I'm writing a QCPlugin to write images to disk (roughly based on the ImageWriterPlugin example). Capturing higher bit depth images is really throwing me though.
First I'm locking the buffer rep with [QCPlugInInputImageSource lockBufferRepresentationWithPixelFormat: colorSpace: forBounds:]. Colorspace/bounds I get from the image, pixelformat I'm using QCPlugInPixelFormatARGB8 for 8 bit or QCPlugInPixelFormatRGBAf for 16/32 (I suspect a problem here: isn't the 16 bit format in QC 16bit integer? 8/32bit should be OK though. And I'm not worrying about PPC support.) Then, I'm creating a CGImage provider, followed by a CGImage using CGImageCreate: cgImage = CGImageCreate([qcImage bufferPixelsWide], [qcImage bufferPixelsHigh], bitdepth, bitdepth * 4, [qcImage bufferBytesPerRow], colorSpace, bitmapInfo, dataProvider, NULL, false, kCGRenderingIntentDefault); I think that's all OK (with the exception of the pixel format being wrong for 16 bit, but with 16 bit being somewhat broken in QC I'm more concerned with 8/32 bit modes) apart from the CGBitmapInfo part. For 8 bit it's just kCGImageAlphaPremultipliedFirst, but what should I be using for the 16/32 bit modes? I thought "kCGBitmapFloatComponents | kCGBitmapByteOrder32Host" should cover it, but no matter what I try it seems to either not use float values or has wrong byte ordering :/ Thanks Chris _______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [email protected]
