Intel Arc Graphics 24EU vs NVIDIA Jetson Orin Nano Super Comparison

Where Each One Wins

The recorded data splits these two processors into very different roles. The Intel Arc Graphics 24EU holds a single benchmark result: a 3DMark Steel Nomad DX12 score of 733 points. That places it in the 3rd percentile of all GPUs in the database. The NVIDIA Jetson Orin Nano Super has no benchmark scores recorded, but it sits in the 50th percentile of all GPUs, a far stronger relative position. In direct head-to-head tests, neither part records a win, as no shared benchmark data exists between them.

For the Intel part, its single score of 733 puts it level with the Intel Arc Graphics 32EU and Intel Arc Graphics 64EU, both also at 733, and only 1.4% ahead of the AMD Radeon HD 6470M at 723. The NVIDIA GeForce GT 415M scores 751, which is 2.4% higher than the Intel Arc 24EU. This means the Intel solution lands in a narrow performance band among older discrete mobile GPUs, despite being an integrated graphics block on a desktop processor.

The NVIDIA Jetson Orin Nano Super, by contrast, is a purpose-built edge computing module. Its 50th percentile ranking suggests it outperforms half of the GPUs in the database, a much stronger standing than the Intel part's 3rd percentile. The Jetson carries 8 GB of LPDDR5 memory on a 128-bit bus with 102.4 GB/s bandwidth, whereas the Intel Arc 24EU uses system shared memory with bandwidth described only as system dependent. That memory arrangement alone gives the Jetson a structural advantage for workloads that depend on memory throughput.

For use-case separation, the Intel Arc Graphics 24EU appears suited to lightweight desktop graphics, media playback, and basic 3D acceleration, given its low shading unit count and integrated nature. The Jetson Orin Nano Super, with its tensor cores and dedicated memory, targets embedded AI inference, robotics, and portable devices. The benchmark data shows the Intel part has at least one measurable graphics result, while the Jetson has none in the database, so comparisons must rely on architectural and memory differences rather than direct test scores.

Architecture Differences

The two parts come from different manufacturers, process nodes, and design philosophies. Intel builds the Arc Graphics 24EU on a 3 nm process at TSMC, using the Xe-LPG architecture within the Arrow Lake-S chip. The chip contains 17,800 million transistors on a 243 mm² die, yielding a transistor density of 73.3 million per square millimeter. NVIDIA builds the Jetson Orin Nano Super on an 8 nm process at Samsung, using the Ampere architecture with the GA10B chip. Its die measures 200 mm², though transistor count is listed as unknown.

Clock behavior differs substantially. The Intel part has a base clock of 300 MHz and a boost clock of 2000 MHz. The NVIDIA module lists no base or boost clocks in the database, only a memory clock of 800 MHz with 6.4 Gbps effective transfer. Memory configurations are starkly different: Intel uses system shared memory with no fixed size or type, while NVIDIA uses 8 GB of LPDDR5 on a 128-bit bus with 102.4 GB/s bandwidth.

Execution resources favor the NVIDIA part significantly. The Jetson Orin Nano Super contains 1024 shading units, 32 texture mapping units, 16 raster output units, and 32 tensor cores. The Intel Arc 24EU has 192 shading units, 12 TMUs, 6 ROPs, and no tensor cores. Pixel rate for the Jetson is 16.32 GPixel/s versus 12.00 GPixel/s for Intel. Texture rate is 32.64 GTexel/s versus 24.00 GTexel/s. Floating point performance shows a clear gap: the Jetson delivers 2.089 TFLOPS FP32 and 4.178 TFLOPS FP16, while the Intel part manages 768.0 GFLOPS FP32 and 1.536 TFLOPS FP16. That puts the Jetson at roughly 2.7 times the FP32 throughput of the Intel part.

Power envelopes also diverge. The Intel Arc 24EU carries a 65 W TDP, while the Jetson Orin Nano Super lists a 25 W TDP. Both are integrated form factors, with the Intel part as an IGP on the CPU and the Jetson as a portable device module measuring 70 mm by 45 mm. The Intel part uses a Ring Bus interface, while the Jetson uses PCIe 4.0 x4. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical.

The Verdict

The data directs each processor to a different audience. The Intel Arc Graphics 24EU, with its 733-point Steel Nomad score and 3rd percentile ranking, serves as a basic integrated graphics solution for everyday desktop tasks. It matches the Intel Arc Graphics 32EU and 64EU exactly in score, suggesting that within this specific workload, those extra execution units do not translate to a measurable advantage. It trails the NVIDIA GeForce GT 415M by 2.4% and leads the AMD Radeon HD 6470M by 1.4%. For a desktop user who needs display output and light 3D acceleration without a discrete card, this part is adequate.

The NVIDIA Jetson Orin Nano Super, despite having no benchmark scores, holds a 50th percentile position among all GPUs. Its 1024 shading units, 32 tensor cores, and dedicated 8 GB LPDDR5 memory with 102.4 GB/s bandwidth make it a substantially more capable compute device on paper. The 25 W TDP, compared to 65 W for the Intel part, indicates higher efficiency per watt for the NVIDIA module. Its compact 70 mm by 45 mm dimensions and portable device display outputs point to embedded and edge applications rather than desktop graphics.

Who should pick which comes down to the workload. The Intel Arc 24EU fits a desktop PC where integrated graphics are a convenience, not a primary feature. The Jetson Orin Nano Super fits a developer building an AI inference box, a robot, or a portable device that needs serious compute in a small package. The launch MSRP of the Jetson is 249 USD, and the Intel part has no listed launch MSRP. Neither part has head-to-head benchmark data, so the decision rests on the architectural advantages documented above.

FAQ

Q: How does the Intel Arc Graphics 24EU compare to the NVIDIA Jetson Orin Nano Super in raw compute?

A: The Jetson delivers 2.089 TFLOPS FP32 and 4.178 TFLOPS FP16, while the Intel Arc 24EU delivers 768.0 GFLOPS FP32 and 1.536 TFLOPS FP16. The Jetson has roughly 2.7 times the FP32 throughput.

Q: Which processor has more memory bandwidth?

A: The Jetson Orin Nano Super uses 8 GB of LPDDR5 on a 128-bit bus with 102.4 GB/s bandwidth. The Intel Arc 24EU uses system shared memory with system dependent bandwidth, so no fixed figure exists in the database.

Q: Are there any benchmark scores for the NVIDIA Jetson Orin Nano Super?

A: No. The database lists no benchmark results for the Jetson, though it holds a 50th percentile ranking among all GPUs. The Intel Arc 24EU has one result, a 733 score in 3DMark Steel Nomad DX12.

Q: How does the Intel Arc 24EU rank against its nearest rivals?

A: It matches the Intel Arc Graphics 32EU and Intel Arc Graphics 64EU at 733 points. It is 1.4% ahead of the AMD Radeon HD 6470M at 723 and 2.4% behind the NVIDIA GeForce GT 415M at 751.

Q: What are the power requirements for each part?

A: The Intel Arc 24EU has a 65 W TDP. The NVIDIA Jetson Orin Nano Super has a 25 W TDP. Both are integrated, with no external power connectors listed.

Q: Do both parts support the same graphics APIs?

A: Yes. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Intel part uses the Xe-LPG architecture, and the Jetson uses Ampere.

Head-to-Head Benchmarks

No direct head-to-head benchmark results exist in the database for these two parts. The Intel Arc 24EU has one recorded test: 3DMark Steel Nomad DX12 at 733 points. The Jetson Orin Nano Super has zero recorded tests. Without shared workloads, a direct performance comparison relies on the architectural specifications.

The closest reference points for the Intel part come from its nearest rivals. The Intel Arc Graphics 32EU and 64EU both score 733, identical to the 24EU. That is a surprising result: it indicates that in this particular DX12 test, adding shading units from 192 to whatever the 32EU and 64EU carry does not change the score. The AMD Radeon HD 6470M scores 723, which is 1.4% lower. The NVIDIA GeForce GT 415M scores 751, which is 2.4% higher. These margins are small, placing the Intel Arc 24EU in a tight cluster with older discrete mobile GPUs.

The Jetson's 50th percentile ranking, while lacking a score, implies it outperforms many discrete GPUs in the database. Its 1024 shading units and 32 tensor cores suggest compute capability far beyond the Intel part's 192 shading units. The Jetson's pixel rate of 16.32 GPixel/s and texture rate of 32.64 GTexel/s both exceed the Intel part's 12.00 GPixel/s and 24.00 GTexel/s. The Jetson also has 16 ROPs versus 6 on the Intel part, which affects fill-rate-bound workloads.

Memory is another decisive factor. The Jetson's 102.4 GB/s bandwidth versus the Intel part's system dependent bandwidth means the NVIDIA module can feed its shaders and tensor cores more consistently. For FP16 workloads, the Jetson's 4.178 TFLOPS dwarfs the Intel part's 1.536 TFLOPS, a 2.7 times advantage. This matters for AI inference and machine learning tasks where FP16 is common.

The Intel part's only advantage in the data is its boost clock of 2000 MHz, but that clock operates on far fewer execution units. The Jetson lists no boost clock, so a clock-for-clock comparison is impossible. The Intel part also has a smaller die at 243 mm² versus 200 mm², though it packs more transistors at 17,800 million versus unknown for NVIDIA, due to the denser 3 nm process.

Specification Differences

The table below highlights only the fields where the two parts differ, based on the recorded data.

| Specification | Intel Arc Graphics 24EU | NVIDIA Jetson Orin Nano Super |

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

| Manufacturer | Intel | NVIDIA |

| Chip | Arrow Lake-S | GA10B |

| Architecture | Xe-LPG | Ampere |

| Generation | Arc Graphics (Arrow Lake) | Tegra (Ampere) |

| Process Node | 3 nm | 8 nm |

| Foundry | TSMC | Samsung |

| Transistors | 17,800 million | unknown |

| Die Size | 243 mm² | 200 mm² |

| Transistor Density | 73.3M / mm² | null |

| Base Clock | 300 MHz | null |

| Boost Clock | 2000 MHz | null |

| Memory Clock | System Shared | 800 MHz 6.4 Gbps effective |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | LPDDR5 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 102.4 GB/s |

| Shading Units | 192 | 1024 |

| TMUs | 12 | 32 |

| ROPs | 6 | 16 |

| Tensor Cores | null | 32 |

| Pixel Rate | 12.00 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 24.00 GTexel/s | 32.64 GTexel/s |

| FP32 | 768.0 GFLOPS | 2.089 TFLOPS |

| FP16 | 1.536 TFLOPS (2:1) | 4.178 TFLOPS (2:1) |

| TDP | 65 W | 25 W |

| Bus Interface | Ring Bus | PCIe 4.0 x4 |

| Display Outputs | Motherboard Dependent | Portable Device Dependent |

| Dimensions | null | 70 mm 2.8 inches by 45 mm 1.8 inches |

| Release Date | 2024-10-23 | 2024-12-16 |

| Launch MSRP | null | 249 USD |

| Benchmark Score | 733 | null |

| Percentile vs All GPUs | 3 | 50 |

| Avg Benchmark Score | 733 | 0 |

Both parts share identical API support, with DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both are integrated form factors with no power connectors and no suggested PSU. The Intel part has no RT cores or tensor cores listed, while the Jetson has 32 tensor cores and no RT cores. The production status for both is Active. The Intel part has a predecessor in HD Graphics, while the Jetson has none listed. The Intel part has no successor, and neither does the Jetson.