Intel Arc A380 vs NVIDIA GeForce 945M Comparison

The Intel Arc A380 and NVIDIA GeForce 945M represent two distinct eras of mobile and small-form-factor graphics, separated by nearly seven years of architectural evolution. The data is unequivocal: in the only directly comparable benchmark, the Arc A380 outperforms the 945M by a staggering 372%, scoring 38,224 to 8,099 in Geekbench OpenCL. This is not a close contest; it is a generational gap made manifest in silicon. The A380’s average benchmark score of 8,558 places it in the 44th percentile of all GPUs, while the 945M’s 8,099 sits at the 42nd percentile, but the raw compute delta tells the real story. For any workload leveraging general-purpose compute, the Arc A380 is in a different league entirely.

Where Each One Wins

The Arc A380 wins every measurable category in this comparison, but the nature of its victories varies. In raw compute throughput, the win is absolute and overwhelming. The Geekbench OpenCL result shows a 372% advantage, which translates to the A380 delivering nearly five times the compute performance. This makes the A380 the clear choice for any application that leverages GPU acceleration, from content creation to scientific computing.

The 945M, conversely, has no benchmark victories in the provided data. Its only score, the 8,099 in Geekbench OpenCL, is a baseline that the A380 crushes. However, the 945M’s architecture does have a theoretical advantage in compatibility with older software. Its DirectX 12 (11_0) support means it can run legacy titles that might not be optimized for the A380’s DirectX 12 Ultimate feature set. For purely legacy 32-bit applications or very old game engines, the 945M’s simpler feature set could be seen as a benefit, though the performance gap would likely still favor the A380 in any compute-heavy task.

The A380 also wins on feature support. It supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, alongside hardware ray tracing via its 8 RT cores. The 945M supports DirectX 12 (11_0), Vulkan 1.4, and OpenGL 4.6, but lacks ray tracing entirely. For modern gaming and future-proofing, the A380 is the only viable option.

Architecture Differences

The architectural gap between these two GPUs is vast. The Arc A380 is built on Intel’s Xe-HPG architecture, specifically the DG2-128 chip, fabricated on a 6 nm process at TSMC. This allows for 7,200 million transistors packed into a 157 mm² die, resulting in a transistor density of 45.9 million per square millimeter. The 945M, in contrast, uses NVIDIA’s Maxwell architecture with the GM107 chip, built on a dated 28 nm process. It houses just 1,870 million transistors on a 148 mm² die, yielding a density of only 12.6 million per square millimeter.

The core configurations differ drastically. The A380 has 1,024 shading units, 64 texture mapping units (TMUs), and 32 raster operation pipelines (ROPs), plus 8 dedicated ray tracing cores. The 945M has 640 shading units, 40 TMUs, and 16 ROPs, with no ray tracing hardware. This translates to a theoretical pixel rate of 65.60 GPixel/s and a texture rate of 131.2 GTexel/s for the A380, versus 16.32 GPixel/s and 40.80 GTexel/s for the 945M. The FP32 compute is equally lopsided: 4.198 TFLOPS for the A380 versus 1,305.6 GFLOPS for the 945M.

Memory architecture is another major differentiator. The A380 uses 6 GB of GDDR6 on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The 945M uses just 2 GB of DDR3 on a 128-bit bus, with a paltry 28.80 GB/s of bandwidth. The A380’s memory is faster in both type and effective speed (15.5 Gbps effective versus 1,800 Mbps effective), making it superior for texture-heavy workloads and higher resolutions.

FAQ

Q: Is the Intel Arc A380 faster than the NVIDIA GeForce 945M in all benchmarks?

A: Yes. In the only head-to-head benchmark available (Geekbench OpenCL), the A380 scores 38,224 versus 8,099 for the 945M, a 372% advantage. The A380 also has a higher average benchmark score of 8,558 versus 8,099.

Q: Which GPU has better memory specifications?

A: The Arc A380 is vastly superior. It has 6 GB of GDDR6 memory with a bandwidth of 186.0 GB/s, while the 945M has 2 GB of DDR3 with only 28.80 GB/s of bandwidth.

Q: Does the NVIDIA GeForce 945M support ray tracing?

A: No. The 945M has no RT cores (null in the specification). The Arc A380 has 8 RT cores and supports DirectX 12 Ultimate (12_2), which includes ray tracing features.

Q: What is the process node difference between these two GPUs?

A: The Arc A380 is built on a 6 nm process at TSMC, while the 945M uses a 28 nm process at TSMC. This is a major factor in the A380’s efficiency and performance advantage.

Q: Are these GPUs still in production?

A: No. Both are listed as "End-of-life" production status. The A380 was released on 2022-06-13, and the 945M on 2015-10-26.

Q: Which GPU has a higher transistor count?

A: The Arc A380 has 7,200 million transistors, while the 945M has 1,870 million. The A380 also has a higher transistor density at 45.9M per mm² versus 12.6M per mm².

Specification Differences

| Field | Intel Arc A380 | NVIDIA GeForce 945M |

| :--- | :--- | :--- |

| Architecture | Xe-HPG | Maxwell |

| Process Node | 6 nm | 28 nm |

| Transistors | 7,200 million | 1,870 million |

| Die Size | 157 mm² | 148 mm² |

| Base Clock | 2000 MHz | 928 MHz |

| Boost Clock | 2050 MHz | 1020 MHz |

| Memory Clock | 1937 MHz (15.5 Gbps effective) | 900 MHz (1800 Mbps effective) |

| Memory Size | 6 GB | 2 GB |

| Memory Type | GDDR6 | DDR3 |

| Memory Bus | 96 bit | 128 bit |

| Memory Bandwidth | 186.0 GB/s | 28.80 GB/s |

| Shading Units | 1024 | 640 |

| TMUs | 64 | 40 |

| ROPs | 32 | 16 |

| RT Cores | 8 | null |

| Pixel Rate | 65.60 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 131.2 GTexel/s | 40.80 GTexel/s |

| FP32 Performance | 4.198 TFLOPS | 1,305.6 GFLOPS |

| TDP | 75 W | 75 W |

| Slot Width | Dual-slot | MXM Module |

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

| Bus Interface | PCIe 4.0 x8 | MXM-B (3.0) |

| DirectX Support | 12 Ultimate (12_2) | 12 (11_0) |

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and it is a rout. The Intel Arc A380 scores 38,224, while the NVIDIA GeForce 945M scores 8,099. This represents a deltaPct of 372% in favor of the A380. To put this in perspective, the A380’s score is more than four and a half times higher, indicating that compute-heavy tasks like physics simulations, video encoding, and machine learning inference will be dramatically faster on the A380.

Looking at the broader benchmark landscape, the A380’s average score of 8,558 is 5.7% higher than the 945M’s 8,099. While the 945M sits near its rivals (NVIDIA GRID K2 at 8,080, AMD Radeon R9 M360 at 8,129), the A380 sits in a slightly higher tier, competing with the AMD FirePro W5170M (8,595) and NVIDIA GeForce MX330 (8,458). The A380’s individual benchmark results are consistently strong: it scores 6,252 in Passmark G3D, 2,762 in Passmark GPU Compute, and 808 in 3DMark Steel Nomad DX12. The 945M has no comparable results in these tests, further highlighting its compute deficit. The A380’s advantage is not marginal; it is a wholesale generational leap.

The Verdict

The data leads to a single, unambiguous conclusion: the Intel Arc A380 is the superior GPU in every measurable way. Its 372% lead in OpenCL compute, combined with a 6 GB GDDR6 memory subsystem that offers nearly 6.5 times the bandwidth of the 945M’s 2 GB DDR3, makes it the only rational choice for any modern workload. The A380’s 6 nm process node and 7,200 million transistors give it a density advantage that the 28 nm Maxwell chip cannot hope to match.

Who should pick the 945M? Strictly speaking, only those with a legacy MXM-B (3.0) slot who cannot upgrade their hardware. Its 75 W TDP matches the A380, but its 2015 release date and lack of ray tracing cores, tensor cores, and modern memory mean it is a relic. The A380’s DirectX 12 Ultimate support and RT cores make it future-proof for gaming, while its FP32 performance of 4.198 TFLOPS is over three times that of the 945M.

The verdict is simple: for any user choosing between these two, the Intel Arc A380 wins on compute, memory, features, and architecture. The 945M’s only virtue is its existence in older portable devices. The benchmark data does not support any other conclusion. The A380 is the decisive winner.