The text on the start screen isn't positioned using PETSCII in R43

[mobluse]

Here you see the output from ./x16emu -echo raw | ~/x16-petscii2utf8/petscii2utf8 running R42 and R43, respectively:

x16emu R42

2023-07-05-171622_655x598_scrot.png (38.68 KiB) Viewed 5617 times

x16emu R43

2023-07-05-171646_655x598_scrot.png (26.88 KiB) Viewed 5617 times

As you can see the positioning of the text is wrong in R43. This is probably because they use another cursor positioning method than before. They probably did this because they wanted the screen after reset to work with different SCREEN modes stored in NVRAM. I know it is possible to position using PETSCII control characters e.g. the cursor keys and perhaps LF and CR. I think this should be fixed so that it can look good when using my filter program, petscii2utf8, because you might want to run Commander X16 using a terminal with this filter or some other PETSCII compatible terminal. I cannot solve it on my side because petscii2utf8 doesn't get information about the positions.

More information about petscii2utf8:

• viewtopic.php?t=6627
• https://github.com/mobluse/x16-petscii2utf8

[TomXP411]

This is not a bug.

The system can now boot to different screen modes, so the boot screen looks a little different for different screen modes. Using PETSCII to position text is less efficient than using the PLOT routine, so I imagine this explains that change.

If you want a PETSCII based boot banner, you are free to create one yourself and save it as AUTOBOOT.X16.

[mobluse]

Yes, I know it's more an issue than a bug, but I don't think the screen after reset appears too slowly; it could use PETSCII positioning without anyone thinking it is too slow, and that would also solve my issue. It's a good idea to write your own boot screen program in AUTOBOOT.X16. I would get the data in the same way as the ROM.

The ROM could have an extra PLOT subroutine for PETSCII positioning that would go to the position using the fewest PETSCII control codes.

[TomXP411]

Yes, I know it's more an issue than a bug, but I don't think the screen after reset appears too slowly; it could use PETSCII positioning without anyone thinking it is too slow, and that would also solve my issue. It's a good idea to write your own boot screen program in AUTOBOOT.X16. I would get the data in the same way as the ROM.

The ROM could have an extra PLOT subroutine for PETSCII positioning that would go to the position using the fewest PETSCII control codes.

You're free to submit a PR, of course.

I get why you want this - to be able to log and re-create the screen via STDIO, but I can't imagine the volunteers working on KERNAL and BASIC ROM code would set aside other issues for this - not without a compelling use case.