Paul E Collins wrote:
> In the Spectrum character set (and approximately on the keyboard), the
> block graphics are laid out in this order:
>
> [ ] [ '] [' ] [''] [ .] [ :] ['.] [':] [. ] [.'] [: ] [:'] [..] [.:]
> [:.] [::]
>
> What's the rule here? I can't see any coherent pattern to it.
>
> Eq.
>
>
looks like it's 0 to 15 in binary to me

the blocks are
2^1 2^0

2^3 2^2

--
link my boring website http://alistairsserver.no-ip.org/

"Alistair C" wrote in message
news:fnatvp$9la$1@netty.york.ac.uk...
> Paul E Collins wrote:
>> In the Spectrum character set (and approximately on the keyboard), the
>> block graphics are laid out in this order:
>>
>> [ ] [ '] [' ] [''] [ .] [ :] ['.] [':] [. ] [.'] [: ] [:'] [..] [.:]
>> [:.] [::]
>>
>> What's the rule here? I can't see any coherent pattern to it.
>>
>> Eq.
>>
>>
> looks like it's 0 to 15 in binary to me
>
> the blocks are
> 2^1 2^0
>
> 2^3 2^2


It is indeed - the characters are generated from the low four bits of the
charset index. This short program uses this system to generate large
graphics from the character set:

10 REM 4x4 Pixel magnifier
20 CLS
30 INPUT "Characters to render: "; LINE a$
40 IF LEN (a$)>8 THEN PRINT AT 0,0;"Up to 8 Characters only!": PAUSE 0: CLS
: GO TO 30
50 PRINT AT 21,0; INK 7;a$
60 FOR x=0 TO (LEN (a$)*8)-1 STEP 2
70 FOR y=0 TO 8 STEP 2
90 PRINT AT (8-y)/2,x/2;CHR$ (128+8*POINT (x,y)+4*POINT (x+1,y)+2*POINT
(x,y+1)+POINT (x+1,y+1))
100 NEXT y
110 NEXT x
120 PAUSE 0
130 GO TO 20

Line 90 is the part you'd be interested in.

D.

"Alistair C" wrote:

> looks like it's 0 to 15 in binary to me

D'oh. Blindingly obvious now. I had assumed, for some reason, that the
only possible way to number the blocks would have been most significant
bit at the top left and LSB at the bottom right, but they mirrored that
layout.

Eq.