Difference between revisions of "Getting started with AI on NXP i.MX8M Plus/Introduction/Overview/i.MX8M Plus hardware specifications"
Line 22: | Line 22: | ||
|- | |- | ||
| NPU | | NPU | ||
− | | - Reaches up to 2.3 TOP/s.<br />- Optimized for voice and image<br /> recognition, object detection. <br />- Uses Winograd Algorithm. | + | | - Reaches up to 2.3 TOP/s.<br />- Optimized for voice and image<br /> recognition, object detection. <br />- Uses Winograd Algorithm. <br />- Supports 8 and 16 bits operations. |
|- | |- | ||
| RAM | | RAM |
Revision as of 16:35, 21 November 2021
Getting started with AI on NXP i.MX8M Plus RidgeRun documentation is currently under development. |
|
The board used in this entire research is the Variscite DART-MX8M-PLUS based on NXP i.MX8M Plus, and some of the main characteristics are listed in the table below.
Hardware module | Specification |
---|---|
CPU | - 4 cores ARM Cortex A53. - Frequency: [1200, 1800] Hz. - Supported 32 and 64 bits. - One thread per core. |
GPU | - 3D GPU GC7000UltraLite. - 2D GPU GC520L. |
NPU | - Reaches up to 2.3 TOP/s. - Optimized for voice and image recognition, object detection. - Uses Winograd Algorithm. - Supports 8 and 16 bits operations. |
RAM | - 4 GB. - LPDDR4-4000. |
Note: For more information please visit the Official Variscite site.