SDI Technologies iH30 Especificaciones descargar pdf (Pagina 18)

Data Transfer

Quadro SDI Output PG-03776-001_v06 | 13

Code Listing 7: Selecting the GPU that is connected to the Output

Card on Windows

while(wglEnumGpusNV(GPUindex,&GPUHandle)) // First call this

//function to get a handle to the gpu

{

//get detailed GPU info

while(wglEnumGpuDevicesNV(GPUHandle,DisplayIdx,&gpuDevice))

{

NvAPI_GetAssociatedNvidiaDisplayHandle(gpuDevice.DeviceName,

&hNvDisplay);

NvAPI_GetPhysicalGPUsFromDisplay(hNvDisplay,&hNvPhysicalGPU,

&count);

if(m_hPhysicalGpu == hNvPhysicalGPU)

{

//save the GPU affinity GPU handle to use

// for creating the OpenGL context

m_gpuHandle = GPUHandle;

break;

}

…

}

//Now an OpenGL context can be created

HGPUNV handles[2];

handles[0] = hGpu;

handles[1] = NULL;

// Create Affinity context

hDC = wglCreateAffinityDCNV(handles);

setPixelFormat(hDC);

// Create rendering context from the affinity device context

hRC = wglCreateContext(hDC);

// Make the affinity context current

wglMakeCurrent(hDC, hRC);

On Linux things are much simpler since all the important

NVCtrl X extension calls and

the GLX context creating call are already using an X screen as one of the parameters.

5.2 BUFFER OBJECT INITIALIZATION

Use of the GL_NV_present_video extension to send GPU rendered content to the SDI

device requires the use of one or more buffer objects. These buffer objects can be texture

objects or renderbuffers bound to one or more frame buffer objects (FBOs). Creation of

an FBO with a texture or render buffer attachment is demonstrated in Code Listing 8.

1 2 ... 13 14 15 16 17 18 19 20 21 22 23 ... 146 147

Sin comentarios

SDI Technologies iH30 Especificaciones Pagina 18