Wiki editing discussion

Page 46/50
39 | 40 | 41 | 42 | 43 | 44 | 45 | | 47 | 48 | 49 | 50

By gdx

Prophet (3090)

gdx's picture

20-09-2019, 09:39

Done, thanks. Before the routine worked by chance.
This is not ideal as example, if someone has simpler to suggest it's welcome.

By Manuel

Ascended (15835)

Manuel's picture

22-09-2019, 17:02

People were talking here about the Toshiba HX-10. I found some weirdness in the keyboard matrix of the dump I have of the HX-10. Can the people who own a Toshiba HX-10 dump their main ROM so I can investigate? Please make clear of which exact version (bottom plate) the ROM belongs.

By gdx

Prophet (3090)

gdx's picture

23-09-2019, 12:13

By Manuel

Ascended (15835)

Manuel's picture

23-09-2019, 21:29

Thanks. I'm also especially interested in European ones.

The one you posted isn't special, it's a very commonly used ROM in many Japanese MSX1 computers: 302afb5d8be26c758309ca3df611ae69cced2821

According to my info used in:
Canon_V-10.xml
Canon_V-20_JP.xml
Casio_MX-10.xml
Casio_PV-16.xml
Casio_PV-7.xml
Fujitsu_FM-X.xml
Hitachi_MB-H1.xml
Hitachi_MB-H50.xml
Mitsubishi_ML-F110.xml
Mitsubishi_ML-F120.xml
National_CF-2000.xml
Pioneer_PX-7.xml
Pioneer_PX-V60.xml
Sanyo_MPC-10.xml
Sanyo_MPC-6.xml
Sony_HB-101.xml
Sony_HB-10.xml
Sony_HB-201.xml
Toshiba_HX-10D.xml
Toshiba_HX-22.xml
Victor_HC-30.xml
Yamaha_CX11.xml
Yamaha_CX5F.xml

By Manuel

Ascended (15835)

Manuel's picture

27-09-2019, 00:18

Manuel wrote:

I do wonder who generated these nice character table figures in our MSX font article. Looks better than the tables from the MSX Technical Databook...

Anyone who knows?

I wrote an MSX-BASIC program to display the font (MSX character set) in a similar way to the MSX Technical Databook and the results for some Arabic machines are quite different than on the Wiki article.
Also, it appears several models have the cursor on 0x7F, and not on 0xFF! Interesting....!

My program:

10 SCREEN1:KEYOFF:COLOR1,15,15:DEFINTA-Z
20 RS=32:WIDTHRS:OX=(RS-&HF)\2:OY=(24-&H
F)\2
30 FOR R=0 TO &HF
40 LOCATE OX+R,OY-1:PRINTHEX$(R)
50 FOR C=0 TO &HF
60 X=OX+C
70 Y=OY+R
80 IF R=0 THEN LOCATE OX-1,OY+C:PRINT HEX$(C)
90 VPOKE BASE(5)+X+(RS*Y),C*&H10+R
100 NEXT C,R
110 GOTO 110

By gdx

Prophet (3090)

gdx's picture

27-09-2019, 09:36

Manuel wrote:

some Arabic machines are quite different than on the Wiki article.

Wiki article shows the font from AX170. I noticed that two or three characters are missing (0x9E and 0x9F). 0x9E is same as 0x7E and 0x9F is same as 0x7F (DEL). Character 0xFF seems also same as DEL.

I had used the following program to print it.

10 SCREEN1:WIDTH31: KEY OFF
20 FOR Y=0 TO 15
30 FOR X=0 TO 15:LOCATE X,Y
40 IF C=127 THEN 80
50 IF C=255 THEN 80
60 IF Y<2 THEN PRINT CHR$(1)+CHR$(C);
70 IF Y>1 THEN PRINT CHR$(C);
80 C=C+1: NEXT: NEXT

but recently I noticed that it's different in SCREEN2 mode.

10 COLOR5,15
20 SCREEN2:OPEN"grp:"AS#1
30 FOR Y=0 TO 15
40 FOR X=0 TO 15:PRESET(32+X*9,Y*9)
50 IF C=127 THEN 120
60 'IF C=159 THEN 120
70 'IF C=255 THEN 120
80 IF C=32 THEN COLOR 13
90 IF C=128 THEN COLOR 12
100 IF Y<2 THEN PRINT#1,CHR$(1)+CHR$(C);
110 IF Y>1 THEN PRINT#1,CHR$(C);
120 C=C+1: NEXT: NEXT
130 IF STRIG(0)=0 THEN 130

Weird.

By Manuel

Ascended (15835)

Manuel's picture

27-09-2019, 18:22

This is what I get with my program for the AX-170:

Several of these machines have the cursor not at 0xFF but at 0x7F!

As you see in column 1 there are accented characters. But other Arabic ROMs have box drawing characters there, for instance, the Bawareth has the box drawing characters there and the cursor at 0xFF. The Yamaha AX500 has the box drawing characters there and the cursor at 0x7F.

So much for standards...

By Manuel

Ascended (15835)

Manuel's picture

27-09-2019, 23:17

Anyway, as you see, it's quite different from what's on the Wiki now. The extended characters are missing and column F seems different as well.

Anyway, on a page about MSX font, I'd still prefer to only show the character set and not control codes, just like what the MSX Technical Databook does.

By gdx

Prophet (3090)

gdx's picture

28-09-2019, 09:29

Your table is also wrong since PRINT CHR$(&H10) does not display "é" and it's the same for the 1xh column. In addition PRINT "NO"+CHR$(&H7F) clears the left character ("O") in SCREEN0/1. The cursor is not displayed.

For the rest you get the same thing as my program in SCREEN2.

In short, one thing is certain, there is something that escapes us. Wiki is incomplete.

Quote:

Anyway, on a page about MSX font, I'd still prefer to only show the character set and not control codes, just like what the MSX Technical Databook does.

The purpose of this page is to show both the character code and their form. It is to inform the developers so it speaks of both. What is important is how to display them and where to find them.

By Manuel

Ascended (15835)

Manuel's picture

28-09-2019, 09:56

The table is not wrong, it is just how all character codes are displayed when you put them in VRAM. As I said, it reproduces the table for international character set from the MSX Technical Databook.

As Rebecca Bettencourt, who is submitting proposals to the Unicode committee to get legacy character sets included, is doing might be better to split it up:

Quote:

This is something rather unique to 8-bit micros that computers today do not deal with: when the character codes used in video memory to represent a character differ from the character codes sent to the BIOS to get those characters displayed. The MSX is far from the only platform to do this. Smile We've taken to calling the former the "video mapping" and the latter the "interchange mapping." In this case we'll now have two mappings for MSX (the international version, that is to say): the "interchange" version that uses the control codes for characters 00-1F and the "video" version that maps 00-1F directly. Cool

And 0x7f as well of course.

Page 46/50
39 | 40 | 41 | 42 | 43 | 44 | 45 | | 47 | 48 | 49 | 50