NVIDIA Jetson Orin Nano Super vs NVIDIA RTX A400 Comparison

Head-to-Head Benchmarks

The data shows no direct head-to-head benchmark results are recorded for the NVIDIA Jetson Orin Nano Super, meaning the database has no comparative scores for this unit. What is available is the complete benchmark profile for the NVIDIA RTX A400, which allows for a detailed analysis of that card's standing relative to its nearest rivals.

The RTX A400 delivers an average benchmark score of 6078. That places it at the 35th percentile of all GPUs in the database. The nearest rival by score is the NVIDIA GeForce MX230, which averages 6077, a delta of 0 percent. The Quadro P2000 sits just behind at 6049, a 0.5 percent deficit. The Intel Iris Pro Graphics 6200 scores 6117, which is 0.6 percent ahead of the A400. The AMD Radeon 760M scores 6019, putting it 1 percent behind.

Breaking the A400's results down by individual test, the strongest performance comes from Geekbench OpenCL with a score of 22844. Geekbench Vulkan follows at 22237. These are the two highest recorded scores for the card and indicate strong compute and graphics API performance in those workloads. The Passmark G3D score is 5983, which aligns closely with the average benchmark score. Passmark GPU Compute returns 2557, a notably lower figure that suggests the card's compute throughput is weaker than its 3D graphics performance in that suite.

The older DirectX tests show a different picture. Passmark DirectX 9 returns 87, DirectX 11 returns 37, DirectX 10 returns 32, and DirectX 12 returns 27. These are low scores relative to the G3D result, and the trend is downward as the API generation advances. The DirectX 12 score of 27 is the lowest of the set, which is unusual for a modern Ampere-based card. The 2D test, Passmark G2D, returns 899.

What stands out is the delta between the Geekbench scores and the Passmark scores. The OpenCL score of 22844 is roughly 3.8 times the G3D score of 5983, while the Vulkan score of 22237 is similarly elevated. This indicates the card's raw compute capability is significantly stronger than its rasterized gaming performance as measured by Passmark. The nearest rivals all cluster within 1 percent of the A400's average score, meaning the card sits in a tightly competitive band where no single rival has a decisive lead.

Architecture Differences

Both the NVIDIA Jetson Orin Nano Super and the NVIDIA RTX A400 use the Ampere architecture, and both are manufactured on an 8 nm process at Samsung. The die size for each is 200 mm², making them physically identical in that respect. The transistor counts differ, however. The RTX A400 lists 8,700 million transistors, giving it a transistor density of 43.5 million per mm². The Jetson Orin Nano Super lists its transistor count as unknown in the database.

The chip names differ. The Jetson uses the GA10B, while the RTX A400 uses the GA107. The Jetson belongs to the Tegra (Ampere) generation, whereas the RTX A400 is part of the Workstation Ampere (Ax000) lineup. The Jetson's predecessor and successor fields are empty, while the RTX A400 lists its predecessor as Quadro Turing and its successor as Workstation Ada.

Core counts show clear separation. The Jetson Orin Nano Super has 1024 shading units, 32 texture mapping units, and 16 raster output units. It includes 32 tensor cores. The RTX A400 has 768 shading units, 24 TMUs, and 16 ROPs. It includes 6 ray tracing cores and 24 tensor cores. The Jetson has no ray tracing cores listed, while the A400 does.

The floating-point figures differ by workload. The Jetson delivers 2.089 TFLOPS in FP32 and 4.178 TFLOPS in FP16, the latter at a 2:1 ratio. The RTX A400 delivers 2.706 TFLOPS in FP32 and 2.706 TFLOPS in FP16 at a 1:1 ratio. That means the A400 has higher FP32 throughput, but the Jetson has higher FP16 throughput when the 2:1 ratio is active.

Pixel and texture rates also differ. The Jetson has a pixel rate of 16.32 GPixel/s and a texture rate of 32.64 GTexel/s. The RTX A400 has a pixel rate of 28.19 GPixel/s and a texture rate of 42.29 GTexel/s. The A400 is ahead in both categories.

Memory architecture is distinct. The Jetson uses 8 GB of LPDDR5 on a 128-bit bus, with 102.4 GB/s of bandwidth. The RTX A400 uses 4 GB of GDDR6 on a 64-bit bus, with 96.00 GB/s of bandwidth. The Jetson has twice the memory capacity and a wider bus, while the A400's GDDR6 runs at a higher effective speed of 12 Gbps versus the Jetson's 6.4 Gbps.

The power envelopes are far apart. The Jetson has a TDP of 25 W, while the RTX A400 has a TDP of 50 W. The Jetson is an integrated graphics processor with a slot width listed as IGP, while the RTX A400 is a single-slot card with no power connectors and a suggested PSU of 250 W.

FAQ

Q: Which GPU has higher FP32 performance?

A: The NVIDIA RTX A400 delivers 2.706 TFLOPS in FP32, while the NVIDIA Jetson Orin Nano Super delivers 2.089 TFLOPS. The A400 is ahead by roughly 0.6 TFLOPS.

Q: How does memory capacity compare between the two?

A: The Jetson Orin Nano Super has 8 GB of LPDDR5 on a 128-bit bus, while the RTX A400 has 4 GB of GDDR6 on a 64-bit bus. The Jetson has double the capacity and double the bus width.

Q: Which card has more shading units?

A: The Jetson Orin Nano Super has 1024 shading units, compared to 768 on the RTX A400. The Jetson also has 32 TMUs versus 24, while both have 16 ROPs.

Q: Does the RTX A400 support ray tracing?

A: Yes, the RTX A400 includes 6 ray tracing cores. The Jetson Orin Nano Super lists no ray tracing cores in the database.

Q: What is the power draw difference?

A: The Jetson Orin Nano Super has a TDP of 25 W, while the RTX A400 has a TDP of 50 W. The A400 uses twice the power budget.

Q: How do the tensor core counts compare?

A: The Jetson Orin Nano Super has 32 tensor cores, while the RTX A400 has 24 tensor cores. The Jetson has a higher count in this category.

Specification Differences

The following fields differ between the two products:

• Chip: GA10B (Jetson) versus GA107 (RTX A400)
• Generation: Tegra (Ampere) versus Workstation Ampere (Ax000)
• Transistors: Unknown (Jetson) versus 8,700 million (RTX A400)
• Transistor Density: Not listed (Jetson) versus 43.5M / mm² (RTX A400)
• Base Clock: Not listed (Jetson) versus 1417 MHz (RTX A400)
• Boost Clock: Not listed (Jetson) versus 1762 MHz (RTX A400)
• Memory Clock: 800 MHz 6.4 Gbps effective (Jetson) versus 1500 MHz 12 Gbps effective (RTX A400)
• Memory Size: 8 GB (Jetson) versus 4 GB (RTX A400)
• Memory Type: LPDDR5 (Jetson) versus GDDR6 (RTX A400)
• Memory Bus Width: 128 bit (Jetson) versus 64 bit (RTX A400)
• Memory Bandwidth: 102.4 GB/s (Jetson) versus 96.00 GB/s (RTX A400)
• Shading Units: 1024 (Jetson) versus 768 (RTX A400)
• TMUs: 32 (Jetson) versus 24 (RTX A400)
• RT Cores: Not listed (Jetson) versus 6 (RTX A400)
• Tensor Cores: 32 (Jetson) versus 24 (RTX A400)
• Pixel Rate: 16.32 GPixel/s (Jetson) versus 28.19 GPixel/s (RTX A400)
• Texture Rate: 32.64 GTexel/s (Jetson) versus 42.29 GTexel/s (RTX A400)
• FP32: 2.089 TFLOPS (Jetson) versus 2.706 TFLOPS (RTX A400)
• FP16: 4.178 TFLOPS 2:1 (Jetson) versus 2.706 TFLOPS 1:1 (RTX A400)
• TDP: 25 W (Jetson) versus 50 W (RTX A400)
• Slot Width: IGP (Jetson) versus Single-slot (RTX A400)
• Power Connectors: Not listed (Jetson) versus None (RTX A400)
• Suggested PSU: Not listed (Jetson) versus 250 W (RTX A400)
• Bus Interface: PCIe 4.0 x4 (Jetson) versus PCIe 4.0 x8 (RTX A400)
• Display Outputs: Portable Device Dependent (Jetson) versus 4x mini-DisplayPort 1.4a (RTX A400)
• Dimensions: 70 mm length, 45 mm height (Jetson) versus 163 mm length, 69 mm height (RTX A400)
• Release Date: 2024-12-16 (Jetson) versus 2024-04-15 (RTX A400)
• Launch MSRP: 249 USD (Jetson) versus not listed (RTX A400)
• Predecessor: Not listed (Jetson) versus Quadro Turing (RTX A400)
• Successor: Not listed (Jetson) versus Workstation Ada (RTX A400)

Fields that match include manufacturer (NVIDIA), architecture (Ampere), process node (8 nm), foundry (Samsung), die size (200 mm²), ROPs (16), API support (DirectX 12 Ultimate 12_2, OpenGL 4.6, Vulkan 1.4), and production status (Active).

Where Each One Wins

The NVIDIA Jetson Orin Nano Super wins in memory capacity with 8 GB versus 4 GB, and in memory bandwidth with 102.4 GB/s versus 96.00 GB/s. It also has a wider memory bus at 128 bit versus 64 bit. The Jetson has more shading units (1024 versus 768), more TMUs (32 versus 24), and more tensor cores (32 versus 24). Its FP16 throughput of 4.178 TFLOPS is substantially higher than the RTX A400's 2.706 TFLOPS, making it the stronger choice for FP16 workloads. The Jetson's power draw of 25 W is half that of the RTX A400, and its compact 70 mm by 45 mm form factor suits embedded or portable applications. The database lists no benchmark scores for the Jetson, so its percentile ranking and average score are not recorded.

The NVIDIA RTX A400 wins in FP32 performance with 2.706 TFLOPS versus 2.089 TFLOPS. It also leads in pixel rate (28.19 GPixel/s versus 16.32 GPixel/s) and texture rate (42.29 GTexel/s versus 32.64 GTexel/s). The A400 includes 6 ray tracing cores, which the Jetson lacks entirely. Its memory clock runs at 12 Gbps effective versus 6.4 Gbps, and it has explicit base and boost clocks of 1417 MHz and 1762 MHz. The A400's PCIe 4.0 x8 interface is wider than the Jetson's PCIe 4.0 x4. The card provides 4x mini-DisplayPort 1.4a outputs, while the Jetson's display outputs are dependent on the portable device. The A400 has recorded benchmark data: an average score of 6078, a 35th percentile standing, and a Geekbench OpenCL score of 22844.

The recorded data indicates the RTX A400 is the more capable desktop workstation card for rasterization and FP32 compute, while the Jetson Orin Nano Super is optimized for low-power, high-memory-capacity embedded use with superior FP16 throughput and tensor core count. The two devices target different physical and electrical environments, with the Jetson operating at 25 W as an IGP and the A400 operating at 50 W as a single-slot PCIe card.