> By the way, there are some bits in a VDP register controlling "VRAM
> config". Have you tried them? Any effect?
>
I tried these once, I can't remember having noticed ANY effect. I can
imagine this: some older DRAM-IC's (like ...K x 1 bit types) have
separate data inputs and outputs, and/or require different timings
for the control signals. My guess is, that these VDP-bits indicate to
the V9938, what timings/control methods to use.
In practice, it's probably so, that the usually used 64 K x 4 bit
DRAM's also accomodate timings as used for older DRAM's, making the
exact method used a don't care (and therefore don't show any effect
on common MSX machines).
Then somebody mentioned that he couldn't find any 'new' screen modes,
well I know of one, but I guess this must be more or less known
already, it's MSX-1 specific:
In SCREEN 1, colors for the characters are set for groups of 8
characters. In SCREEN 2 (graphic mode), this is for each 'character'
separate (background/foreground-color set per horizontal line of 8
pixels, in both cases), and the screen consists of 3 groups of 256
'characters'. I know of an 'in-between' of this, giving a textmode,
with 256-element character-set, where for each character separately,
a backgrond/foreground-color can be set per line of 8 pixels.
The BASIC listing below uses this effect, and you could study it, to
see how it's done (not that I care, just maybe an interesting fact):
10 DEFINT A-Z
20 ' DE TRUC: (VARIATIE OP SCREEN 1)
30 BASE(5)=6144: BASE(6)=8192: BASE(7)=0: BASE(8)=6912:
BASE(9)=4096
40 SCREEN 1: COLOR 10,1,1: CLS
50 VDP(0)=VDP(0) OR 2: VDP(3)=159: VDP(4)=0
60 A=&HF3E9:K=16*PEEK(A)+PEEK(A+1)
70 FOR A=BASE(6) TO BASE(6)+2047: VPOKE A,K: NEXT
80 ' DRUK KARAKTERSET AF
90 FOR K=32 TO 255: PRINT CHR$(K);: NEXT K
100 ' STEL KLEURKODES IN VOOR 'SCHADUWLETTERS'
110 FOR K=0 TO 255
120 A=BASE(6)+8*K
130 VPOKE A+3,144: VPOKE A+4,144
140 VPOKE A+2,128: VPOKE A+5,128
150 VPOKE A+1,96: VPOKE A+6,96
160 VPOKE A,208: VPOKE A+7,208
170 NEXT K
180 ' ZET NOG EVEN EEN SPRITE OP HET SCHERM
190 SCREEN ,2: S$="": RESTORE 220
200 FOR N=0 TO 31: READ B$: S$=S$+CHR$(VAL("&H"+B$)): NEXT N
210 SPRITE$(0)=S$: PUT SPRITE 0,(220,160),10,0
220 END
230 DATA
3,F,1F,3F,3F,7F,7F,FF,FF,FF,7F,7F,3F,3F,1F,7,C0,F0,F8,38,7E,F8,E0,80,
8 0,E0,F8,FC,F8,F0,E0,80
Ofcourse, on the V9938 and higher you can combine this with color
palette functions. Maybe this works as well with SCREEN 4 (SCREEN 2,
with improved sprite-functions), but I haven't tried that (who uses
SCREEN 2 or 4 anyway).
What WOULD be interesting to know:
If there is a number of bits controlling the screen modes, and there
are less screen modes, than possible combinations of these bits, WHAT
screen-modes DO you get, when setting one of these undefined
combinations that are possible?
If anyone ever figures this out, I'm sure it will interest some of
us, so dump it on this mailing-list, should you find any answers to
this....
Greetings,
Alwin Henseler ([EMAIL PROTECTED])
http://www.twente.nl/~cce/index.htm (computerclub Enschede)
http://huizen.dds.nl/~alwinh/msx (MSX Tech Doc page)
Aaahhh, I think the weather clears up after all, and you can leave
your umbrella's at home when going to Zandvoort... :->
****
MSX Mailinglist. To unsubscribe, send an email to [EMAIL PROTECTED] and put
in the body (not subject) "unsubscribe msx [EMAIL PROTECTED]" (without the
quotes :-) Problems? contact [EMAIL PROTECTED] (www.stack.nl/~wiebe/mailinglist/)
****
