BME688 BSEC2.2.0.0 pi3g segmentation fault

[BSTRobin]

Hi Prop4et,

Could we know detailed information about your host architecture?

[BSTRobin]

Hi Prop4et,

Could you give further feedback? Thanks.

[BSTRobin]

Hi Prop4et,

Could you give further feedback? Thanks.

[Prop4et]

It compiles the pithree version and by forcing a call to the library function it starts to record data that is kinda real. I stuck to that because the problem i'm facing is now related to loading the data into bme ai studio, eventhough the format is right, the software doesn't seem to load the whole duration (about 4 hours) but only an hour of sampling.

Because of that i bought a development kit to at least be able to create an algorithm and i manage to do so.

When i have to get back to the pi i'll let you know everything

[BSTRobin]

Hi Prop4et,

Do you mean you changed you hardware platform to others and BSEC2 software had been worked well on you new platform?

[Prop4et]

I mean i am still using the same raspberry pi 4B that i was using at the beginning, but i changed the version of the library. The library i am using is the one compiled for the PiThree_ArmV6. This one is compiled by the raspberry pi and it runs. The only issue i have with this is that if i want to read values from the bsec library i need to force a call to a library function before starting the get data process used for sampling by the pi3g code. Otherwise i have TPH readings that are fakes (for example H is always 100). I know they can be ignored when generating the algorithm, but i was wondering why there is this behaviour.

If i try and compile the PiFour_ArmV8 i get an error telling me that it can't be compiled, and by searching the internet it seems like the error is returned when the hardware is not correct for the library (.a file) i am using.