IPC Specifications¶
Mech-Mind provides the following IPC configurations:
Configuration |
Deluxe |
Advanced |
Standard |
Basic |
|
CPU |
Recommended model |
i7-11700 (F) and above |
i7-8700 and above |
i5-12400 # UHD 730 and above |
i7-1165G7 # Iris Xe 96EU |
Cores/Threads |
12 cores/20 threads |
6 cores/12 threads |
6 cores/12 threads |
4 cores/8 threads |
|
CPU TDP |
65/180 W |
65 W |
65 W |
12–28 W |
|
Discrete graphics |
Recommended model |
RTX 3060 |
GTX 1650 |
None |
None |
Video RAM |
12 GB |
4 GB |
|||
Discrete graphics TDP |
170 W |
75 W |
|||
RAM |
≥ 16 GB |
16 GB |
16 GB |
8 GB |
|
ROM |
512 GB SSD + 1 TB |
512 GB SSD |
256 GB SSD (expandable) |
256 GB SSD |
The application scenarios of each IPC configuration:
IPC configuration |
Application scenario |
Deluxe |
Exhibitions |
Projects that require high product performance and extremely short cycle time |
|
Deep learning model training |
|
Advanced |
Applications with relatively strict requirements, such as requiring multiple Mech-Vision projects or short cycle time |
Standard |
Requiring deep learning model inference; regular applications such as depalletizing and machine tending |
Basic |
Running simple Mech-Vision projects |
Recommended version of Mech-DLK for different IPC configurations:
IPC configuration |
Recommended Mech-DLK version |
Deluxe |
Mech-DLK Pro-Train (Deep learning training) |
Mech-DLK Pro-Train (Deep learning training) |
|
Mech-DLK Pro-Run (Deep learning inference) |
|
Advanced |
Mech-DLK Pro-Train (Deep learning training) |
Mech-DLK Pro-Run (Deep learning inference) |
|
Standard |
Mech-DLK Pro-Run (Deep learning inference) |
Basic |
Not applicable |