Intel Arc B580 vs NVIDIA Jetson Orin Nano Super Comparison
Intel Arc B580
Jetson Orin Nano Super
PERFORMANCE BENCHMARKS
Analysis: Intel Arc B580 vs NVIDIA Jetson Orin Nano Super
Head-to-Head Benchmarks
The database contains no shared benchmark entries for the Intel Arc B580 and the NVIDIA Jetson Orin Nano Super. The Arc B580 has a complete set of recorded measurements across DirectX, OpenCL, Vulkan, and compute workloads, while the Jetson Orin Nano Super returns an empty benchmark array. This absence of overlapping test data makes a direct score-for-score comparison impossible.
What can be established is the Arc B580's position within the broader GPU landscape. Its average benchmark score of 23,021 places it at the 68th percentile among all GPUs tracked by the database. Its nearest recorded rivals show a remarkably tight cluster: the AMD Radeon RX 580 2048SP averages 23,061 (a 0.2% difference), the NVIDIA GeForce RTX 3080 averages 23,172 (0.7% ahead), the NVIDIA P106-100 averages 23,249 (1% ahead), and the NVIDIA GeForce RTX 2080 averages 22,895 (0.6% behind). The Arc B580 sits essentially at parity with this group, with all deltas within a single percentage point.
The Jetson Orin Nano Super, by contrast, has an average benchmark score of 0 and sits at the 50th percentile. This is not a failing score in the conventional sense; it simply reflects the absence of submitted results in the database. The module is designed for embedded and edge deployments, and its measurement profile in this database is blank. Any attempt to derive a head-to-head delta would be fabrication, and the data does not support such a claim.
The Arc B580's individual scores show substantial variation across workloads. Its PassMark G3D score of 15,748 is the strongest traditional 3D figure, while its PassMark DirectX 9 score of 183 and DirectX 11 score of 128 indicate legacy API performance. The DirectX 10 and DirectX 12 scores both sit at 76. The compute-oriented PassMark GPU Compute score of 7,729 is less than half the G3D figure, suggesting the card's compute throughput is a secondary characteristic relative to its rasterization strength. Geekbench results are notably higher: 92,821 in OpenCL and 109,672 in Vulkan. The 3DMark Steel Nomad DX12 score of 3,068 rounds out the suite.
Architecture Differences
The two devices share no architectural lineage. The Intel Arc B580 uses the BMG-G21 chip built on Xe2-HPG architecture, part of the Battlemage generation. It is fabricated on a 5 nm process at TSMC with 19,600 million transistors on a 272 mm² die, yielding a transistor density of 72.1 million per square millimeter. The NVIDIA Jetson Orin Nano Super uses the GA10B chip on Ampere architecture from the Tegra family. It is built on an 8 nm process at Samsung with a 200 mm² die size; its transistor count is not recorded in the database.
The memory subsystems differ in both capacity and type. The Arc B580 carries 12 GB of GDDR6 on a 192-bit bus with 456.0 GB/s of bandwidth. The Jetson Orin Nano Super carries 8 GB of LPDDR5 on a 128-bit bus with 102.4 GB/s of bandwidth. That is a 353.6 GB/s advantage for the Intel part, a gap that directly reflects the different design goals: a discrete desktop graphics card versus an integrated system-on-module.
Compute resources diverge sharply. The Arc B580 has 2,560 shading units, 160 texture mapping units, 80 raster output units, and 20 ray tracing cores. The Jetson Orin Nano Super has 1,024 shading units, 32 TMUs, 16 ROPs, and no dedicated ray tracing cores. Instead, it carries 32 tensor cores, a resource the Arc B580 does not list. Pixel rate for the Arc B580 is 213.6 GPixel/s against 16.32 GPixel/s for the Jetson module. Texture rate is 427.2 GTexel/s versus 32.64 GTexel/s. FP32 throughput is 13.67 TFLOPS versus 2.089 TFLOPS, and FP16 is 27.34 TFLOPS versus 4.178 TFLOPS, both at a 2:1 ratio.
Power envelopes reinforce the positioning. The Arc B580 has a 190 W TDP, a dual-slot cooler, a single 8-pin power connector, and a suggested 450 W power supply. The Jetson Orin Nano Super has a 25 W TDP and is an integrated graphics processor with no power connector listed. The Intel card measures 272 mm in length, 115 mm in height, and 45 mm in width. The NVIDIA module measures 70 mm by 45 mm with no width recorded. Both use PCIe 4.0, but the Arc B580 exposes an x8 link versus x4 on the Jetson module. Display outputs also differ: the Arc B580 provides one HDMI 2.1a and three DisplayPort 2.1 outputs, while the Jetson module's display outputs are listed as portable device dependent.
Both devices support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The release dates are close: the Arc B580 launched on December 12, 2024, and the Jetson Orin Nano Super on December 16, 2024.
The Verdict
The recorded data supports a clear separation of roles. The Intel Arc B580 is a desktop graphics card with a full benchmark profile, a 68th percentile ranking, and performance parity with the RTX 2080, RTX 3080, RX 580 2048SP, and P106-100. Its 12 GB memory configuration, 456.0 GB/s bandwidth, and 13.67 TFLOPS FP32 throughput position it for conventional rendering workloads. The 190 W TDP and dual-slot form factor assume a desktop chassis with adequate power delivery.
The NVIDIA Jetson Orin Nano Super is a different class of device. With a 25 W TDP, an 8 GB LPDDR5 configuration on a 128-bit bus, and 2.089 TFLOPS FP32, it targets embedded and edge scenarios where power and physical footprint dominate. Its 32 tensor cores and 4.178 TFLOPS FP16 throughput suggest a compute-oriented design, but the database contains no benchmark results to quantify that capability. The 50th percentile figure is a placeholder derived from an empty result set, not a measured performance rank.
For a user selecting a part for a desktop build with PCIe slots, standard power delivery, and display connectivity, the Arc B580 is the only option with documented performance data. For a user integrating compute into a compact, power-constrained system, the Jetson Orin Nano Super offers a dramatically smaller physical footprint (70 mm versus 272 mm length) and an 87% reduction in TDP, but no benchmark evidence in this database to validate expected throughput. The choice depends entirely on deployment context, and the data only fully supports one side of that decision.
FAQ
Q: Which GPU has a higher average benchmark score?
A: The Intel Arc B580 has an average benchmark score of 23,021. The NVIDIA Jetson Orin Nano Super has no recorded benchmark scores, so its average is 0 in the database.
Q: What is the memory bandwidth difference between the two?
A: The Arc B580 has 456.0 GB/s of bandwidth from 12 GB of GDDR6 on a 192-bit bus. The Jetson Orin Nano Super has 102.4 GB/s from 8 GB of LPDDR5 on a 128-bit bus.
Q: Do both support ray tracing?
A: The Arc B580 has 20 dedicated ray tracing cores. The Jetson Orin Nano Super lists no ray tracing cores in its specifications.
Q: What are the TDP figures for each device?
A: The Arc B580 is rated at 190 W with a suggested 450 W power supply. The Jetson Orin Nano Super is rated at 25 W and lists no power supply requirement.
Q: How does the Arc B580 compare to nearby GPUs in the database?
A: Its average score of 23,021 is 0.2% behind the AMD Radeon RX 580 2048SP, 0.6% behind the NVIDIA GeForce RTX 2080, 0.7% behind the NVIDIA GeForce RTX 3080, and 1% behind the NVIDIA P106-100.
Q: Which device has tensor cores?
A: The Jetson Orin Nano Super has 32 tensor cores. The Arc B580 does not list a tensor core count.
Where Each One Wins
The Intel Arc B580 wins in every measured performance category available in the database. Its FP32 throughput of 13.67 TFLOPS is more than six times the Jetson module's 2.089 TFLOPS. Its FP16 throughput of 27.34 TFLOPS is more than six times the 4.178 TFLOPS recorded for the NVIDIA part. Pixel rate favors the Arc B580 at 213.6 GPixel/s versus 16.32 GPixel/s, and texture rate favors it at 427.2 GTexel/s versus 32.64 GTexel/s. Memory bandwidth is 456.0 GB/s versus 102.4 GB/s. The Arc B580 also has more shading units (2,560 versus 1,024), more TMUs (160 versus 32), more ROPs (80 versus 16), and the only ray tracing cores in the comparison.
The Jetson Orin Nano Super wins in physical and electrical characteristics. Its 25 W TDP is 165 W lower than the Arc B580's 190 W rating. Its 70 mm length is 202 mm shorter than the 272 mm Intel card. Its integrated form factor requires no power connector, whereas the Arc B580 requires a single 8-pin connector. The Jetson module also carries 32 tensor cores, a feature absent from the Arc B580's specification sheet. The 8 nm Samsung process and 200 mm² die are smaller in area than the 5 nm TSMC process and 272 mm² die, though transistor density is not recorded for the NVIDIA part.
The Arc B580 further wins on display connectivity with one HDMI 2.1a and three DisplayPort 2.1 outputs, while the Jetson module's outputs are portable device dependent. The Arc B580's PCIe 4.0 x8 interface provides double the lane width of the Jetson module's PCIe 4.0 x4 interface.
Specification Differences
The two devices differ across nearly every specification field. The Arc B580 uses the BMG-G21 chip on Xe2-HPG architecture from the Battlemage generation, while the Jetson Orin Nano Super uses the GA10B chip on Ampere architecture from the Tegra family. Process nodes differ: 5 nm at TSMC for Intel, 8 nm at Samsung for NVIDIA. Die size is 272 mm² versus 200 mm². Transistor count is 19,600 million for the Arc B580, while the Jetson module's count is unknown.
Clock behavior differs in the database. The Arc B580 has a base and boost clock both at 2670 MHz, plus a memory clock of 2375 MHz with 19 Gbps effective. The Jetson module has no base or boost clock recorded, only a memory clock of 800 MHz with 6.4 Gbps effective.
Memory configuration differs in size, type, bus width, and bandwidth: 12 GB GDDR6 on 192-bit at 456.0 GB/s versus 8 GB LPDDR5 on 128-bit at 102.4 GB/s. Shading units are 2,560 versus 1,024. TMUs are 160 versus 32. ROPs are 80 versus 16. Ray tracing cores are 20 versus none. Tensor cores are none versus 32.
Pixel rate is 213.6 GPixel/s versus 16.32 GPixel/s. Texture rate is 427.2 GTexel/s versus 32.64 GTexel/s. FP32 is 13.67 TFLOPS versus 2.089 TFLOPS. FP16 is 27.34 TFLOPS versus 4.178 TFLOPS, both at 2:1 ratios.
TDP is 190 W versus 25 W. Slot width is dual-slot versus IGP. Power connectors are one 8-pin versus none. Suggested PSU is 450 W versus none. Bus interface is PCIe 4.0 x8 versus PCIe 4.0 x4. Display outputs are one HDMI 2.1a plus three DisplayPort 2.1 versus portable device dependent.
Dimensions differ substantially: 272 mm by 115 mm by 45 mm versus 70 mm by 45 mm with no width recorded. Both share API support for DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have a launch MSRP of 249 USD. Both are marked as active in production. Both released in December 2024, with the Arc B580 four days earlier. The Arc B580 lists Alchemist as its predecessor; the Jetson module lists no predecessor or successor.