The database therefore provides no empirical validation of these theoretical figures, but the specification-derived deltas are consistent across all measured subcomponents.

Architecture Differences

The two GPUs come from different Intel generations. The Arc A380E uses the DG2-128 chip built on the Xe-HPG architecture, belonging to the Alchemist (Arc 3) generation. The Arc Pro B70 uses the BMG-G31 chip built on the Xe2-HPG architecture, belonging to the Battlemage (Pro Series) generation. This is a generational leap from Alchemist to Battlemage.

The manufacturing process differs accordingly. The A380E uses a 6 nm process at TSMC with 7,200 million transistors on a 157 mm² die. The B70 uses a 5 nm process at TSMC with a 368 mm² die, though its transistor count is listed as unknown. The transistor density for the A380E is 45.9M / mm², while the B70's density is not recorded. The die size difference is large: 368 mm² versus 157 mm², a 2.34x increase in silicon area.

The memory subsystem changes completely. The A380E uses 6 GB of GDDR6 on a 96-bit bus. The B70 uses 32 GB of GDDR6 on a 256-bit bus. The B70's memory capacity is 5.33x larger, and its bus width is 2.67x wider. The effective memory speed also rises from 15.5 Gbps to 19 Gbps.

The execution resource counts scale up dramatically on the B70. Shading units go from 1024 to 4096. Texture mapping units go from 64 to 256. Raster output units go from 32 to 128. Ray tracing cores go from 8 to 32. The tensor core field is null for both, so no comparison is possible there. The B70 has four times the shaders, four times the TMUs, four times the ROPs, and four times the RT cores compared to the A380E.

The bus interface changes as well. The A380E uses PCIe 4.0 x8. The B70 uses PCIe 5.0 x16. This gives the B70 a much wider and faster connection to the host system, which matters for data transfer in professional workloads.

Display outputs differ. The A380E provides four DisplayPort 2.0 outputs. The B70 provides one HDMI 2.1a and three DisplayPort 2.1 outputs. The B70 adds HDMI support and upgrades the DisplayPort standard from 2.0 to 2.1.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The API feature sets are identical, so software compatibility is not a differentiator.

Power delivery changes significantly. The A380E has a TDP of 75 W with no power connectors, drawing everything from the PCIe slot. The B70 has a TDP of 230 W and requires one 8-pin power connector. The suggested PSU also rises from 250 W to 550 W. The physical size follows suit: the A380E is a single-slot card measuring 254 mm by 127 mm by 20 mm, while the B70 is a dual-slot card measuring 267 mm by 110 mm by 39 mm.

FAQ

Q: What is the difference in FP32 compute between the two cards?

A: The Intel Arc A380E delivers 4.096 TFLOPS of FP32 performance, while the Intel Arc Pro B70 delivers 22.94 TFLOPS. The B70 has 5.6 times the raw FP32 throughput.

Q: How much memory does each card have and what is the bandwidth?

A: The A380E has 6 GB of GDDR6 on a 96-bit bus with 186.0 GB/s bandwidth. The B70 has 32 GB of GDDR6 on a 256-bit bus with 608.0 GB/s bandwidth.

Q: Which card has more ray tracing cores?

A: The A380E has 8 ray tracing cores. The B70 has 32 ray tracing cores, which is four times the count.

Q: What are the boost clock speeds?

A: The A380E has a fixed clock of 2000 MHz for both base and boost. The B70 has a base clock of 2280 MHz and a boost clock of 2800 MHz.

Q: Do both cards support the same DirectX version?

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

Q: What is the power requirement difference?

A: The A380E has a 75 W TDP and requires no power connectors. The B70 has a 230 W TDP and requires one 8-pin power connector. The suggested PSU for the A380E is 250 W, while the B70 suggests 550 W.

Specification Differences

| Specification | Intel Arc A380E | Intel Arc Pro B70 |

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

| Chip | DG2-128 | BMG-G31 |

| Architecture | Xe-HPG | Xe2-HPG |

| Generation | Alchemist (Arc 3) | Battlemage (Pro Series) |

| Process Node | 6 nm | 5 nm |

| Die Size | 157 mm² | 368 mm² |

| Transistors | 7,200 million | unknown |

| Transistor Density | 45.9M / mm² | null |

| Base Clock | 2000 MHz | 2280 MHz |

| Boost Clock | 2000 MHz | 2800 MHz |

| Memory Clock | 1937 MHz 15.5 Gbps effective | 2375 MHz 19 Gbps effective |

| Memory Size | 6 GB | 32 GB |

| Memory Type | GDDR6 | GDDR6 |

| Memory Bus Width | 96 bit | 256 bit |

| Memory Bandwidth | 186.0 GB/s | 608.0 GB/s |

| Shading Units | 1024 | 4096 |

| TMUs | 64 | 256 |

| ROPs | 32 | 128 |

| Ray Tracing Cores | 8 | 32 |

| Pixel Rate | 64.00 GPixel/s | 358.4 GPixel/s |

| Texture Rate | 128.0 GTexel/s | 716.8 GTexel/s |

| FP32 | 4.096 TFLOPS | 22.94 TFLOPS |

| FP16 | 8.192 TFLOPS (2:1) | 45.88 TFLOPS (2:1) |

| TDP | 75 W | 230 W |

| Slot Width | Single-slot | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | 250 W | 550 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 5.0 x16 |

| Display Outputs | 4x DisplayPort 2.0 | 1x HDMI 2.1a, 3x DisplayPort 2.1 |

| Dimensions | 254 mm 10 inches, 127 mm 5 inches, 20 mm 0.8 inches | 267 mm 10.5 inches, 110 mm 4.3 inches, 39 mm 1.5 inches |

| Release Date | 2024-03-31 | 2026-03-25 |

| Production Status | End-of-life | null |

| Launch MSRP | null | 949 USD |

Where Each One Wins

The Intel Arc A380E wins in the domain of low-power, compact installations. Its 75 W TDP requires no auxiliary power connectors, allowing it to fit into systems with a 250 W suggested PSU. Its single-slot design at 254 mm length and 20 mm width occupies minimal physical space. The four DisplayPort 2.0 outputs support multi-monitor setups without needing HDMI. Its 6 GB memory and 96-bit bus are sufficient for basic display output and light compute tasks. The end-of-life production status indicates it is a legacy part, but its low power draw and small footprint remain advantages for constrained environments.

The Intel Arc Pro B70 wins in every performance category. Its 22.94 TFLOPS FP32 throughput, 608.0 GB/s bandwidth, and 32 GB memory capacity position it for large data sets and heavy rendering workloads. The 32 ray tracing cores and 128 ROPs handle complex scenes with high pixel output. The 2800 MHz boost clock provides headroom beyond its base rate. The PCIe 5.0 x16 interface supports high-throughput data exchange with the host. The dual-slot cooler and 230 W TDP reflect its performance ambitions, and the 550 W suggested PSU accommodates the additional power draw. The one 8-pin connector is required for operation. The display output suite includes both HDMI 2.1a and DisplayPort 2.1, covering modern monitors and professional displays.

The Verdict

The data indicates two GPUs at opposite ends of the Intel professional lineup. The Arc A380E delivers 4.096 TFLOPS with 6 GB memory and a 75 W TDP, suitable for basic multi-display workstations where power and space are limited. Its 2000 MHz fixed clock and single-slot profile make it a straightforward add-on for systems with modest power budgets. The 96-bit memory bus and 186.0 GB/s bandwidth are adequate for desktop composition but not for intensive compute.

The Arc Pro B70 is the clear performance leader across every metric. Its 22.94 TFLOPS, 608.0 GB/s bandwidth, and 32 GB memory make it suitable for demanding professional workloads such as large-scale rendering, simulation, or AI inference, subject to software support. The 32 ray tracing cores and 2800 MHz boost clock provide substantial compute headroom. The 5 nm process and 368 mm² die indicate a high-end implementation of the Xe2-HPG architecture. The launch MSRP of 949 USD positions it as a premium workstation component.

Users should choose the Arc A380E if their primary need is display output with minimal power draw and no external power connectors. Users should choose the Arc Pro B70 if their workloads require the full range of compute and memory capabilities, and if the system can supply 230 W and a 550 W PSU. The two cards serve different tiers of the market, and the specification data shows no overlap in their intended use cases.