BeaverNewby Loc: Memphis, Tn

While looking at the file formats options for various Nikon cameras I found the following Info in there Camera Specifications:
I understand that compression methods toss away redundant information in order to keep file sizes smaller. Question: Can we expect to see a visible difference when comparing images from 12 bit versus 14 bit file formats or compressed, Lossless Compressed, and Uncompressed file formats?

File Formats info from Nikon Website

Nikon D3300
NEF (RAW): 12 bit, compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression
NEF (RAW) + JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D5500
NEF (RAW): 12 or 14 bit, compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D7000
NEF (RAW): 12 or 14 bit, lossless compressed or compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW) + JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D7200
NEF (RAW): 12 or 14 bit, lossless compressed or compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D500
NEF (RAW): 12 or 14 bit (lossless compressed, compressed or uncompressed); large, medium and small available (medium and small images are recorded at a bit depth of 12 bits using lossless compression)
TIFF (RGB)
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression; Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Full Frame
Nikon D600
NEF (RAW): 12 or 14 bit, lossless compressed or compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D750
NEF (RAW): 12 or 14 bit, lossless compressed or compressed
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D800
NEF (RAW): 12 or 14 bit, lossless compressed, compressed or uncompressed
TIFF (RGB)
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Nikon D810
NEF (RAW): 12 or 14 bit, lossless compressed, compressed or uncompressed; small size available (12-bit uncompressed only)
TIFF (RGB)
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression (Size priority); Optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats
Nikon D5
NEF (RAW): 12 or 14 bit (lossless compressed, compressed or uncompressed); large, medium, and small available (medium and small images are recorded at a bit depth of 12 bits using lossless compression)
TIFF (RGB)
JPEG: JPEG-Baseline compliant with fine (approx. 1:4), normal (approx. 1:8) or basic (approx. 1:16) compression; optimal quality compression available
NEF (RAW)+JPEG: Single photograph recorded in both NEF (RAW) and JPEG formats

Winslowe

Take a shot in each, and compare the results!

BeaverNewby Loc: Memphis, Tn

That sound simple but I have a lowly D3300 which only has the option to record a 12 bit compressed file. I am asking those that have that option and use it there views.

Apaflo Loc: Anchorage, Alaska

There is virtually no valid reason to ever save an uncompressed file. It may have use for a very high speed automatic data transfer to real time processing. I've never heard of anyone on UHH doing that.

Nikon's lossy compression changes the gamma curve in the two highest recorded fstops. Really bad for flowers and wedding dresses, but it does make a smaller file. Probably never a good idea unless you understand it well.

Hence the Lossless Compressed formats are the default best choice.

The difference between 12-bit and 14-bit is how much dynamic range can be recorded. If you shoot low contrast scenes and stay below ISO 800 then 12-bit encoding will do fine. If you shoot higher ISO's in low light, or higher contrast scenes at any ISO you want 14-bit encoding all the time.

Trying to shoot random images to detect the differences is very unlikely to be productive.

Winslowe

Understood. Apaflo has said it pretty well.

JimH123 Loc: Morgan Hill, CA

If you are just looking at an unprocessed file, you are unlikely to see any difference with the various options. The difference comes when you push the image to emphasize the dynamic range.

In astrophotography, we really push (we call it stretch) images to bring out really dim and low contrast deep sky objects. Having more bits can make a big difference.

BeaverNewby Loc: Memphis, Tn

I can see that would help capture the very subtle differences in hues and tones of flowers and wedding dresses. Can you restate what you mean by the two "highest recorded fstops?" Do you mean f22, and f32. or are you referring to f1.4 or f1.8.



Sounds, like you are in favor of cameras that have the option.
Thanks Apaflo

Apaflo Loc: Anchorage, Alaska

The two or three brightest fstops of recorded data are the ones that are gamma adjusted. It varies with different models and I don't remember exact specifics for any one model. With 12-bit data there would be 4096 steps in the highest fstop, 2048 in the second and 1024 in the third... but with Lossy Compression that is compressed to something like 750 steps (in the RAW data, keep in mind there are only 69 levels in the highest fstop in JPEG data).

The ability to distinguish high frequency detail is reduced.

Perfectionist will see a difference, but perhaps the average photographer will not. If, for example, you shoot JPEG or if all of your images are down sampled to 1200x1600 or smaller it would have absolutely no visible effect!

BeaverNewby Loc: Memphis, Tn

I always try to get the best images possible from my equipment. Meaning I shoot raw and until recently I just cropped them, adjusted them for exposure or color temperature, then I converted them to jpg's. Because I didn't know that shooting in Raw required some degree of sharpening I discovered I was losing some potential Image quality! Since my camera doesn't allow a choice in saving files in a higher bit format, I was curious how much image quality I was losing. I was hoping somebody would post a picture, say a static scene shot in 12bit versus 14bit so that I could see if I could appreciate the difference.

JimH123 Loc: Morgan Hill, CA

In a normal static scene, you are not going to see any difference in 12-bit vs 14-bit. There needs to be some heavy processing done to the image that needs to reach for deep down information that you can't see in normal viewing. If you want to see a difference, over expose an image so that the sky is as close to clipping as you can make it. And then in processing, lower the exposure to try to make it normal looking. You can only try 8-bit JPEG vs your 12-bit RAW, but you should see the sky develop bands of brightness instead on being a gradual shift in brightness. The JPEG will be much worse looking. A 14-bit version would be more tolerant than what you can see on the 12-bit.

I searched the internet to find an example of sky banding which I am showing here. If you look at the sky, you see abrupt changes in the sky. I have seen worse than this, but it will suffice for discussion purposes. The fewer the bits, the more prone this is to happen. The more bits, the less prone to happen. Also be aware that a monitor could show banding due to the inability of the monitor to display sufficient color gradients. Again, the original JPEG is not likely to show this. But if you over expose to the right (without clipping - watch the histogram) and then correct in the photo editor, you can see it created when dealing with fewer bits.

(Download)

Math78 Loc: Scottsdale, AZ

I agree with this. I've been switching to RAW 12-bit compressed when I need to maximize the frames per second that I can shoot AND write to the SD card. The 12-bit file is about half the size of the 14-bit, and that makes a big difference in continuous shooting with my D7100. I can't tell the difference in image quality as long as the exposure is reasonably close to being correct.

Check out some comparison photos here: http://photographylife.com/14-bit-vs-12-bit-raw

JimH123 Loc: Morgan Hill, CA

Good example. But the author didn't expose to the extreme right and try to recover. Only to the extreme left. It is harder to see the banding in the darks than it is in the brights such as the sky due to the gritty nature of the dark areas. I have tried the expose to the right exercise and was able to see the banding. But I didn't keep that image to use it again.

BeaverNewby Loc: Memphis, Tn

I read through all the posts and agree that the 14bit raw has the potential to capture more image information than the 12bit raw. It's the post processing skill that will determine whether you will see a difference in the final photo. I recently started using photoshop elements 14 and noted that for some of the effects it reduces the images from 16bit to 8bit. There is definitely a shift in image quality visible on my 23" monitor when this happens.

JimH123 Loc: Morgan Hill, CA

There is actually very little that does not get downsized in Photoshop Elements to 8-bits. This is intentional by Adobe as they want you to use Photoshop. But there are other photo editing products that are inexpensive that don't play that 8-bit game.

BeaverNewby Loc: Memphis, Tn

I have resisted the temptation to use the more expensive product because I really like the Photoshop Elements Program. I am still trying to master the Camera basics part of it. I am also focusing on building my kit of lenses.