The present and future of post production business and technology

Are Pixels Doomed?

Pixels – those little dots that make up all our video images – are hard to encode and push down pipelines, even with ever-increasing encoding efficiency. On the other hand, vectors are small and very efficient, but so far have proved difficult to apply to video content.

The team behind a new project - University of Bath, Root6 Technology, Smoke & Mirrors and Ovation Data Services – have launched a new vector-based video codec and are claiming that the pixel will be dead within five years. That’s a very bold claim.

The difficulty is defining vectors to represent the intricacies of the image. Imaging trying to create a photo-real image in Illustrator, versus in Photoshop? That’s the challenge. But the pay-off is huge:

 The researchers at Bath have developed a new, highly sophisticated codec which is able to create and fill between contours, overcoming the problems preventing their widespread use. The result is a resolution-independent form of movie or image, capable of the highest visual quality but without a pixel in sight.

One encode, a small file size, and plays back sharp at all sizes from very tiny to huge. I can’t wait to see if this plays out. If it does it will solve those distribution problems that Dan Rayburn was talking about the other day – the Internet has plenty of bandwidth for vectors!






5 responses to “Are Pixels Doomed?”

  1. Tim Johnston

    Oh man, yeah. It’s funny, I was actually thinking about this today as a solution to the upres/downres problem. Yeah, the pixel (in terms of video encoding) will be totally dead. Of course, screens will still be pixel-based for a while. But with this type of video, so will be dead the concept of the ‘frame’. If you can move the vectors and reference points through infinite space, along continuous time intervals, there’s no longer any need for frames, or frame rates. Whaetever playback device you have will just interpolate for its own screen’s frame rate.


    Great news! Wait… will FCPX support it? 😉

    1. Philip

      As I understand it, anyone can make an input module (aka codec) so I guess it will be up to the developers

  3. Patrick James

    As a UK person I’m quite pleased to read this research has been taking place at the University of Bath.

    Unfotunately in the UK we have a history of developing things, but once developed, they are taken up by overseas companies. I think this is perhaps the short-sightedness of UK companies.

    So I hope that this time UK companies look at this and have the imagination to see how they could exploit the technological development.

  4. Mr. Francis Zuccarello

    Before you start to write the obituaries, you ought to show us the entire image, and not just the wireframe, or whatever you might call it. Thanks, and may everybody rest in piece.