|
by Nigel Spratling, President of Echolab with Dale Matthews
|
Everything in video is compressed. The first place you see this is in how your original material is recorded. In the case of the HDCAM, it’s mildly compressed to probably 5 to 1. The P2 HD uses mild to medium compression at about 8 to 1. The P2 HD can be set somewhere between 10 to 1 and 20 to 1. The XDCAM is heavily compressed at almost 100 to 1. And HDV is compressed to almost 200 to 1. And that’s just the compression taking place in the camera. First, a word of caution when selecting cameras: check the native resolution of the camera imager. All HD cameras will output either 720 or 1080 formats. However, the resolution of the imager may be neither! Many of the images are 1440 by 1080 – or some other number by 720, but not actually the same resolution. You need to make sure you are actually outputting 1920 by 1080. 720 sources need to be upmscaled for display on 1080 screens. In fact, they have to be scaledmup more than twice. And, of course, you can’t get what is not there in the first place. If it’s 1080i, there’s no display scaling required. Scaling refers to when the pixels are resized or resampled to match the output display characteristics. De-interlacing 1080i to 1080p introduces motion aliasing; there are odd effects generated by the de-interlacing. 720 sources need to be upscaled; they actually have to be scaled up more than twice. And, once again, you can’t get what’s not there in the first place. So you’re simply upscaling to a picture that’s effectively defocused. That’s because you’re repeating pixels – or you’re applying math to pixels – to try to interpret something that wasn’t there.
|
