Screen streaming

CopyFromBuffer(self, data, format=BitmapBufferFormat_RGB, stride=-1)
Copy data from a buffer object to replace the bitmap pixel data.
Default format is plain RGB, but other formats are now supported as
well.

// Grabs the entire desktop. Make sure to set video_alpha to 0 before and to 100 after,
// to grab the desktop without the engine. Returns a BMAP* with the grabbed desktop.
//
BMAP* bmap_for_desktop ()
{
BMAP* bmap = NULL;

// get the device context of the entire desktop, including windows bar and everything
HDC hDesktopDC = GetWindowDC(HWND_DESKTOP);
if (hDesktopDC != 0)
{
// get desktop size in pixels
int desktopSizeX = GetSystemMetrics(SM_CXSCREEN);
int desktopSizeY = GetSystemMetrics(SM_CYSCREEN);

// creates a bitmap compatible with the desktop device context, in which we will later
// blit the desktop content into

HBITMAP hBitmap = CreateCompatibleBitmap(hDesktopDC, desktopSizeX, desktopSizeY);
if (hBitmap != NULL)
{
// create a memory device context compatible with the device context of the desktop. This
// context will be used to select the blit target
HDC hBitmapdc = CreateCompatibleDC(hDesktopDC);

// select the target bitmap into a the desktop-alike device context and copy the desktop
SelectObject(hBitmapdc, hBitmap);
BitBlt(hBitmapdc, 0, 0, desktopSizeX, desktopSizeY, hDesktopDC, 0, 0, SRCCOPY);

// create Gamestudio bitmap with size of the desktop
bmap = bmap_createblack(desktopSizeX, desktopSizeY, 888);
if (bmap != NULL)
{
// lock bitmap to blit from windows bitmap to Gamestudio bitmap
var format = bmap_lock(bmap, 0);
if (format > 0)
{
// we blit the bitmap pixelwise by fetching and converting the RGB components from the
// Windows bitmap and throwing it with pixel_to_bmap into the Gamestudio bitmap. This
// procedure might be faster by using GetDIBits and bmap->finalbits, but this seems to be
// safer for now

int iRow, iCol;
COLORREF colorRef;
COLOR color;
var pixel;

for (iRow = 0; iRow < bmap->height; iRow++)
{
for (iCol = 0; iCol < bmap->width; iCol++)
{
// retrieve the RGB color value as of the pixel at the specified coordinate as hexadecimal
// value 0x00bbggrr. We use the standard GDI macros to extract the component
colorRef = GetPixel(hBitmapdc, iCol, iRow);

if (colorRef != CLR_INVALID)
{
color.red = GetRValue(colorRef);
color.green = GetGValue(colorRef);
color.blue = GetBValue(colorRef);
}
// else: the pixel is outside of the current clipping region

// convert and set the retrieved color to a pixel in the format of the Gamestudio bitmap
pixel = pixel_for_vec(&color, 100, format);
pixel_to_bmap(bmap, iCol, iRow, pixel);
}
}

// blitting done, unlock Gamestudio bitmap
bmap_unlock(bmap);
}
}

// lets delete the desktop-alike device context
DeleteDC(hBitmapdc);
}

// an application must not -delete- a DC whose handle was obtained by calling the GetDC
// function. Instead, it must call the ReleaseDC function to free the DC. If such a DC is
// not freed, serious effects on painting requested by other applications can happen!
ReleaseDC(HWND_DESKTOP, hDesktopDC);
}

return(bmap);
}

wxBitmap is intended to be a wrapper of whatever is the native image format that is quickest/easiest to draw to a DC or to be the target of the drawing operations performed on a wxMemoryDC. By splitting the responsibilities between wxImage/wxBitmap like this then it's easier to use generic code shared by all platforms and image types for generic operations and platform specific code where performance or compatibility is needed.

class RGBValue
A simple class which stores red, green and blue values as 8 bit unsigned integers in the range of 0-255.

I know it'll work well because its a stream and pixel perfection isn't needed, image formats are better suited for a still image, right?

Ok, but what do you mean I don't want bitmaps, exactly? Sorry, but I'm new to the D3D stuff
I can still safely use GetPixel(hdc,col,row) to access the raw data and pass it pixel by pixel to be encoded, or is there some sort of array that holds all this data?

PS.: And, another quick question, haven't asked google yet, but how do I get the device context that the Acknex uses? Didn't find any pointers in the manual... Sorry if this is a stupid question.

PS.: And, another quick question, haven't asked google yet, but how do I get the device context that the Acknex uses? Didn't find any pointers in the manual... Sorry if this is a stupid question.

The LPDIRECT3DSURFACE9 pointer to the created z buffer is available through the render_zbuffer variable.

PS.: Malice, you're suggesting to create a bmap_zbuffer, but isn't bitmap creation slow? I also don't have a clue how render targets work. Used them a lot of years ago, and I have no memory of that process.

#include <acknex.h>
#include <d3d9.h>
void main()
{
.............................
BMAP *bmpMap = bmap_createblack(1920,1200,32);
BMAP *bmpScreen = bmap_createblack ( screen_size.x, screen_size.y, 24 );
bmap_zbuffer(bmap_createblack(2048,2048,32));

while(1)
{
bmap_rendertarget(bmpMap,0,0);
GetDC(render_zbuffer);
bmap_rendertarget(NULL,0,0);
wait(1);
}
}

I'm trying several methods to capture screen with DirectX and I've come up with 3 methods so far:
•GetFrontBufferData() - Average execution times: •GetFrontBufferData(): 0.83598 s
•D3DXSaveSurfaceToFile(): 0.0036 s
•Total: 0.83958 s

•GetBackBuffer() - Average execution times: •GetBackBuffer(): 0 s <-- INTERESTING - WHY?
•D3DXSaveSurfaceToFile(): 0.2918 s
•Total: 0.2918 s

•GetRenderTargetData() - Average execution times: •GetRenderTargetData(): 0.00928 s
•D3DXSaveSurfaceToFile(): 0.00354 s
•Total: 0.01282 s


Average times have been computed by taking 50 screenshots and by measuring time with clock().

...
Error in 'features.h' line 173:
< #if (!defined __STRICT_ANSI__ && !defined _ISOC99_SOURCE && \
>
.. 0.436 sec
Error compiling...

I understand those libraries come by default with IDEs, so its normal not to have them with our SED, right? How can I add them, or something?