I read in many places people saying that recording RAW pictures in 12 bit or in 14 bit do not make any difference in the quality of the picture. However, I am not certain to understand if this is true by just viewing RAW pictures with, for example, Lightroom or Photshop.
My question is : when converting RAW into JPG a 12 bit RAW file compared with a 14 bit file, will the JPG quality be exactly the same also ?
#1. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0stappy Nikonian since 06th Aug 2009Fri 22-Mar-13 09:35 PM
Short answer: no.
That said, a jpg file is an 8 bit image. So, in most cases the conversion from a 12 bit or 14 bit raw file to the jpg will be visually indistinguishable. Depending on the raw coversion algorithm and the image, they may in fact be identical. However, if the image is processed they can be quite different. This is especially true if the post processing includes exposure adjustments, pulling back highlights or pulling up shadows.
#2. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0km6xz Nikonian since 22nd Jan 2009Sat 23-Mar-13 04:33 AM
You likely will not see a difference, not due to not being there but because the way you are viewing it. Monitors and JPG have even less tone range than a 12 bit file contains data for.
RAW 12bit captures 4096 individual steps in luminance while the 14 bit files has a finer step value of 16384 individual values spanning the same dynamic range. The addition detail information does not increase the dynamic range but does increase the number of steps you can extract information from immediately before the hard cut off of clip occurs. Between the clip level and the highest level before clipping a 12 bit file, 14 bit has 4 addition steps carrying information.
JPGs store rendered images as 8 bits per channel per pixel: 8 bits of red, 8 of green, 8 of blue, 24 bits in total. RAW stores 12/14 bits per channel per pixel: 36/42 bits per pixel in total.
With higher and higher res monitors and printers coming into use all the time, older files should be showing more difference between 12 bit and 14 bit.
#4. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
Another aspect, as I understand it and has perhaps been alluded to, is that the greater bit depth allows for heavier processing. You don't lose as much information as a percentage of the overall information available when you save an edited file.
Powder Springs, GA, USA
Nikonians Team Member
#5. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
It's the shadows that benefit the most from 14-bit, primarily if you edit the image to pull additional detail from them. This is the part of the raw file that has the least precision, so the extra bits help the most in this tonal zone. Coincidentally, this is also one of the D800's strengths, so if you really want to harness its full power in terms of dynamic range, 14-bit is the way to go. If not, you're fine with 12-bit. Personally, I have my D800e and D800 set to 14-bit and never change it.
A jpeg resulting from the editing of a 14-bit raw file that's had its shadows lifted substantially will look better and have less posterization than a jpeg resulting from a 12-bit raw file that's been edited in the same manner. TIFFs would be the same way. The key is the raw editing step - not the final format.
#6. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
>I read in many places people saying that recording RAW
>pictures in 12 bit or in 14 bit do not make any difference in
>the quality of the picture. However, I am not certain to
>understand if this is true by just viewing RAW pictures with,
>for example, Lightroom or Photshop.
You can't see a 12-bit or a 14-bit color gamut on a monitor that only displays something between a 6-bit or 10-bit color gamut. Forget about that aspect of it unless you're going to invest in a very, very expensive graphics monitor.
>My question is : when converting RAW into JPG a 12 bit RAW
>file compared with a 14 bit file, will the JPG quality be
>exactly the same also ?
The JPG quality will be slightly better - as long as you process the RAW properly. Just as Rick noted, the more color gamut you work with (14-bit) the smoother and cleaner your shadow detail is liable to be. As well, gradations such as high overhead sky fading to a lighter or darker horizon benefits from greater bit depth because there are more and finer tonal steps. When a conversion to TIF or JPG is then done, everything looks better. Just as important, the more color depth you've got in the RAW file, the more accurately the converter can work when you create your TIF or JPG.
IMO, I think the most important reason for making photos at the greatest possible bit depth is to improve the quality of your prints. The greater the gamut, the better the print. Todays multi-ink, wide format printers are capable of outputting a huge gamut. Starting the image selection and post-processing workflow with a 14-bit RAW file makes the best sense and will eventually result in the best possible print output.
Mario Georgiou (one of the genuises at Harrod's graphics design group in London, one of the internal store key ad design managers, and the Harrod's in-store brand guardian), has taught photographers, graphic artists and designers for years to forget about space concerns and to forget about trying to predict what something is going to be used for at some point in an undefinable future. Always work with original digital photos and scans created at the highest possible resolution and greatest possible bit-depth. Once you have all the data that can possibly be captured with a DSLR, video camera or scanner, you can then create anything you need - from giant billboards, large ultra high res posters or art prints for viewing in high detail at only arm's length, down to the 8"x10" framers, down to small web graphics. If you don't have all the color depth and resolution in the first place, you can't recreate it or put it in later.
Make sure your monitor isn't letting you down. Spend $75-$125 on a good monitor calibrator such as the ColorSync Spyder. It's fully automated and it takes only 10 minutes to fully calibrate your monitor in a darkened room. You'll then be looking at the best it can do for you.
#8. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
I agree with much of what has already been said.
As one who likes to make large, gallery quality prints, I use an Eizo 24 inch monitor that claims to present 98 percent of the Adobe RGB color space, and I calibrate it with the Colormunki system.
I print with an Epson P7900 printer (24 inch wide carriage, roll feed) that includes the HDR Ultra Chrome ink set (eleven colors).
I shoot everything possible in base ISO (usually 100 or 200, depending on camera), uncompressed RAW format, 14 bit color depth, Adobe color space, and all in-camera image processing functions set to off, or their neutral positions. Post processing is done in Adobe CS5 Extended. In recent memory, the only time I shot at a higher ISO was at night from a law enforcement helicopter circling a staged felony traffic stop on the ground, for some promotional publication images.
The fine shadow and other detail possible with this workflow shows up in large prints on quality papers. Yesterday I took a series of large prints to the gallery where some of my work hangs. The owner has a client who is interested in one of my images. He gasped when he saw one of them, which he had not seen before. He commented at length about the richness and sublety of the colors, the detail, resolution ... This from a man who views a lot of prints.
In my experience, the advantages of RAW, 14-bit color depth, etc., do not become apparent until one enters the large format printer world. And then, it takes a carefully designed work flow from start (pre-camera visualization of scene/subject under pre-processing ambient illumination) to finish (print hanging on wall under post-processing ambient illumination) to avoid the many traps that will affect image quality if not recognized and managed along the way.
One of the goals of my workflow is to retain as many bits of image data as possible, from start to finish. This includes capturing the image in-camera as close as possible to the finished aspect ratio, and requiring minimal cropping. Yes, my file sizes are larger, and it takes a bit longer to prepare large image print files, but it is clearly worth it in my experience.
The only time I enter the JPG world is when I post images on this site.
Hope this helps a bit.
HBB in Phoenix, Arizona
Nikonian Team Member
Photography is a journey with no conceivable destination.
#9. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
When converting a 12 bit RAW compared to a 14 bit RAW photo to JPEG; more than likely the output will look the same unless … you use a color managed system with a high-end monitor. Then you may see subtle differences.
However shooting in either 12 or 14 bit RAW and outputting to sRGB (8 bit) there are benefits as pointed out by many post before this one.
The one major benefit of shooting in 14 bit RAW vs 12 bit RAW - future image editing and viewing possibilities may make very good use of the extra recorded color gamut.
Visit my Nikonians gallery.
Visit my Nikonians gallery.
#10. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
Yes, the JPG would be virtually identical give or take a few color swaps. The main limiting factor of JPG is it's tiny 8bit capture range.
A JPG image can only capture 256 unique colors. A 14bit image (such as any advanced SLR) can capture 16,384 colors.
Notice I use the word 'can', because the actual color accuracy, rendition, and capture is also limited by the sensor and processing power contained in the camera.
Cheap cameras can make JPG's nearly identical to the spendy ones. But it's with RAW imagery where the big gun SLR's shine.
#12. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 10DigitalDarrell Charter MemberSun 07-Apr-13 08:43 PM | edited Sun 07-Apr-13 09:17 PM by DigitalDarrell
>A JPG image can only capture 256 unique colors. A 14bit image
>(such as any advanced SLR) can capture 16,384 colors.
Actually, Phil, that's a common misconception.
An 8-bit JPEG can capture over 16 million colors. However, it can only capture 256 unique colors per RGB channel. That's 256 reds (R), 256 greens (G), and 256 blues (B). When you combine all three channels (all the shades you have) you will see this: 256 x 256 x 256 = 16,777,216 colors.
A 12-bit file can contain 4096 colors per RGB channel or over 68 billion colors. When you combine the RGB channels you have: 4096 x 4096 x 4096 = 68,719,476,736 colors.
A 14-bit file can contain 16,385 colors per RGB channel or over 4 trillion shades. When you combine the RGB channels you have: 16385 x 16385 x 16385 = 4,398,851,866,625 colors.
Therefore, a JPEG file can contain a high degree of color compression while still maintaining proper luminance levels so that the image looks great. It just cannot be modified over and over and resaved without degrading from compression losses.
Personally, I would shoot in 14-bit RAW mode and take advantage of all those potential color storage bits (for subjects that need it) and later make a conversion to JPEG from my vastly superior RAW container. It is especially important to do so if, as Rick Walker mentioned, you plan on using data in the low-order bits (darks) for maximum dynamic range.
Moral of the story: Shoot and store your images as RAW if you have the time, temperament, and training to deal with postprocessing.
Darrell Young (DigitalDarrell) www.pictureandpen.com
"Better too many words than not enough understanding."
#13. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 12Vox Sciurorum Registered since 18th Oct 2007Mon 08-Apr-13 04:40 PM
>A 14-bit file can contain 16,385 colors per RGB channel
>or over 4 trillion shades. When you combine the RGB
>channels you have: 16385 x 16385 x 16385 = 4,398,851,866,625
A 14 bit value can represent 2^14 numbers, but camera sensors can not generate 2^14 truly different values to fill that space. These sensors work by counting photons. There is a fundamental limit to accuracy called "shot noise." When the sensor says 10,000 photons that should be read as 10,000±100 (random error is related to square root of value).
In bright areas the 14 bit value is about 7 bits of measurement and 7 bits of noise. Nikon's NEF compression works by discarding some of that random noise. I have seen claims that it does so badly but never backed by sound methodology.
#11. "RE: RAW 12 bit versus 14 bit" | In response to Reply # 0
I also have a "forward fear" that I may discard something that may be important some day. So basically, I shoot 14-bit RAW. That's all there is -- there ain't no more! I do edit images after a shoot, deleting obvious bad captures -- bad focus, framing, blur, etc. Then I edit the best or what I need for a project and will probably discard a lot of what's left. I do this for three main reasons: to winnow down my folders for future perusal; to save storage space (I know, it's cheap, but...) and because one way to be seen as a good photographer is to get rid of your mistakes! But I always save the RAW NEF, from which I can later produce any file format that may be required. That starts with 14 bits.
Waterloo, IL, USA
Visit my Nikonians gallery.