8-Bit vs 16-Bit Images

The term bit is common in any form of digital media. With respect to digital imaging, bit depth goes by many names like -pixel depth or color level. In digital photography, the discussion of 8-bit vs 16-bit files has been going on as much as Nikon vs Canon. This short article intends to give you a better understanding of what bit depth is certainly. It will also take you by means of whether we need 16-bit images or not, and if we perform, when we need them.

Desk of Contents

What is bit depth?

Many of us are aware of the fact that pixels are basic elements of any image. Specifically, any color within digital imaging is symbolized by a combination of red, green, and blue shades. One particular combination is used per pixel, and millions of pixels make an image. It is for this reason that bit depth is also called color depth. For example , 100 % pure red is represented using the numbers “255, 0, 0. ” Pure green is usually 0, 255, 0, plus pure blue is 0, 0, 255. In photography, each primary color (red, green or blue) is definitely represented by an integer between 0 and 255. Any non-primary colors are usually represented by a combination of the primary colors, such as “255, 100, 150” for a particular shade of pink.

Let us consider the largest quantity that represents red, which is 255. When I convert 255 into binary, I get 111111, which is eight numbers long. Now, when I try to convert the next decimal, 256, I would get 1000000, that is a 9 digit binary amount. That is why any integer in between 0-255 is considered “8 bit”; it can be represented within eight binary digits.

So , the definition of little bit depth is the number of bits used by each color component to represent a pixel. For instance , 8 bits can symbolize up to 256 shades (or, 2^8) of a given principal color.

Bit depth vs Color gamut

Some photographers confuse colour depth with color range. Color gamut is a selection of colors, usually used in the context of which range of shades a given device can display or printer can output. Electronics and printers are not able to display nearly as many colors since the human eye can see. The range of colours which they can display is usually restricted to a color gamut like sRGB or AdobeRGB, or a specific gamut based on the printer/ink/paper at hand. You can read more about color gamut at Spencer’s write-up on sRGB vs Adobe RGB compared to ProPhoto RGB .

Bit depth, however, can be visualized as the distance between colors within the gamut. In other words, you could have two pictures of rainbows that each go from red to violet – i. electronic., the same gamut. But the very first rainbow may be a gentle lean with many thousands of individual colours if you zoom in at the pixels, whereas the second range may be made up of just seven or eight colors and appear much blockier. In that illustration, the second rainbow would have a far smaller bit depth.

1-Bit pictures

In order to visualize bit depth easier, let us take a simple sort of a 1-bit image. As you may have gathered already, little bit depth is merely 2 to the power of that number. So , a 1-bit image might have only 2^1 values. Since 2^1 = 2, there are only two values accessible here: 0 and one – AKA black and white.

_V5A7411F
1-bit image that only offers pure black and genuine white

Take a look at the image below for the similar example. The left side of the image is usually 8-bit whereas the right part is 1-bit.

8B1B
8-bit vs 1-bit

The right side of the image contains only black and white. A few areas of the particular 1-bit image might show up grey, but once bigger to pixel peep, the difference becomes apparent as observed below. The 8-bit picture can hold 256 shades of grey whereas the image on the right can only hold either black or white.

8B1BPixel
8-bit images enable up to 256 tones, while 1-bit images can have just two

Bits vs Pieces per channel

In the above area, we saw that an 8-bit image can only hold 256 different shades of gray in total. But I described at the start of this article that 8-bit color images actually have 256 shades  per principal color . So , a typical color image that we commonly call “8-bit” actually can fit well more than just 256 shades. It’s more accurate in order to call it an 8-bit  per channel   image. If your color image has 8 bits per channel, and there are 3 channels (red, green, plus blue), the overall image can actually fit a total of 256 × 256 × 256 shades, which equals 16, 777, 216 (or 2^24). That’s why you may from time to time hear an 8-bit per channel image referred to at a 24 bit image, despite the fact that this is not the most commonly used term for it.

Nevertheless confusing? Let me take the assist of Photoshop to make it crystal clear. Take a look at the illustrative image below.

RGB_Channels
The image has three channels, every channel has a bit level of 8-bits.

In the Channels tabs, marked red in the picture above, you can see that although this is a greyscale image, they have four channels: one station each for red, green, and blue, and a good RGB channel for the entire picture. It’s not possible to know whether or not I can recover the color picture in this case (for all we know, I applied a B& W adjustment layer and flattened the image). Yet at least in some form, presently there remain three primary color channels here, and each one has eight bits of information.

As such, the entire picture here is technically still 24 bit. However , I could remove all color information by going to the top menu and selecting Image > Mode > Greyscale. Once I do, you will see that only one channel exists today, as shown in the image below:

GreyscaleChannels
As soon as an image is converted into the greyscale image, it will no more hold 4 channels but only one.

The picture above is a true 8-bit image; there are only 256 shades associated with gray in this photo, and there is no way to get back the color version. This also reduced my quality to 1/3 of what it was before.

16-bits/channel or 48-bits RGB

Now that you understand bit depth, you can easily calculate the little bit depth of 16-bits per channel images. An image along with 16 bits per station will have up to 2^16 tones per channel, or 65536. If you have an RGB image where each of Red, Eco-friendly, and Blue has sixteen bits, you must multiply 65536 ×  65536  ×  65536 to see that the image holds up to 281 trillion colors in total.

Even though theoretically, 16-bits/channel bit level is supposed to hold 281 trillion colors, Photoshop’s 16-bit does not hold that much. As per description, the maximum possible tonal worth for each of the primary shades should be 65, 536. However the maximum possible number of tones in Photoshop’s 16-bit/channel RGB is (2^15)+1=32769. So when you are working with Photoshop in 16-bit mode, a pixel holds any of 35. 2 trillion colors instead of 281 trillion.

Will be 16-bits/channel really usable?

Even though Photoshop’s 16-bit/channel images can only hold 12. 5% of the theoretical maximum value, 35. two trillion colors is still a great deal. The million dollar question that will arises now is, can a persons eye resolve so many colours? The answer is NO. Research has demonstrated that the human eye can resolve a maximum of 10 million colors. Take a look at the image below.

Reds
Shades of 8-bit red

Can you see any noticeable difference between the three rounded squares? Most of you might spot the tonal difference between the a single in the middle and the one in the right. But I certainly cannot find any visible difference between the left 1 and the middle one.

The leftmost sq . is 255, 0, 0, while the middle square can be 254, 0, 0. That is one step of difference in an 8-bit image, no place near even Photoshop’s 16-bit images!   Had the above mentioned image been a 16-bits/channel image in Photoshop, you can fit more than 32, 500 tones between the left and center images.

Since 16-bits/channel images hold an exceptionally large number of colors, these people obviously are space consuming. For example , Nikon’s NX software program outputs 130 MB JPEG files when I choose to foreign trade it as 16-bit, while, the file size shrinks to about 70 MB after i choose 8-bit with one of my images.

In addition , very few output products – monitors, prints, and so forth – can display more than almost eight bits per channel anyway. But that doesn’t imply the higher bit depths are usually unimportant.

Where does 16-bits/channel really matter?

The section above might give an impression that nobody would ever need a lot more than 8 bits per route. Nevertheless, 16-bit images have their uses. Let us consider the picture below.

Curve1_2

I have opened an image plus converted it into 8-bit by using the menu option Picture > Mode > 8-bits/channel. Now I apply two curves adjustment layers to the opened up image. In Curves one, I select input as 255 and change the output to 23. To put it simply, I have underexposed the picture. Using Figure 2, I have selected the particular input as 23 and changed the output to 255. This brings back the exposure to where it was before underexposing it – but at the expense of “crunching” plenty of colors. This leads to the banding effect that you can see above and clouds in  the image above.

Once i do the same edit to some 16-bit image, there is no visible banding in the sky. You can see that will in the comparison below, exactly where I put both images through the same adjustments:

8B16B_1
8-bits/channel image displays noticeable banding which is nearly avoided in the 16-bit

This is where 16-bit images find their make use of. The more drastic your modifying is, the more helpful it will be to have as many shades associated with color as possible.

You can still avoid banding on 8-bit images with careful processing – for example not doing the extreme curves adjustments I did above – but 16-bit images give you more room for mistake. That’s why, if you’re editing in software like Photoshop, it is good practice to work alongside 16-bit images. Only once the particular editing work is done could it be a good idea to convert it to an 8-bit image for result. (Although it’s still best to keep the 16 bit JPEG or PSD in your store, in case you decide to do more editing later. )

So , in general, the particular useful scope of 16-bit per channel images starts and ends with publish processing.

Conclusion

I hope this article gave the readers a basic understanding of exactly what bit depth is as well as the difference between 8-bit and 16-bit per channel pictures. Even though 16 bits might sound like overkill, we saw here that it finds the use in post-processing images. But 8-bit per channel pictures take up much less file space, so it’s worth exporting your images, especially for the web, to 8-bit per channel to save space.

Make sure you let me know in the comments area if you have questions or enhancements so that other readers can benefit from it.

Leave a Reply

Your email address will not be published. Required fields are marked *