On v48.0.0.1, at 8MHz, writing all black pixels as a 1bpp bitmap results in a garbled screen on the 65816 writing 16-bits across Data0 and Data1 in one instruction (see attached photo).
Works correctly on v47.0.2 at 8MHz, or at slower clock speeds on v48.0.1.
This assembly program demonstrates the issue: https://gist.githubusercontent.com/catharinejm/0322454e0001ce1ee13355c8154f6277/raw/c749efea391f763f2db9c543c9383583fc67050d/vera.s
It has a BASIC header, so it can be loaded and run normally once assembled.
Assembly command for convenience:
ca65 -tnone -o vera.o vera.s && ld65 -tnone -o VERA.PRG vera.o
One of the times I ran it, the white pixels were cyan. Possibly the 8-bit write to L0_HSCROLL_H was also corrupted?
On v48.0.0.1, at 8MHz, writing all black pixels as a 1bpp bitmap results in a garbled screen on the 65816 writing 16-bits across Data0 and Data1 in one instruction (see attached photo).
Works correctly on v47.0.2 at 8MHz, or at slower clock speeds on v48.0.1.
This assembly program demonstrates the issue: https://gist.githubusercontent.com/catharinejm/0322454e0001ce1ee13355c8154f6277/raw/c749efea391f763f2db9c543c9383583fc67050d/vera.s
It has a BASIC header, so it can be loaded and run normally once assembled.
Assembly command for convenience:
ca65 -tnone -o vera.o vera.s && ld65 -tnone -o VERA.PRG vera.oOne of the times I ran it, the white pixels were cyan. Possibly the 8-bit write to L0_HSCROLL_H was also corrupted?