A new study published in Nature by University of Cambridge researchers just dropped a pixelated bomb on the entire Ultra-HD market, but as anyone with myopia can tell you, if you take your glasses off, even SD still looks pretty good :)
A new study published in Nature by University of Cambridge researchers just dropped a pixelated bomb on the entire Ultra-HD market, but as anyone with myopia can tell you, if you take your glasses off, even SD still looks pretty good :)
I went looking for a quick explainer on this and that side of youtube goes so indepth I am more confused.
I’ll add another explanation for bitrate that I find understandable: You can think of resolution as basically the max quality of a display, no matter the bitrate, you can’t display more information/pixwls than the screen possess. Bitrate, on the other hand, represents how much information you are receiving from e.g. Netflix. If you didn’t use any compression, in HDR each pixel would require 30 bits, or 3.75 bytes of data. A 4k screen has 8 million pixels. An HDR stream running at 60 fps would require about 1.7GB/s of download wihout any compression. Bitrate is basically the measure of that, how much we’ve managed to compress that data flow. There are many ways you can achieve this compression, and a lot of it relates to how individual codecs work, but put simply, one of the many methods effectively involves grouping pixels into larger blocks (e.g. 32x32 pixels) and saying they all have the same colour. As a result, at low bitrates you’ll start to see blocking and other visual artifacts that significantly degrade the viewing experience.
As a side note, one cool thing that codecs do (not sure if literally all of them do it, but I think most by far), is that not each frame is encoded in its entirety. You have, I, P and B frames. I frames (also known as keyframes) are a full frame, they’re fully defined and are basically like a picture. P frames don’t define every pixel, instead they define the difference between their frame and the previous frame, e.g. that the pixel at x: 210 y: 925 changed from red to orange. B frames do the same, but they use both previous and future frames for reference. That’s why you might sometimes notice that in a stream, even when the quality isn’t changing, every couple of seconds the picture will become really clear, before gradually degrading in quality, and then suddenly jumping up in quality again.
For an ELI5 explanation, this is what happens when you lower the bit rate: https://youtu.be/QEzhxP-pdos
No matter the resolution you have of the video, if the amount of information per frame is so low that it has to lump different coloured pixels together, it will look like crap.
On codecs and bitrate? It’s basically codec = file type (.avi, .mp4) and bitrate is how much data is sent per second for the video. Videos only track what changed between frames, so a video of a still image can be 4k with a really low bitrate, but if things are moving it’ll get really blurry with a low bitrate even in 4k.
The resolution (4k in this case) defines the number of pixels to be shown to the user. The bitrate defines how much data is provided in the file or stream. A codec is the method for converting data to pixels.
Suppose you’ve recorded something in 1080p (low resolution). You could convert it to 4k, but the codec has to make up the pixels that can’t be computed from the data.
In summary, the TV in my living room might be more capable, but my streaming provider probably isn’t sending enough data to really use it.