8x AI performance: a giant leap forward for Edge AI compare to Jetson AGX Xavier
? Exciting news! NVIDIA Jetson AGX Orin Developer Kit is available to at Seeed now for pre-order, be the first one get AGX Orin Dev kit now!
Jetson AGX Orin module is compact with the same form factor at 100mm x 87mm and same pin compatibility as Jetson AGX Xavier, however, AGX Orin delivers 8 times of the AI performance up to 275 TOPS. Besides AI performance, let’s take a look at all improvements in GPU, CPU, Memory & Storage, Video Codecs, PVA & VIC and I/Os!
|AGX Xavier||AGX Orin|
|GPU||512-core Volta GPU with 64 Tensor Cores||NVIDIA Ampere architecture |
with 2048 NVIDIA® CUDA® cores and
64 Tensor Cores
|CPU||8-core ARM |
v8.2 64-bit CPU,8MB L2 + 4MB L3
|12-core Arm® Cortex®-A78AE |
v8.2 64-bit CPU3MB L2 + 6MB L3
|Memory||32GB 256-Bit LPDDR4x137GB/s||32GB 256-bit LPDDR5204.8 GB/s|
|Storage||32GB eMMC 5.1||64GB eMMC 5.1|
|DL Accelerator||2x NVDLA Engines||2x NVDLA v2.0|
|Vision Accelerator||7-way VLIW Vision Processor||PVA v2.0|
4x 4K @ 60 (HEVC)
8x 4K @ 30 (HEVC)
16x 1080p @ 60 (HEVC)
32x 1080p @ 30 (HEVC)
|2x 4K604x 4K30|
16x 1080p30 (H.265)
2x 8K @ 30 (HEVC)
6x 4K @ 60 (HEVC)
12x 4K @ 30 (HEVC)
26x 1080p @ 60 (HEVC)
52x 1080p @ 30 (HEVC)
30x 1080p @ 30 (H.264)
24x 1080p30 (H.265)
|Size||100.0 mm x 87.0 mm||100.0 mm × 87.0 mm|
NVDIA Ampere GPU: 3rd Generation Tensor Cores, Sparsity, Streaming Multiprocessor
Jetson AGX Orin contains an integrated Ampere GPU, providing a total of 2048 CUDA cores and 64 Tensor cores with up to 131 Sparse TOPs of INT8 Tensor compute, and up to 4.096 FP32 TFLOPs of CUDA compute. New Tensor Float 32 (TF32) precision provides up to 5X the training throughput over the previous generation to accelerate AI and data science model training without requiring any code changes.
Ampere GPU also brings support for sparsity, a fine-grained compute structure that doubles throughput and reduces memory usage. New Sparsity feature can take advantage of fine-grained structured sparsity in deep learning networks to double the throughput of Tensor Core operations over the prior generation Turing Tensor Cores.
The NVIDIA Ampere GPU introduces a new design for the Streaming Multiprocessor (SM) that can improve performance per watt and performance per area, along with supporting 3rd generation tensor cores and TensorRT cores. Ampere GPUs improve on the previous NVIDIA Turing™ generation; and are software compatible so that the same APIs are used.
CPU: 1.7x the performance in advance
The biggest change in the CPU in Jetson AGX Orin is replacing the NVIDIA Carmel CPU
clusters with the Arm Cortex-A78AE. Jetson AGX Orin CPU is made up of 12 CPU Cores, enabling 1.7x the performance compared to the 8-core NVIDIA Carmel CPU on Jetson AGX Xavier.
Each core consists of 64KB Instruction L1 Cache and 64KB Data Cache, and 256 KB of L2 Cache. Same as Jetson AGX Xavier, each cluster consists of 2MB L3 Cache. The max frequency supported on the CPU is 2 GHz.
Memory & Storage: 1.4x the memory bandwidth and 2x the storage of the Jetson AGX Xavier
Jetson AGX Orin brings support for 1.4x the memory bandwidth and 2x the storage of the
Jetson AGX Xavier, enabling 32 GB of 256-bit LPDDR5 and 64 GB of eMMC. The DRAM supports a max
clock speed of 3200 MHz, with 6400 Gbps per pin, enabling 204.8 GB/s of memory bandwidth.
Multi-Standard Video Encoder & Decoder and a JPEG processing block (NVJPEG)
Jetson AGX Orin contains a Multi-Standard Video Encoder (NVENC), a Multi-Standard Video Decoder
(NVDEC), and a JPEG processing block (NVJPEG). NVENC enables full hardware acceleration for various
encoding standards including H.265, H.264, and AV1. NVDEC enables full hardware acceleration for
various decoding standards including H.265, H.264, AV1, VP9. NVJPG is responsible for JPEG
(de)compression calculations (based on the JPEG still image standard), image scaling, decoding
(YUV420, YUV422H/V, YUV444, YUV400) and color space conversion (RGB to YUV).
Next-generation Vision Accelerator Engine: PVA v2.
The PVA enables support for various computer vision kernels such as filtering, warping, image pyramid, feature detection, and FFT, which is essential for computer vision applications on the detector, tracker, object tracker, stereo disparity, and visual perception.
Jetson AGX Orin is ready embedded into next-gen autonomous machines, robotics or drones working for retail, manufacturing, transportation, agtech, smart city, aerospace, etc.
Coming soon in Q1 2022
According to the NVIDIA website, the Jetson AGX Orin module and developer kit are coming in Q1 2022. Stay tuned with NVIDIA’s notification and updates at Seeed!
Join Seeed Ecosystem
At Seeed, you will find everything you want to work with NVIDIA Jetson Platform – official NVIDIA Jetson Dev Kits, Seeed-designed carrier boards, and kits, as well as third-party boards and accessories.
Seeed will continue working on the Jetson product line and will be ready to combine our partners’ unique skills with Seeed’s hardware expertise for an end-to-end solution. Read Jetson powered ML application and stories from our ecosystem partner.
Medical imaging and healthcare
Remote edge AI device management
Deploying an AI idea can be faster, flexible, even scalable for everyone. Seeed Jetson Platform targets on helping educators, developers and enterprises deploy ML in the real world. By consolidating Seeed’s best-in-class hardware, cutting-edge technology from our software partners and all developers from the community, we aim at emerging all kinds of AI scenarios in our open-source platform to faster industries digital transformation. We are looking for partners to join our ecosystem together to deliver solutions to different industries together.
Please contact coope[email protected] for more partnership ecosystem information.