OAK-FFC-6P: firmware crash when trying to run example scripts that use neural networks

[aheintz-lmco]

Hi,

On the OAK-FFC-6P with a single RGB camera connected to socket SA, I am able to run simple example scripts. These include:

• cam_test.py
• rgb_preview.py
• latency_measurement.py
• rgb_rotate_warp.py

However, more complex examples that use neural networks reach the following error:

" RuntimeError: Communication exception - possible device error/misconfiguration. Original message 'Couldn't read data from stream: '...' (X_LINK_ERROR)' "

Examples I have tested that encounter such an issue includes:

• rgb_mobilenet.py
• rgb_mobilenet_4k.py
• tiny_yolo.py
• object_tracker.py

No changes are made to these examples before running them.

The traceback always links to some form of calling the ".get()" method. In the context of "rgb_mobilenet.py", the traceback links to line 100: "inNN = qNN.tryGet()"

Additionally, the firmware crashes. It looks like the core dump is too large to attach, but I'm happy to pass it along if you can tell me where to send it.

Other notes:

• I am using the rvc3_develop branch
• I have updated the OAK-FFC-6P OS to 1.12 and the firmware to rvc3-1.2.0

How can I remedy this issue? Any help is much appreciated. Thanks in advance!

[Erol444]

Hi @aheintz-lmco ! NNs inside examples folder are all meant for RVC2, and the same model won't work on RVC3. We have a separate folder that does have RVC3 models/examples, see here: https://github.com/luxonis/depthai-python/tree/rvc3_develop/examples/tmp_s3 I hope this helps!