Assuming I already have an unsigned char * array for an RGB image of
width W and height H (so there are 3WH elements in the array), how
do I go about assigning a surface pointing to that array ?
I’m assuming I use:
extern SDL_Surface *SDL_CreateRGBSurfaceFrom( void *pixels,
int width, int height, int depth, int pitch,
Uint32 Rmask, Uint32 Gmask, Uint32 Bmask, Uint32 Amask);
but what is pitch ? And are the masks specified as longs ? (so for
…RGBRGBRGB… data, would they be:
Rmask: 0xFF0000 Gmask: 0x00FF00 Bmask: 0x0000FF Amask: 0
Is ‘pitch’ W or W*3 and is depth in bits/pixel or bytes ?
The examples I’ve seen tend to load an image using an SDL function,
so it’s all set up for them. I’ve tried to copy the library function
semantics, but my program still crashes 
So I’m just trying to
clear up this bit to make sure I’ve interpreted the library code
properly…
Cheers,
Simon.
but what is pitch ? And are the masks specified as longs ? (so for
…RGBRGBRGB… data, would they be:
Rmask: 0xFF0000 Gmask: 0x00FF00 Bmask: 0x0000FF Amask: 0
Is ‘pitch’ W or W*3 and is depth in bits/pixel or bytes ?
this problem screwed my newbie for days, trial and error gave me:
pitch seems to be the real width in pixels, width is still a mystery to me,
height seems to be the number of rows (pitch) so for 24bit data :
surface->pitchsurface->heightsurface->BytesPerPixel=whole bytesOn 18-Jan-00 Simon Gornall wrote:
Assuming I already have an unsigned char * array for an RGB image of
width W and height H (so there are 3WH elements in the array), how
do I go about assigning a surface pointing to that array ?
I’m assuming I use:
extern SDL_Surface *SDL_CreateRGBSurfaceFrom( void *pixels,
int width, int height, int depth, int pitch,
Uint32 Rmask, Uint32 Gmask, Uint32 Bmask, Uint32 Amask);
A surface is assumed to be a linear block of memory allocated large enough
for the desired width and height. You pass a pointer to this memory to the
SDL_CreateRGBSurfaceFrom() function to create an SDL surface from preallocated
memory, or you can use SDL_CreateRGBSurface() to have SDL allocate the memory
for you.
but what is pitch ?
I added this to the FAQ:
http://www.devolution.com/~slouken/SDL/faq/FAQ-General.html#GENERAL_VIDEO1
And are the masks specified as longs ? (so for
…RGBRGBRGB… data, would they be:
Rmask: 0xFF0000 Gmask: 0x00FF00 Bmask: 0x0000FF Amask: 0
That’s correct.
Is ‘pitch’ W or W*3 and is depth in bits/pixel or bytes ?
W*3 in your case.
See ya!
-Sam Lantinga (slouken at devolution.com)
Lead Programmer, Loki Entertainment Software–
“Any sufficiently advanced bug is indistinguishable from a feature”
– Rich Kulawiec
depet at gmx.de wrote:
this problem screwed my newbie for days, trial and error gave me:
pitch seems to be the real width in pixels, width is still a mystery to me,
height seems to be the number of rows (pitch) so for 24bit data :
surface->pitchsurface->heightsurface->BytesPerPixel=whole bytes
Ooops!
And there was me thinking that
surface->widthsurface->heightsurface->BytesPerPixel=whole bytes
and that
surface->pitch = surface->width*surface->BytesPerPixel
Now I’m confused!
-Lea.
but what is pitch ? And are the masks specified as longs ? (so for
…RGBRGBRGB… data, would they be:
Rmask: 0xFF0000 Gmask: 0x00FF00 Bmask: 0x0000FF Amask: 0
Is ‘pitch’ W or W*3 and is depth in bits/pixel or bytes ?
Pitch is the “physical length of the width of the screen”, or the amount of
memory it takes. The 24-bit “truecolor” mode you’re describing probably
actually has a pitch of 32, because really what it looks like is:
Rmask: 0x00FF0000 Gmask:0x0000FF00 …
So pitch is really (usually) w*4. I just rip the code out of the example SDL
programs and hack it to the right settings, so I don’t remember exactly what I
use.
Nichola
vining at pacificcoast.net wrote:
but what is pitch ? And are the masks specified as longs ? (so for
…RGBRGBRGB… data, would they be:
Rmask: 0xFF0000 Gmask: 0x00FF00 Bmask: 0x0000FF Amask: 0
Is ‘pitch’ W or W*3 and is depth in bits/pixel or bytes ?
Pitch is the “physical length of the width of the screen”, or the amount of
memory it takes. The 24-bit “truecolor” mode you’re describing probably
actually has a pitch of 32, because really what it looks like is:
Rmask: 0x00FF0000 Gmask:0x0000FF00 …
So pitch is really (usually) w*4. I just rip the code out of the example SDL
programs and hack it to the right settings, so I don’t remember exactly what I
use.
No, I’m sure that my data is RGBRGB… etc so I guess it’s 24-bit, but
thanks for
the info 
ATB,
Simon
Ooops!
And there was me thinking that
surface->widthsurface->heightsurface->BytesPerPixel=whole bytes
and that
surface->pitch = surface->width*surface->BytesPerPixel
Now I’m confused!
You are right, I was wrong, please excuse I was in a complete wierdo stateOn 18-Jan-00 Lea Anthony wrote:
Actually the width and the pitch aren’t always the same.
For example usually in 24bpp mode the width of a scanline
(in bytes) is width*3. So for a 640 pixel scanline the
width in bytes is 1920. However for various reasons
video card manufacturers pad scanlines for alignment
and other reasons. So the pitch might actually be larger
then the width. For 640 @ 24bpp the pitch is often
2560 (640 *4 bytes per pixel). The pitch is the amount
of bytes that must be added to a pointer to get to the
same pixel position on the next scanline…This is not
always the width.
Adam Meyerowitz
In article <38849DCF.8D4D67C5 at prismtechnologies.com>,
lea.anthony at prismtechnologies.com says…>
depet at gmx.de wrote:
this problem screwed my newbie for days, trial and error gave me:
pitch seems to be the real width in pixels, width is still a mystery to me,
height seems to be the number of rows (pitch) so for 24bit data :
surface->pitchsurface->heightsurface->BytesPerPixel=whole bytes
Ooops!
And there was me thinking that
surface->widthsurface->heightsurface->BytesPerPixel=whole bytes
and that
surface->pitch = surface->width*surface->BytesPerPixel
Now I’m confused!
-Lea.