Qbasicnews.com

Hello everyone

I've always been able to use the green and red colour palettes, but whenever I try that for the blue colour, an error message appears. It's been years since I last tried it, but I think the message was "Invalid Function Parameter", but in portuguese Smile

. I think it has to do with the number being too big for the function, but I used the procedure described in tutorials, something like: "Green + Red * 64 + Blue * 128" (again, this is from memory so it may not be entirely correct). Is it because I'm using the Interpreter, and it only works with the Compiler (technically making it a bug), or am I doing something wrong? Could you give a very brief code snippet that works? Sorry if this is discussed in tutorials or faqs that I haven't looked into. Thanks.

(I've read the palette tutorial on the site, but it doesn't work for me, since I usually avoid direct port calls and assembly statements).

Here you go dude:

Code:
'' QB palette shifting example

'' cha0s Jun '06

'' save as QBRGB.BAS

'' ----------------------------

DECLARE FUNCTION QBRGB& (r AS INTEGER, b AS INTEGER, g AS INTEGER)

DEFINT A-Z

CONST colorCurrent = 4 '' random palette location to modify.

SCREEN 13

'' Display 50, 50, 50

PALETTE colorCurrent, QBRGB&(50, 50, 50)

PAINT (0, 0), colorCurrent

LOCATE 1: PRINT "light-gray"

WHILE LEN(INKEY$) = 0: WEND

'' Display 10, 35, 50

PALETTE colorCurrent, QBRGB&(10, 45, 50)

PAINT (0, 0), colorCurrent

LOCATE 1: PRINT "blue and greenish"

WHILE LEN(INKEY$) = 0: WEND

'' Display 50, 35, 8

PALETTE colorCurrent, QBRGB&(50, 35, 8)

PAINT (0, 0), colorCurrent

LOCATE 1: PRINT "orange"

WHILE LEN(INKEY$) = 0: WEND

FUNCTION QBRGB& (r AS INTEGER, g AS INTEGER, b AS INTEGER)

  QBRGB& = 65536 * b + 256 * g + r

END FUNCTION

RGB is implemented in FB as a macro, which means it's lots faster than using a function to shift the values, like in the QB version. You can calculate the rgb value inline of course, but that gets messy. ;p

I'll try that Chaos, thanks.

This is generally faster (in most computers):

Code:
SUB SetDAC (Index%, r%, g%, b%)

   OUT &H3c8, Index%

   OUT &H3c9, r%

   OUT &H3c9, g%

   OUT &H3c9, b%

END SUB

In QBasic 1.1 , the PALETTE command doesn't appear to accept extended integers. Could someone try it, using the blue colour, and corroborate this?

I still haven't tried the code snippet (QBasic is installed in my 486, not my net machine), but I can predict that it won't work, because the PALETTE command won't work itself, unless you use only the green and red colours. Thanks Nathan, but I won't do port output, my 486 is way too valuable for me to do that, no safety-net there if I do a typing mistake.

OK, let me correct myself, the code snippet does work with QBasic 1.1. I've never been able to do it because, for some reason, QBasic doesn't think 256*256 is the same thing as 65536. Go figure.

Keep in mind that the pallete command performs error checking on the supplied RGB values to ensure that they are in the range of 0 to 63, not 0 to 255. Consequently not every 24 bit value (3 eight bit fields) is a legal parameter.

This is done because the VGA DAC only provided six bits for the RGB values.

Man, you won't break your computer OUTing to the VGA BIOS I/O ports. Every program you use on it does it for everything.

Try this:

Code:
a& = 256*256

PRINT a&

Always use the correct data type.

torstum

Anonymous

torstum

na_th_an

torstum

torstum

RyanKelly

na_th_an