The two modules have the same FP32 performance (0.666 TFLOPS). The Jetson TX2 has twice the memory (8 GB vs. 4 GB) and more memory bandwidth (59.7 GB/s vs. 51.2 GB/s).
Jetson TX2 4GB vs. Jetson TX2 developer... +1developer.nvidia.com ↗“memory.memory_gb=4 GB, memory.memory_type=LPDDR4, memory.memory_bandwidth_gb_per_s=51.2”developer.nvidia.com ↗“memory.memory_gb=8 GB, memory.memory_type=LPDDR4, memory.memory_bandwidth_gb_per_s=59.7”
| Spec | Jetson TX2 4GB | Jetson TX2 |
|---|---|---|
| Memory | 4 GB LPDDR4 | 8 GB LPDDR4 |
| Memory bandwidth | 51.2 GB/s | 59.7 GB/s |
| Max FP32 | 0.666 TFLOPS | 0.666 TFLOPS |
- Both modules use a Pascal GPU with 256 CUDA cores, so their FP32 figure is the same. developer... +1developer.nvidia.com ↗“compute.gpu_cuda_cores=256, compute.gpu_architecture=Pascal”developer.nvidia.com ↗“compute.gpu_cuda_cores=256, compute.gpu_architecture=Pascal”
- Both modules are listed as end of life (EOL). developer... +1developer.nvidia.com ↗“identity.lifecycle_status=EOL”developer.nvidia.com ↗“identity.lifecycle_status=EOL”