display | more...
Now that we have ubiquitous video compression for network transmission, the ability to handle a dropped frame is a requirement for multimedia software.

Fast internet transmission is not reliable. You can slow it down and make it more reliable, but producers are starting to choose getting most of the data on time rather than getting all of the data a little late.

Current encoding standards tend to encode a full frame, and then only the portions of that frame that change over the next X frames, where X can vary greatly (1 to hundreds). The effect is much less data to transfer, but more importantly you can cut the video at any point, and regain the full frame when another one comes around, or build the video up with the updates before the full frame comes. You can drop pieces of the updates and it only affects a small portion of the image. There are methods of encoding the image at several different points consequtively, so if a packet is dropped the overall quality suffers but the missing parts aren't localized to one part of the screen.

Certianly dropped frames aren't good in general, but recent developments have placed much less stress on this point than used to be the case. Of course, eventually, we'll have no dropped frames on real-time network transmissions, but the technology in use to deal with it now is tremendous.

Log in or register to write something here or to contact authors.