I was referring to Rosetta 2, but thinking about it, it doesn’t really change the situation as I’ve found a couple of API that need to be different on ARM than Intel, so you’ll still need an Intel Mac to test your applications with and debug specific Intel issues.
Update:
After a lot of messing about, I find that the event order (and events at all) ,
differ in debug versus compiled mode,
and also between Big Sur versus Monterey.
I removed all code from New Document , since I never want that to fire by itself before initialisation has happened, (only fires on Monterey)
then moved all inititalisation into a method,
then trapped all app events like open/ open document and ensured that they went through this process before showing any screens.
Now working on my intel Monterey machines, and I’m having some folk check on M1
1 Like
That’s odd and something I never encountered. Do you have an example that shows this?
‘Shows’, no, sorry.
I regrettably don’t have the time to try and create a stripped down file at the moment.
But the flow of methods and events was different when debugging, than when running the compiled version, based on logging files I generated.
The crash log showed a problem with XML … the newdocument event was firing while an XML based loop -through-nodes section was being processed.
That is the reason why code should not rely on that order, as advised in the LR.
In my experience this is about 100% true.
In my experience it is not. I have a project that works on Intel and not on M1 or the other way around 