Intel Arc B570 vs NVIDIA Jetson Orin Nano Super Comparison
Intel Arc B570
Jetson Orin Nano Super
PERFORMANCE BENCHMARKS
Analysis: Intel Arc B570 vs NVIDIA Jetson Orin Nano Super
FAQ
Q: What is the core architectural difference between the Intel Arc B570 and the NVIDIA Jetson Orin Nano Super?
A: The Intel Arc B570 uses the Xe2-HPG architecture on a 5 nm TSMC process with the BMG-G21 chip, while the NVIDIA Jetson Orin Nano Super uses the Ampere architecture on an 8 nm Samsung process with the GA10B chip.
Q: How do their memory subsystems compare?
A: The Arc B570 has 10 GB of GDDR6 on a 160-bit bus delivering 380.0 GB/s of bandwidth, whereas the Jetson Orin Nano Super has 8 GB of LPDDR5 on a 128-bit bus delivering 102.4 GB/s.
Q: Which device has higher raw compute throughput?
A: The Arc B570 delivers 11.52 TFLOPS of FP32 performance and 23.04 TFLOPS of FP16 performance, compared to 2.089 TFLOPS FP32 and 4.178 TFLOPS FP16 for the Jetson Orin Nano Super.
Q: What are the power requirements for each?
A: The Arc B570 has a TDP of 150 W and requires a 450 W suggested PSU with a single 8-pin power connector. The Jetson Orin Nano Super has a TDP of 25 W and is an integrated graphics processor with no separate power connector or suggested PSU.
Q: What is the physical size difference?
A: The Arc B570 is a dual-slot card measuring 272 mm in length and 115 mm in height, while the Jetson Orin Nano Super measures 70 mm by 45 mm as an IGP.
Q: Which device supports ray tracing?
A: The Arc B570 includes 18 dedicated RT cores, while the Jetson Orin Nano Super has no dedicated RT cores listed in the database, though it does include 32 tensor cores.
Architecture Differences
The Intel Arc B570 and NVIDIA Jetson Orin Nano Super come from different design philosophies. The Arc B570 is a discrete desktop graphics card built around the BMG-G21 chip using Intel's Xe2-HPG architecture, fabricated on a 5 nm TSMC process. The Jetson Orin Nano Super is an embedded system-on-module using the GA10B chip with NVIDIA's Ampere architecture, fabricated on an 8 nm Samsung process.
The transistor counts tell a clear story of scale. The Arc B570 packs 19,600 million transistors into a 272 mm² die, yielding a density of 72.1M transistors per mm². The Jetson Orin Nano Super has a die size of 200 mm² with an unknown transistor count and no density figure recorded in the database. The process advantage of the Arc B570 is substantial, enabling both a larger transistor budget and higher density.
Shader resources diverge sharply. The Arc B570 has 2304 shading units, 144 texture mapping units, and 80 ROPs. The Jetson Orin Nano Super has 1024 shading units, 32 TMUs, and 16 ROPs. The Arc B570 also carries 18 RT cores for hardware ray tracing, while the Jetson Orin Nano Super has no RT cores listed. Instead, the Jetson module has 32 tensor cores, which are absent from the Arc B570's specification sheet. This reflects different target workloads: raster and ray tracing for the Arc, AI inference acceleration for the Jetson.
Memory architecture differs in type and capacity. The Arc B570 uses 10 GB of GDDR6 across a 160-bit bus at 2375 MHz, producing 380.0 GB/s of bandwidth. The Jetson Orin Nano Super uses 8 GB of LPDDR5 across a 128-bit bus at 800 MHz, producing 102.4 GB/s. The nearly fourfold bandwidth advantage of the Arc B570 matters for texture-heavy and high-resolution workloads.
Clock behavior also differs. The Arc B570 has a fixed 2500 MHz base and boost clock. The Jetson Orin Nano Super has no base or boost clock figures recorded, which suggests a variable or thermally constrained operating profile typical of embedded modules. Memory clock is similarly fixed on the Arc at 2375 MHz, while the Jetson runs its LPDDR5 at 800 MHz.
The form factors reflect the intended deployment. The Arc B570 is a dual-slot PCIe 4.0 x8 card with a 1x 8-pin power connector and a 450 W suggested PSU. The Jetson Orin Nano Super is an IGP with PCIe 4.0 x4 connectivity, no power connector, and no suggested PSU, sized at 70 mm by 45 mm. Display output also differs: the Arc B570 provides 1x HDMI 2.1a and 3x DisplayPort 2.1, while the Jetson's display outputs are listed as portable device dependent.
Both support the same API generation: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The release dates sit about a month apart, with the Arc B570 launching on 2025-01-15 and the Jetson Orin Nano Super on 2024-12-16.
Head-to-Head Benchmarks
The database contains benchmark results for the Intel Arc B570 but no recorded benchmarks for the NVIDIA Jetson Orin Nano Super. The head-to-head benchmark field is empty, and the Jetson's average benchmark score is listed as 0. This makes a direct numerical comparison impossible across the standard benchmark suite.
The Arc B570's recorded data provides a reference point. Its average benchmark score is 20556, which places it at the 65th percentile of all GPUs in the database. The nearest rivals in the database are the NVIDIA GeForce RTX 3070 Mobile with an average score of 20534 (0.1% behind the Arc), the Intel Arc A750 at 20582 (0.1% ahead), the NVIDIA Quadro M4000M at 20480 (0.4% behind), and the AMD Radeon R9 M390X at 20662 (0.5% ahead). These delta percentages indicate the Arc B570 sits in a tightly packed performance band, within half a percent of all four nearest rivals.
Individual benchmark scores for the Arc B570 show its strengths. In Geekbench Vulkan it scores 96844, and in Geekbench OpenCL it scores 83514. The Passmark G3D score is 14195, while Passmark GPU Compute is 7281. The 3DMark Steel Nomad DX12 score is 2649. Legacy DirectX tests show 164 for DirectX 9, 118 for DirectX 11, 72 for DirectX 12, and 65 for DirectX 10. The Passmark G2D score is 661.
The Jetson Orin Nano Super has no benchmark entries in the database, so there are no wins to assign in either direction. The wins counter for both devices is 0. The practical interpretation is that the Arc B570 has a measured performance profile in the database, while the Jetson's performance cannot be quantified from recorded data. The Jetson's 50th percentile ranking appears to be a placeholder based on its inclusion in the database rather than a measured result, given the absence of benchmark scores.
The compute figures from the specification sheet reinforce the gap. The Arc B570's 11.52 TFLOPS FP32 output is roughly 5.5 times the Jetson's 2.089 TFLOPS. The pixel rate difference is even larger proportionally: 200.0 GPixel/s versus 16.32 GPixel/s. The texture rate comparison shows 360.0 GTexel/s against 32.64 GTexel/s. These are specification-level figures rather than benchmark results, but they indicate the scale of the performance gap in raw throughput.
The Verdict
The data supports a clear division of roles. The Intel Arc B570 is a discrete desktop graphics card with measured benchmark results, an average score of 20556, and a 65th percentile ranking among all GPUs in the database. Its nearest rivals are all within half a percent, meaning it performs in line with midrange desktop and mobile discrete GPUs from the previous generation.
The NVIDIA Jetson Orin Nano Super has no recorded benchmark scores, an average score of 0, and a 50th percentile ranking that appears unverified. Its specification sheet shows a 25 W TDP, 8 GB of LPDDR5 memory, and 2.089 TFLOPS of FP32 compute. These figures place it in an entirely different performance class.
The Arc B570 is the appropriate choice for workloads that need rasterization throughput, ray tracing, and high memory bandwidth. The Jetson Orin Nano Super is appropriate for embedded or edge deployments where the 25 W power envelope, 70 mm by 45 mm footprint, and 32 tensor cores for AI inference take priority over raw graphics performance.
Neither device should be selected based on the other's strengths. The Arc B570 cannot match the Jetson's power efficiency in a portable form factor, and the Jetson cannot approach the Arc's compute and bandwidth figures. The database records no head-to-head benchmarks, so any comparison beyond the listed specifications and the Arc's measured scores would require additional data.
Specification Differences
| Specification | Intel Arc B570 | NVIDIA Jetson Orin Nano Super |
|---|---|---|
| Chip | BMG-G21 | GA10B |
| Architecture | Xe2-HPG | Ampere |
| Generation | Battlemage (Arc 5) | Tegra (Ampere) |
| Process node | 5 nm (TSMC) | 8 nm (Samsung) |
| Transistors | 19,600 million | unknown |
| Die size | 272 mm² | 200 mm² |
| Transistor density | 72.1M / mm² | null |
| Base clock | 2500 MHz | null |
| Boost clock | 2500 MHz | null |
| Memory clock | 2375 MHz | 800 MHz |
| Memory size | 10 GB | 8 GB |
| Memory type | GDDR6 | LPDDR5 |
| Memory bus width | 160 bit | 128 bit |
| Memory bandwidth | 380.0 GB/s | 102.4 GB/s |
| Shading units | 2304 | 1024 |
| TMUs | 144 | 32 |
| ROPs | 80 | 16 |
| RT cores | 18 | null |
| Tensor cores | null | 32 |
| Pixel rate | 200.0 GPixel/s | 16.32 GPixel/s |
| Texture rate | 360.0 GTexel/s | 32.64 GTexel/s |
| FP32 | 11.52 TFLOPS | 2.089 TFLOPS |
| FP16 | 23.04 TFLOPS (2:1) | 4.178 TFLOPS (2:1) |
| TDP | 150 W | 25 W |
| Slot width | Dual-slot | IGP |
| Power connectors | 1x 8-pin | null |
| Suggested PSU | 450 W | null |
| Bus interface | PCIe 4.0 x8 | PCIe 4.0 x4 |
| Display outputs | 1x HDMI 2.1a, 3x DisplayPort 2.1 | Portable Device Dependent |
| Dimensions | 272 mm x 115 mm | 70 mm x 45 mm |
| Release date | 2025-01-15 | 2024-12-16 |
| Launch MSRP | 219 USD | 249 USD |
Where Each One Wins
The Intel Arc B570 wins in every measured graphics metric recorded in the database. Its FP32 compute of 11.52 TFLOPS is more than five times the Jetson's 2.089 TFLOPS. Its memory bandwidth of 380.0 GB/s is nearly four times the Jetson's 102.4 GB/s. The pixel rate of 200.0 GPixel/s and texture rate of 360.0 GTexel/s dwarf the Jetson's 16.32 GPixel/s and 32.64 GTexel/s. The Arc also has 18 RT cores for hardware ray tracing, which the Jetson lacks entirely.
The Arc B570's benchmark presence is another clear win. It has ten recorded benchmark scores across 3DMark, Geekbench, and Passmark suites, with an average score of 20556. The Jetson has no recorded benchmarks and an average score of 0. The Arc's 65th percentile ranking versus the Jetson's 50th percentile reflects this measured advantage, though the Jetson's percentile appears to lack supporting data.
The NVIDIA Jetson Orin Nano Super wins in power efficiency and physical footprint. Its 25 W TDP is one-sixth of the Arc B570's 150 W. It requires no power connector and no suggested PSU, while the Arc needs a 1x 8-pin connector and a 450 W PSU. The Jetson measures 70 mm by 45 mm as an IGP, versus the Arc's 272 mm by 115 mm dual-slot card. The Jetson also has 32 tensor cores, a feature absent from the Arc, which positions it for AI inference workloads despite its lower graphics throughput.
The Jetson's 8 nm Samsung process and unknown transistor count suggest a design optimized for low power rather than density. The Arc's 5 nm TSMC process and 72.1M transistors per mm² density show a design focused on maximum throughput within a desktop power envelope.
The API support is identical, with both listing DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This means software compatibility is not a differentiator. The bus interface differs, with the Arc using PCIe 4.0 x8 and the Jetson using PCIe 4.0 x4, which matters for bandwidth-sensitive workloads in systems that support both.
The launch MSRP of the Arc B570 is 219 USD, while the Jetson Orin Nano Super has a launch MSRP of 249 USD. The release dates are close, with the Jetson on 2024-12-16 and the Arc on 2025-01-15.
The use-case split follows the data. Desktop gaming, ray-traced rendering, and high-bandwidth compute workloads belong to the Arc B570. Edge AI inference, portable embedded systems, and low-power deployments belong to the Jetson Orin Nano Super. The database records no head-to-head benchmark results, so the decision rests on the specification sheet and the Arc's measured scores.