Behaviour of katakana keys on VIC-1001?

[srowe]

I'd always assumed that the katakana keys on the original Japanese model were just straightforward replacements for graphics characters but having looked through the source in http://www.zimmers.net/anonftp/pub/cbm/ ... src.tar.gz it has some commented-out code which handles them differently.

Could someone with a VIC-1001 explain how they work?

[srowe]

I think I've found an explanation in the VIC-1001 manual, unfortunately it's in Japanese of course

It seems that there are six mapping tables, which are also shown in the manual.

[Mayhem]

The first branch is holding down Shift and C= keys on boot. Then the branches underneath I think relate to standard keys, with shift, and with C= key held down.

[srowe]

The first branch is holding down Shift and C= keys on boot. Then the branches underneath I think relate to standard keys, with shift, and with C= key held down.

I've not reverse-engineered the VIC-1001 code but I think Shift/C= behaves differently. It toggles between two modes, the left is the usual unshifted/shifted/graphics and the right selects one of two further levels of decode depending on the C= key.

I could do with a Japanese speaker translating this diagram, alternatively I'll try feeding it into an online katakana OCR tool.

[ral-clan]

Yes, I heard that too. The VIC-1001 starts up in Upper Case Latin alphabet mode, then when you press C=/SHIFT it switches to Katakana mode (pressed anytime - not held down on boot). There is no lower case Latin alphabet mode like on the VIC-20.

I have never seen a VIC-1001 in person, this is just what I have inferred from information I've read.

[srowe]

That's useful info, thanks. As well as the diagram there's some evidence in the commented out KERNEL code that indicates other bits of the MODE byte (usually only used to lock the Shift/C= combination) were used to store this state.
If I get bored I'll dissassemble the VIC-1001 KERNAL in this area to answer the remaining questions.

[srowe]

With the aid of VICE I was able to resolve the mystery. SHIFT/C= switches between two character sets (as in the VIC-20) but when in the second character set the C= alone toggles between two submodes.

The state is stored in MODE ($0291)

• $00 default Western alphabetic
• $08 katakana submode A
• $10 katakana submode B

Interestingly mode switching via the keyboard cannot be disabled/enabled using CHR$(8) and CHR$(9).