We all know that when it comes to resolution, the higher the better. That’s what all the UHD TV vendors are telling us at least. Just walking down the TV aisle of an electronics shop, the crisp details of 4K resolution are evidence enough. Why isn’t everything 4K already then?
Putting aside the lack of genuine 4K content, the sad truth is that when it comes to compression, higher resolution isn’t always better. And strong compression cannot be avoided, since without that, there are no means to distribute the content to the living rooms.
For example, using technologies like H.264 (MPEG), when it comes to the choice of either compressing the video at native resolution (like 2160p) or downscaling it into a lower resolution (like 1080p) first and then compressing it, the outcome can be as surprising as this:
That’s right, using lower resolution can actually improve the quality at lower bitrates, but how come? The world of compression technology is full of complicated dilemmas, and when working with a system with as many choices to make and levers to pull as H.264, it can be very hard to make the right decisions. (Not to mention the built-in inefficiencies of scaling, or the varying efficiency of motion prediction engines for higher resolutions etc.)
This issue is a source of pain even outside the world of compression. Should you just switch everything to a higher resolution, or should you keep all the lower resolution streams as well? Does it make any sense to upgrade your systems at all?
When choosing a compression technology, the question “How much bitrate do I need for X resolution?” is an especially relevant one. For some, it simply means: “what is the absolute minimum bitrate required to transmit my video, before the compression collapses miserably”?
If you asked an expert on legacy codecs what the minimum bitrate requirement is for 16K (16x UHD resolution) video, you could only get a very complicated answer which would lead to a set of new questions instead of a definitive answer.
We at V-Nova, however, would give you a very simple answer: “exactly as much as you use today”.
How is that possible? Have we found the holy grail of compression technology? No, all that it takes is to try and solve the right problems: providing continuous linear scalability.
Using PERSEUS™ technology gives operators peace of mind – they no longer need to worry about the side-effects of increasing resolution.
For a given bitrate, when faced with the challenge of upgrading to a higher resolution, with PERSEUS…
…You will never lose out by encoding at a higher resolution.
…You will often win, by seeing extra detail.
Knowing you as a very tech-savvy reader I need to admit this between you and I: yes, as you’ve spotted clearly there is a little “cheating” as far as extrapolation goes from a UHD result to 16K. In case you don’t believe statistics and need a solid evidence, well, stay tuned…
Written by Balázs Keszthelyi – Head of Pure Perseus R&D Group at V-Nova