Intel Arc A380M vs NVIDIA Jetson Orin Nano Super Comparison

Head-to-Head Benchmarks

The database does not contain direct head-to-head benchmark results for the Intel Arc A380M and the NVIDIA Jetson Orin Nano Super. Both entries report an average benchmark score of 0 and a percentile rank of 50 among all GPUs, which indicates that no measured performance data has been recorded for either part. Without recorded scores, a direct numeric comparison of their relative performance is not possible from the current dataset. The absence of benchmarks in the record means any claims about which one finishes first in a given workload would be unsupported speculation. The data does, however, provide substantial architectural and specification-level differences that can be analyzed to infer where each device may hold an advantage, even if exact performance deltas remain unquantified.

The Intel Arc A380M delivers a FP32 throughput of 4.096 TFLOPS, which is nearly double the 2.089 TFLOPS of the NVIDIA Jetson Orin Nano Super. In FP16 workloads, the Arc A380M reaches 8.192 TFLOPS (2:1), while the Jetson Orin Nano Super achieves 4.178 TFLOPS (2:1). These compute figures represent the clearest numerical gap between the two parts, with the Intel solution holding a roughly 1.96x advantage in both FP32 and FP16 peak rates. Pixel fill rate also favors the Arc A380M at 64.00 GPixel/s versus 16.32 GPixel/s, a 3.92x difference. Texture fill rate shows an even larger spread: 128.0 GTexel/s for Intel against 32.64 GTexel/s for NVIDIA, a 3.92x margin. These rates indicate that the Arc A380M is designed to sustain substantially higher rasterization and texture throughput per clock.

Memory bandwidth presents a different story. The Arc A380M uses 6 GB of GDDR6 on a 96-bit bus, yielding 186.0 GB/s. The Jetson Orin Nano Super uses 8 GB of LPDDR5 on a 128-bit bus, yielding 102.4 GB/s. The Intel part leads by 1.82x in raw bandwidth, but the NVIDIA part has 2 GB more capacity. The effective memory clock for the Arc A380M is 15.5 Gbps, while the Jetson Orin Nano Super runs at 6.4 Gbps effective. The GDDR6 implementation on Intel's side clearly prioritizes throughput, while the LPDDR5 on NVIDIA's side prioritizes capacity and likely lower power per access.

Where Each One Wins

The Intel Arc A380M wins decisively in raw compute throughput, fill rates, and memory bandwidth. Its FP32 performance of 4.096 TFLOPS positions it as the stronger choice for workloads that rely heavily on general-purpose shader math, such as traditional 3D rendering, pixel shading, and compute-heavy graphics effects. The 64.00 GPixel/s pixel rate and 128.0 GTexel/s texture rate suggest that the Arc A380M can handle higher resolution framebuffers and more complex texture sampling than the Jetson Orin Nano Super. The 186.0 GB/s memory bandwidth supports these high fill rates, reducing the likelihood that memory access becomes a bottleneck in bandwidth-sensitive scenes.

The NVIDIA Jetson Orin Nano Super wins in memory capacity, offering 8 GB versus 6 GB, and in power efficiency per watt. Its TDP is 25 W compared to 35 W for the Arc A380M, a 10 W difference. The Jetson Orin Nano Super also includes 32 tensor cores, a feature that the Arc A380M does not list (its tensorCores field is null). This makes the NVIDIA part the only one of the two with dedicated tensor hardware, which is significant for AI inference, deep learning, and matrix math workloads. The Jetson Orin Nano Super's IGP form factor and PCIe 4.0 x4 interface also differ from the Arc A380M's MXM Module and MXM-A (3.1) bus, which may affect integration flexibility.

In terms of physical footprint, the Jetson Orin Nano Super is specified at 70 mm by 45 mm, while the Arc A380M has no recorded dimensions. The NVIDIA module's compact size and lower TDP make it better suited for space-constrained, low-power embedded applications. The Arc A380M, by contrast, is a mobile graphics module that expects a host device to provide power and cooling, and its 35 W TDP is modest for a discrete GPU but higher than the Jetson Orin Nano Super's 25 W.

Architecture Differences

The two GPUs come from different architectural generations and process nodes. The Intel Arc A380M is built on the Xe-HPG architecture, specifically the DG2-128 chip, and belongs to the Alchemist generation (Arc 3 Mobile). It is fabricated on TSMC's 6 nm process, with 7,200 million transistors on a 157 mm² die, giving a transistor density of 45.9M per mm². The NVIDIA Jetson Orin Nano Super uses the Ampere architecture with the GA10B chip, part of the Tegra (Ampere) generation, and is fabricated on Samsung's 8 nm process. Its die size is 200 mm², but transistor count is listed as unknown, so density cannot be calculated.

The Arc A380M has 1024 shading units, 64 TMUs, and 32 ROPs, along with 8 ray tracing cores. The Jetson Orin Nano Super also has 1024 shading units but only 32 TMUs and 16 ROPs, and it lists no ray tracing cores. Instead, it has 32 tensor cores. This division of resources reflects different design priorities: Intel allocates more of its silicon to TMUs and ROPs for graphics rasterization, while NVIDIA dedicates a portion of its die to tensor operations for AI acceleration. The Arc A380M also lists 8 RT cores, which means it supports hardware-accelerated ray tracing, a feature the Jetson Orin Nano Super does not appear to provide.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API-level feature parity is present. The memory type differs fundamentally: GDDR6 on a 96-bit interface for Intel versus LPDDR5 on a 128-bit interface for NVIDIA. The GDDR6 clock runs at 1937 MHz with 15.5 Gbps effective data rate, while the LPDDR5 runs at 800 MHz with 6.4 Gbps effective. The bus width advantage goes to NVIDIA (128 bit vs 96 bit), but the bandwidth advantage goes to Intel (186.0 GB/s vs 102.4 GB/s). The Jetson Orin Nano Super's release date is 2024-12-16, nearly two years after the Arc A380M's 2023-01-23 release, and the NVIDIA part carries a launch MSRP of 249 USD, while the Intel part has no recorded launch MSRP.

The Verdict

Based strictly on the recorded data, the Intel Arc A380M is the stronger performer in raw graphics and compute throughput. Its FP32 rate of 4.096 TFLOPS doubles the Jetson Orin Nano Super's 2.089 TFLOPS, and its pixel and texture fill rates are nearly four times higher. The 186.0 GB/s memory bandwidth outpaces the NVIDIA part's 102.4 GB/s by a significant margin. For any workload that depends on shader math, rasterization, or texture processing, the Arc A380M has the clear numerical advantage.

The NVIDIA Jetson Orin Nano Super, however, offers capabilities that the Arc A380M does not list. The 32 tensor cores are unique to the NVIDIA part, and its 8 GB memory capacity exceeds the Arc A380M's 6 GB. Its 25 W TDP is lower, and its compact 70 mm by 45 mm dimensions make it a candidate for embedded or edge deployments where power and space are constrained. The 249 USD launch MSRP also appears in the record, while the Arc A380M has no launch price listed, though the analysis here does not weigh cost beyond stating the fact.

The choice between these two parts depends on the workload. The Arc A380M is positioned for graphics-intensive tasks where peak FP32, fill rate, and bandwidth matter most. The Jetson Orin Nano Super is positioned for AI inference and tensor-based computation, with its tensor cores and larger memory pool supporting model weights and activations. The data does not indicate which part would win in a head-to-head benchmark, as no such measurements exist in the database, but the architectural specifications provide a clear directional signal.

FAQ

Q: Which GPU has higher FP32 performance?

A: The Intel Arc A380M delivers 4.096 TFLOPS, while the NVIDIA Jetson Orin Nano Super delivers 2.089 TFLOPS, making the Intel part approximately 1.96x faster in FP32.

Q: Does the NVIDIA Jetson Orin Nano Super support ray tracing?

A: The record lists no ray tracing cores for the Jetson Orin Nano Super. The Intel Arc A380M lists 8 RT cores, which indicates hardware ray tracing support on the Intel side.

Q: What is the memory capacity difference?

A: The NVIDIA Jetson Orin Nano Super has 8 GB of LPDDR5, while the Intel Arc A380M has 6 GB of GDDR6. NVIDIA leads by 2 GB, but Intel leads in bandwidth at 186.0 GB/s versus 102.4 GB/s.

Q: Which GPU has tensor cores?

A: The NVIDIA Jetson Orin Nano Super lists 32 tensor cores. The Intel Arc A380M does not list any tensor cores (the field is null), so tensor acceleration is exclusive to the NVIDIA part.

Q: What are the TDP ratings?

A: The Intel Arc A380M has a TDP of 35 W, and the NVIDIA Jetson Orin Nano Super has a TDP of 25 W. The NVIDIA part is rated for lower power consumption.

Q: What is the launch MSRP of the NVIDIA Jetson Orin Nano Super?

A: The launch MSRP is 249 USD. The Intel Arc A380M has no launch MSRP recorded in the database.

Specification Differences

| Field | Intel Arc A380M | NVIDIA Jetson Orin Nano Super |

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

| Chip | DG2-128 | GA10B |

| Architecture | Xe-HPG | Ampere |

| Generation | Alchemist (Arc 3 Mobile) | Tegra (Ampere) |

| Process Node | 6 nm (TSMC) | 8 nm (Samsung) |

| Transistors | 7,200 million | unknown |

| Die Size | 157 mm² | 200 mm² |

| Transistor Density | 45.9M / mm² | null |

| Base Clock | 1550 MHz | null |

| Boost Clock | 2000 MHz | null |

| Memory Clock | 1937 MHz, 15.5 Gbps effective | 800 MHz, 6.4 Gbps effective |

| Memory Size | 6 GB | 8 GB |

| Memory Type | GDDR6 | LPDDR5 |

| Memory Bus Width | 96 bit | 128 bit |

| Memory Bandwidth | 186.0 GB/s | 102.4 GB/s |

| TMUs | 64 | 32 |

| ROPs | 32 | 16 |

| RT Cores | 8 | null |

| Tensor Cores | null | 32 |

| Pixel Rate | 64.00 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 128.0 GTexel/s | 32.64 GTexel/s |

| FP32 | 4.096 TFLOPS | 2.089 TFLOPS |

| FP16 | 8.192 TFLOPS (2:1) | 4.178 TFLOPS (2:1) |

| TDP | 35 W | 25 W |

| Slot Width | MXM Module | IGP |

| Bus Interface | MXM-A (3.1) | PCIe 4.0 x4 |

| Dimensions | null | 70 mm, 45 mm |

| Release Date | 2023-01-23 | 2024-12-16 |

| Launch MSRP | null | 249 USD |