#90909 - HtheB - Mon Jul 03, 2006 12:43 pm
Hi, I want to convert .bmp to .raw..
I tried many tools that I could find (yes.. even on gbadev)
It didnt work with any application.. (it wont work on the DS)
I had conv.exe.. (it was the best so far..) but.. does anyone has the conv.exe? b/c I think I have a corrupted version of it.. (the images works.. but its VERY weird..)
if someone know where to find it.. please tell me ^_^
tnq! :)
#90921 - silent_code - Mon Jul 03, 2006 2:19 pm
use usenti and/or (win)git (google for it ;) )... should solve your problems (except for 16bit images)...
#90930 - LiraNuna - Mon Jul 03, 2006 3:47 pm
what format you need the .raw as? 16col, 256col 32kcol ?
if it's one of those, gfx2gba is what you're looking for.
| Code: |
| gfx2gba -c[16/256/32k] img.bmp |
_________________
Private property.
Violators will be shot, survivors will be shot again.
#90938 - HtheB - Mon Jul 03, 2006 4:56 pm
| silent_code wrote: |
| use usenti and/or (win)git (google for it ;) )... should solve your problems (except for 16bit images)... |
Tried out allready ;)
didnt work..
| LiraNuna wrote: |
what format you need the .raw as? 16col, 256col 32kcol ?
if it's one of those, gfx2gba is what you're looking for.
| Code: | | gfx2gba -c[16/256/32k] img.bmp |
|
Allready tried out.. didn't work eather..
its 24bit bmp format.. ^_^
so.... I really need that conv.exe its the only way :(
#90942 - silent_code - Mon Jul 03, 2006 5:11 pm
then just save it in 16 or 8 bit ;) that way you'll be able to use gfx2gba or the tools i mentioned... shouldn't be a problem ;)
#90943 - Cearn - Mon Jul 03, 2006 5:29 pm
| HtheB wrote: |
| silent_code wrote: | | use usenti and/or (win)git (google for it ;) )... should solve your problems (except for 16bit images)... |
Tried out allready ;)
didnt work.. |
Could you define 'didn't work' here? Apart from the alpha bit, these tools should give the proper result even for 24bit images.
#91043 - HtheB - Tue Jul 04, 2006 6:41 am
| Cearn wrote: |
| HtheB wrote: | | silent_code wrote: | | use usenti and/or (win)git (google for it ;) )... should solve your problems (except for 16bit images)... |
Tried out allready ;)
didnt work.. |
Could you define 'didn't work' here? Apart from the alpha bit, these tools should give the proper result even for 24bit images. |
Ok, I will make pictures and show them.. (really its horrible :( )
#91151 - silent_code - Tue Jul 04, 2006 10:27 pm
you better do ;)
it sounds like it's a code problem... i'm using 24bit .bmp files with usenti and (win)git and the only problem i have is the 24 to 16bit conversion of my raytraced images :( (look at the other thread about that problem).
but that's nothing to do with the tools.
#91217 - HtheB - Wed Jul 05, 2006 10:55 am
| silent_code wrote: |
you better do ;)
it sounds like it's a code problem... i'm using 24bit .bmp files with usenti and (win)git and the only problem i have is the 24 to 16bit conversion of my raytraced images :( (look at the other thread about that problem).
but that's nothing to do with the tools. |
this is what it says in the conv.cpp
| Code: |
#include <stdio.h>
typedef unsigned short ushort;
typedef unsigned char byte;
typedef unsigned int uint;
#define RGB(r,g,b) ((1 << 15) | (r) | ((g) << 5) | ((b) << 10))
//#define RGB(r,g,b) ((b) | ((g) << 6) | ((r) << 11))
char *dumpFile(char* name) {
char *font;
int size;
FILE *fd = fopen(name, "r");
fseek(fd, 0, SEEK_END);
size = ftell(fd);
fseek(fd, 0, SEEK_SET);
font = new char[size];
fread(font, 1, size, fd);
fclose(fd);
return font;
}
int main(int argc, char** argv) {
byte rgb[3];
ushort *dataOff;
int off, pad;
char *file = dumpFile(argv[1]);
char *fileOff;
uint width, height;
int size;
ushort * data;
width = *(uint*)(file + 18);
height = *(uint*)(file + 22);
data = new ushort[width * height];
size = width * height * 2;
off = 0;
pad = (3000 - width * 3) & 3;
fileOff = file + 0x36;
dataOff = data + (height - 1) * width;
for (int y = 0; y < height; y++) {
for (int x = 0; x < width; x++) {
rgb[0] = (byte)(*fileOff++);
rgb[1] = (byte)(*fileOff++);
rgb[2] = (byte)(*fileOff++);
dataOff[x] = RGB(rgb[2] >> 3,
rgb[1] >> 3,
rgb[0] >> 3);
}
dataOff -= width;
fileOff += pad;
}
FILE *fd = fopen(argv[2], "w");
rgb[0] = height;
rgb[1] = width;
fwrite(rgb, 1, 2, fd);
fwrite(data, size, 1, fd);
fclose(fd);
delete[] (file);
} |
#91239 - HtheB - Wed Jul 05, 2006 3:49 pm
when I compile it.. and use it (conv.exe) I still get corrupted pictures :(
#91246 - Cearn - Wed Jul 05, 2006 5:13 pm
That conv.exe works fine*, you're either not giving it the right kind of bitmap, or using the data it produces wrongly. Check your video mode settings.
* Though it's a minor miracle: no comments, magic numbers galore and no safety checks of any kind. The thing expects a 24bpp, version 3, bottom-up windows bitmap, where the width is a multiple of 4, but doesn't mention that anywhere. These conditions are probably met for your bitmap, but anything else will probably hang or crash the program, or worse: give incorrect output. The fact that the fopen() calls do not explicitly use binary mode is also slightly worrying; I'm not 100% sure, but that may give bad results if bytes are 0x0A ('\n') or 0x0D ('\r').
Also note that the first 2 bytes give the size of the bitmap. Again, not documented, and would be wrong for DS-screensized bitmaps or larger.
#91263 - HtheB - Wed Jul 05, 2006 7:12 pm
| Cearn wrote: |
That conv.exe works fine*, you're either not giving it the right kind of bitmap, or using the data it produces wrongly. Check your video mode settings.
* Though it's a minor miracle: no comments, magic numbers galore and no safety checks of any kind. The thing expects a 24bpp, version 3, bottom-up windows bitmap, where the width is a multiple of 4, but doesn't mention that anywhere. These conditions are probably met for your bitmap, but anything else will probably hang or crash the program, or worse: give incorrect output. The fact that the fopen() calls do not explicitly use binary mode is also slightly worrying; I'm not 100% sure, but that may give bad results if bytes are 0x0A ('\n') or 0x0D ('\r').
Also note that the first 2 bytes give the size of the bitmap. Again, not documented, and would be wrong for DS-screensized bitmaps or larger. |
YES!!! I GOT IT TO WORK!!!
i changed:
| Code: |
| FILE *fd = fopen(name, "r"); |
and
| Code: |
| FILE *fd = fopen(argv[2], "w"); |
into:
| Code: |
| FILE *fd = fopen(name, "rb"); |
and
| Code: |
| FILE *fd = fopen(argv[2], "wb"); |
and it works! tnx for all help everyone! :)