This video is available Free for Everyone.
Running Time: 3 minutes
Digital Fundamentals Tutorial – What is the Difference Between 8-Bit and 16-Bit Color Depth Images?
This Digital Fundamentals tutorial is on the difference between editing images in 8-bit color and 16-bit bit depth. While it should be obvious that 16-bit color is twice as good as the 8-bit color, it just doesn't seem that way to the naked eye. This is because humans can only see less than 10 million colors. Since 8-bit already exceeds 16 million colors, why in the world do we need to work in a bit depth that moves us into trillions of colors? Visually, they both look the same, why use the higher resolution?
This very quick Digital Fundamentals tutorial is only a few minutes long, but guarantees you will never work in the 8-bit bit depth again! With two adjustments using the levels tool, you will see that using the 16-bit bit depth is the only way to go, if you want to achieve the best results possible.
Once you have watched the tutorial, I am sure you will be convinced. Then what should you do to properly make an 8-bit to 16-bit conversion? Well, thats easy. just go up to the pull down menu and select Image – Mode – 16-Bits/channel. Suddenly, you are whisked away to the wonders of color perfection. But, wait, you ask… what about that part, just below, that says 32-Bits/channel? Why talk about 16-bit when we can clearly select 32-bit and bask in all its glory? Well, settle down there, sparky. One thing at a time. I mean, if 16-bit is already trillions of colors, and can save the example in the video tutorial, isnt 32-bit just complete overkill? I sure think it is. But then again, who am I?
Once we have turned the 8-bit image into a 16-bit image, then we can make all of our Adjustment Layers and save our layered PSD or TIF file. When we want to send the file off to the prepress house, or magazine printer, simply go back to Image – Mode – 8-Bits/channel and save the flattened file for them. At this point, all the color changes are done, so no need to have all that extra data. It was only needed to avoid banding or other color corruption before the CMYK conversion.
So then, why do cameras still work in 8-bit bit depth? This is because 8-bit bit depth is old school, and the only one supported by JPGs. If a camera takes a picture, it is good with 8-bit. It does not need to (or can not) make the extra jump to 16-bit. Don't forget, we can only see a fraction of the 8-bit bit depth anyway. The issue comes in when we, the retouchers, start messing with things, and accidentally cause the banding. The camera did a good job, the first time around.
Questions and Answers
- What is 8-Bit Depth in Photoshop?
Each Color of an image in RGB (Red, Green, Blue) is created from a gradient from black to white. There are 256 shades of grey between Absolute Black and Absolute White. This means that Red has 256, Green has 256 and Blue has 256. The math is 256x256x256=16.7 Million Colors possible. However, it is actually possible to see the 256 shades (steps). If you can see them, it is refrerred to as banding.
- What is 16-Bit Depth in Photoshop?
It is the same as 8-Bit, except that you now have 65,536×65,536×65,536=281 Trillion Colors. Now, I am sure you can image that with only 256 shades of grey, you can see the banding (sometimes). With 65,536, banding has been eliminated as a possbile limitation in your images.
- Why use 16-Bit in Photoshop?
To Avoid Banding.
- Which is better, 16-Bit or 32-Bit Depth?
Obviously, 32-Bit will hold many more colors. However, it is this author's professional opinion that 32-Bit Depth is overkill, and leave you with huge files with little benefit. If you add a few adjustment layers, it can bring your computer to a severe slow down.
- Should CMYK be 8-Bit, 16-Bit or 32-Bit Depth?
You can work in 16-Bit RGB, but since CMYK is the last step before sending the image to a printing press, it may as well be 8-Bit. Banding will need to be addressed then.
- Why does Photoshop default to 8-Bit Depth, when 16-Bit is better?
The best reason is that several of the Filters, and 3rd Party Plug-Ins only work in 8-Bit. The better question is "Why isn't there a Preference to Default to 16-Bit Depth?"
- When should I use 8-Bit Depth Images?
There are a few times when 8-Bit is needed; JPGs can only be 8-Bit, when you need to deliver samples to your client, several Filters are only 8-bit, and when you need to convert a file to CMYK for a printing press. These are just examples off the top of my head, I am sure there would be some more.
This Digital Fundamentals Tutorial answers the question: What is the difference between 8-bit and 16-bit, bit depth? This tutorial is also valid for older versions of the Creative Suite series, including CS, CS2, CS3, CS4, CS5 and CS6. If you would like to learn more about Photo Enhancement from a Professional Retoucher, I offer a Photoshop for Beginners Class. Please contact me today, and I will be able to add you to the schedule too. If you would just like to watch online videos, The Art of Retouching Studio offers many Photoshop Tutorials for Beginners and Advanced users.
Very well explained. I really learned a lot reading this. I saw that video tutorial like five minutes ago but with this explanation is soo much better!
Thank you! Being able to see the banding in the video really helped with the understanding – very well explained!
Thanks for the article! I bookmarked this site and plan to come back often. Good work!
Thanks this video really help out, read an article about 8 bit vs 16 bit for 15 mins. You explained it in 3 min.
Thanks again
Can a 8 bit image be converted to 16 bit or do we need to go back to the .RAW and start over from it?
You can make the 8-bit image 16-bit, and continue editing, yes. This will prevent NEW banding from occurring. If the banding is already evident in the base image, then it will not help.
very well explained …
thank you
Hi John,
Thank you for this interesting tutorial! I need your advice please. I am doing graphics for an iphone app (stickers) where images will be either in black or in white (black trees or white silhouettes). What about the depth? 8 bit or 32?
Many thanks
Regards
Adana
I think the correct answer, in your situation, would be 1-Bit Depth. That means if you take your RGB color image, and make it GreyScale; you can further go down to Bitmap, which is only Black and White, no shades of grey. If you make it a 50% threshold, at a higher than normal Resolution (normally around 300 PPI) and make it 600 or 1200 PPI during the conversion, I think you would get what you were looking for. After that, you can save the final file in the format and resolution you normally would. This is, of course, a trickier answer than you were looking for, but I think it would be the correct one.
Thanks for the words. A request – I am unable to check the video – says the link is nonexistent for the video to stream. Is that right or is it just my system?
Thanks fr your comment. It led me to discover a large number of my back catalog videos were offline. I have no idea why, but it has been fixed.
Hi John,
An amazing read indeed.
Would you suggest storing images as a 16bit tiff file or would it be ok to covert the file to a 8 bit Jpeg after all the post processing has been finished with ?
Assuming its fine to finally convert the 16 bit tiff to 8 bit jpeg:
Would i be able to convert an 8 bit jpeg sometime in the future back to a 16 bit tiff for print ?
And do i lose out in anyway in case i choose to preserve images in Jpegs (to preserve hard disk space) and later convert it to a 16 bit tiff just because i come across an occasional rare image that i might find print worthy ?
In general, RAW is long term store of the original. TIF is long term storage of layered files. JPG are samples for email distribution.
If your camera starts with JPG, you can store them as JPG. If your camera starts with RAW, your should store your images as RAW. If you have a layered file you want to save, store it as a 16-Bit TIF. If you really plan on never doing anything with the image, then you can store them anyway you want.
You simply need to keep in mind that RAW will give you much more flexibility later, should you ever want to do anything with the image. With JPGs, dark blacks are black, and blown out whites are white; there is no coming back. With RAW, you can usually save images like that.
You also asked if you can work in 16-Bit, save as 8-Bit JPG, then later back to 16-Bit before printing. This is of no use. Using 16-Bit is intended to avoid banding when making color corrections. Once you make it 8-Bit, you are all done. Make it 16-Bit again will do nothing, unless you edit it more, and then you should have the 16-Bit TIF to go back to.
Thanks………..
Hilarious…
“What is 16-Bit…”
“The same as 8 bit, except…”
“The better question is “Why isn’t there a Preference to Default to 16-Bit Depth?”
Indeed! Now ask it and answer it!
How should I answer it? “Because Adobe Says So?” Once they get a good idea, it takes them years to implement it. While I sit here trying to justify 8-Bit by default, I really can not. When I teach my classes, I can visually create banding on command, just by using levels and saturation set to 8-bit. It’s clear as day. When I do the same thing with 16-bit enabled, there is no banding to be seen. The only time I work in 8-Bit is if I have a full size trade show display (ie: it’s huge), or when I simply forget to make it 16-Bit (there-by, going to back to my original point).
Thank you very much, this video has been so helpful, very useful
Dear John,
I’m an illustrator using photoshop to paint with, and I’ve always used 8-bit just because all the filters, etc. are available in that setting. However, I also noticed that large color spaces with gradients (like sky) have always had very slight banding. Now I know how to eliminate that, I’ll just paint in 16 bit instead, woo! (thanks!)
However, when I’m all done, I usually switch to cmyk, deal with some hue changes, and send it off to print as .psd.
If i’m working in 16/bit rgb and switch down to 8/bit rgb or 8/bit cmyk, won’t that limit my color range and cause banding then…? I am confused why you’d convert down to 8/bit mode after if 8/bit mode causes banding.
Hopefully you can help me out here =/
Also, i guess…why can’t i just print in 16/bit mode?
You work in 16 to avoid banding, while creating gradients, adding adjustment layers, etc. Only at the very end, before sending to press, will you make it 8-bit, and get a “clean” version of your layered file. If at that point, banding is still an issue, Add Noise of 2-3% monochrome gaussian. Note that on screen, it will be more obvious than on press. I always found 3 to do the best job, even though it looks pretty bad on screen, press does not see it that way.
You can send it to press as 16-bit, but the RIP (Raster Image Processor) will automatically convert it to 8-bit anyway. This is because the color separated plates need to be dots. So it’s better to be done in your hands, rather than the computer’s (which wont care about the banding).
If you are working on your local inkjet printer (Epson, Canon), then leaving it 16-bit before sending is OK. You are not faced with the same technical challenges as you are on press.