Some of the former Pixhawk v2 sold by 3DR had an 1MB program size limit.
Does anyone know if that size limit still exist with the Pixhawk now sold by mRo?
And if that is not the case anymore, then will QGC upload “px4fmu-v3_default.px4” instead of “px4fmu-v2_default.px4” or do I need to upload “px4fmu-v3_default.px4” manually instead?
Your link says it’s v3 so not limited flash. You can use either v3 or v2 binaries with that pixhawk.
@metinburak, thank you for pointing that out. I completely missed it.
“including STM32F427 revision 3 that corrects a bug that limited the Pixhawk to 1MB of flash”
A couple of more questions. There are two versions of the px4fmu-v2 (px4fmu-v2_default.px4 & px4fmu-v2_lpe.px4), but only one px4fmu-v3_default.px4. Do you know why is that?
We’re mainly flying indoors without GPS so will it be okay to install px4fmu-v3_default.px4 and just use LPE?
as you can configure the estimators via conf parameters, it is ok to build it with default and select the estimator before build via sys parameters (was like estimator group parameter i think)
i suggest using ekf2.
Can I use px4fmu-v3_default.px4 instead of px4fmu-v2_default.px4, and select the estimator via sys parameters as you suggested?
Or, are you saying that I should still use px4fmu-v2_default.px4 with this new Pixhawk, and select the estimator via sys parameters instead of using px4fmu-v2_lpe.px4?
Yes you can flash px4fmu-v3_default on a newer px4fmu-v2.
Thanks for confirming. Just for curiosity, why isn’t there a px4fmu-v3_lpe.px4 as well?
The difference between the builds (v2,v3) is the linker script which is about the flash size. I think the v3 builds targets the 2MB flash so there is no need to specifically build for one estimator because of sufficient flash on STM F427 CPU v3. It’s my understanding of course. (no need because u have sufficient flash.)