doesn't it seem a bit strange that you have to construct your own differences? I've always wondered why there isn't a video interface that just sends compressed deltas. the number of devices that need to actually scan is pretty small at this point.