I have an odd bug on FMX that has all the hallmarks of a pointer overrun or something similar, writing pixels to a FMX TBitmap. It only occurs on some machines with a specific OS/hardware combination; I can't reproduce it; I can't reproduce even on the same machine with a test cut-down app; a code inspection shows nothing wrong, compiling with range checking etc is fine, as are bound-checking asserts. My customer is in the process of setting up remote access to the machine it occurs on, but I'd appreciate any "Yeah, you're clearly doing X" kind of help :)

I have an odd bug on FMX that has all the hallmarks of a pointer overrun or something similar, writing pixels to a FMX TBitmap. It only occurs on some machines with a specific OS/hardware combination; I can't reproduce it; I can't reproduce even on the same machine with a test cut-down app; a code inspection shows nothing wrong, compiling with range checking etc is fine, as are bound-checking asserts. My customer is in the process of setting up remote access to the machine it occurs on, but I'd appreciate any "Yeah, you're clearly doing X" kind of help :)

http://stackoverflow.com/questions/37325579/odd-artifacts-writing-to-fmx-bitmap-with-tbitmapdata-on-windows-10
http://stackoverflow.com/questions/37325579/odd-artifacts-writing-to-fmx-bitmap-with-tbitmapdata-on-windows-10