Bit depth is a topic that always comes up whenever a new camera is announced. But what does it all actually mean? And why is it important? Given that most monitors can’t even display 14-bit images and jpg files are 8-bit anyway, is it even important at all? In this video, Matt Granger explores the topic, explains what it means, and why it’s relevant for your images (and video).
A bit is essentially a 1 or a 0. Each time you add a bit, you double the amount of data saved. 1 bit is a simple on or off switch. 2 bits is two of them, which offers four possible combinations. 3 bits offers 8 choices. 4 bits, 16, and so on. In a simple black and white image, this equates to different shades of grey available for use in your image.
For colour images, this number is multiplied by three, once for each of the colour channels red, green and blue. When we talk about bits in relation to images, we’re typically talking about the number of shades of a colour each of those channels can hold. In the case of 8-bit, that means 256 options from pure black to pure red, 256 options from black to green and 256 options from black to blue for a total of 16.7 million possible combinations of colour.
That’s a lot of colours, right? But our cameras are capable of producing 12-bit, 14-bit or even 16-bit RAW images. That’s way more than our monitors can handle, way more than we can see in print, and offers a level of colour resolution far greater than even our eyes can distinguish. So, why do we need more than 8 bits? Especially when most of the information gets thrown away when we save out as an 8-bit jpg?
Largely it’s down to editing, as Matt explains. Having that information available to us, even if we can’t see it, allows us to bring it out in the image and enhance it. A higher bit depth allows for more information in the shadows and highlights, so that if we’re shooting a high contrast scene, we can pull back that highlight and shadow detail to make it more prominent.
If we’re only shooting 8-bit, editing in 8-bit and then putting out an 8-bit file, getting that hidden highlight and shadow detail back is impossible because it simply doesn’t exist in the source image.
Even with basic editing, shooting files with a higher bit depth and processing it yourself manually allows you to increase the overall quality of the image, particularly in areas of subtle gradients, like skies, which can often cause banding when shooting 8-bit.
The topic as a whole gets quite in-depth, so if you didn’t quite get what I said above, have a watch of Matt’s video where he shows some examples, and talks about the implications of bit depth in images.
This also applies to video, too, which is why the Blackmagic Pocket 4K and Pocket 6K have become so popular by offering 12-bit BRAW video. And it’s why people are still complaining that new Sony mirrorless cameras only shoot 8-bit video while the rest of the world is going 10-bit or higher.
How important is bit depth to you?
[via FStoppers]
For me this is a threshold thing. If I was buying a stereo I’d listen to them from cheapest to more expensive and when I could no longer tell them apart, that’s the one I’d buy. Spending more on something I can’t hear is foolish. In this case, for me, I don’t out put anything over 1080p. Even spending money for 6K is not needed, so worrying about anything over 8 or 10 bit is also not worth investing in.
You can almost always hear the difference between stereo systems; the issue is whether one system (mostly determined by the speakers) sounds more realistic to your ears. This involves extended listening sessions and a variety of music that you know and like to listen to. Listening fatigue is also a factor. A lot of speakers that grab your attention and have dramatic sound end up being tiring to listen to over a couple of hours, just making you want to turn your stereo off. Really good stereo systems draw you into the music, keeping the listening experience enjoyable.
I agree, the décolletage of the lady in the picture is a bit deep but I didn’t understand the explanation.
Detlev Deninger bit is like on-off switch. In 16 bit you have 16 “switches” and if you count how many possible variations there can be on and off… it is a lot.
1 bit can be on or off. 2 bit can be on-on, on-off, off-on, off-off.
Every time adding 1 bit more, doubles possible variations.
I take images that I want big prints of to an art supply and framing store in TIF format. Even with their excellent monitors and very good printing processes my images sometimes have details out of gamut. They still produce really nice prints though.