cudaDeviceProp prop;
	int device = 0;

	cudaGetDeviceProperties( &prop, device );

	printf( "\nGPU #%d: \'%s\' @ %d MHz\n", (device+1), prop.name, (prop.clockRate/1000) );
	printf( "   Memory Bus Width: %d-bit @ %d MHz ==> ", prop. memoryBusWidth, prop.memoryClockRate/1000 );
	printf( "   Max Bandwidth: %u GB/s\n", (prop.memoryClockRate/1000 * ((prop. memoryBusWidth/8)*2))/1000 ); // DDR2/3/4/5 = *2
	printf( "   Texture Size 1D: %d          \n", prop.maxTexture1D                                                );
	printf( "   Texture Size 2D: %d x %d     \n", prop.maxTexture2D[0], prop.maxTexture2D[1]                       );
	printf( "   Texture Size 3D: %d x %d x %d\n", prop.maxTexture3D[0], prop.maxTexture3D[1], prop.maxTexture3D[2] );
	printf( "   Mem: %u MB (%u bytes)\n", (prop.totalGlobalMem/1024)/1024, prop.totalGlobalMem );

	printf( "\nCUDA: %d.%d\n", prop.major, prop.minor );
	printf( "   Threads/Block: %d\n   MaxThreadsSize: %d, %d, %d\n   MaxGrid: %u, %u, %u\n",   // 1024   
		prop.maxThreadsPerBlock,
		prop.maxThreadsDim[0], prop.maxThreadsDim[1], prop.maxThreadsDim[2],
		prop.maxGridSize[0], prop.maxGridSize[1], prop.maxGridSize[2]
	);
	printf( "   Memory/Block: %d\n", prop.sharedMemPerBlock );
	printf( "   Registers/Block: %d\n", prop.regsPerBlock );
	printf( "\n" );
	printf( "   Multi Processors: %d \n", prop.multiProcessorCount );
	printf( "   Max Threads/Proc: %d \n", prop.maxThreadsPerMultiProcessor );