When I got my D300 and had to choose between shooting in 12 or 14bit I did a series of comparative shots to see if I could see the difference. I know that there is a theoretical difference. But, photographs are about "seeing" and if you can't see the difference, why use up 25% of your storage space for nothing?

So, now with the D700, and with the theory that the FX sensor provides better color rendition per pixel, I am wondering if I can see a difference between 12 and 14 bit. By the way, I got a 16gig Extreme III card today and it gives just over 800 shots (or that is what is says) in 12bit mode, and 600 shots in 14bit mode.

I did a couple of test shots and, again, I am having trouble seeing the difference. See http://www.pbase.com/lmmiller9/d700300_comparisons.

Has anyone else done their own comparisons in any other lighting conditions, or colors, and been able to actually see a difference? If so, under what conditions? Have you decided to shoot in 12 or 14bit and why?

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

My ancient eyes aren't good enough to tell a straight 14-bit print or monitor image from a 12-bit either, but then I can't usually tell the difference between a high-quality jpeg and an uncompressed tiff or a raw file straight out of the camera either. At worst, you'll most likely only see that largely theoretical difference in big color spreads like a clear blue sky where the sampling differentiation needs to be extremely subtle.

Once I start working with the files in Photoshop, however, I appreciate all the extra headroom I can get. If I know I'm going to be enlarging an image past 13x19 or so and/or applying some creative (or just rehabilitative) filtering in PS, I want to start with the cleanest, deepest data possible.

Digital storage is cheap, cheap these days, and it's hard for me to even imagine a scenario where I'd need room for more than a few hundred images before pausing to dump or exchange the card.

Bruce Jones
Writer, rider, shooter

I had asked how to witch between 12 and 14 bit, but I found it. I will invstigate your topic now that I can

Regards,

Don Kesler

http://www.donaldkesler.com

Through the judicious use of adjustment layers, no pixels were actually harmed in the processing of my shots..

i think you should use it all time on d700.
if you adjust exposure in the nef or do postprocessing you'll get more distinct values instead of just dark muddy noise.

http://www.earthboundlight.com/phototips/nikon-d300-d3-14-bit-versus-12-bit.html

so the bigger the bit depth is, no matter what the camera's ADC encoding size is or how it works, the exposure being logarithmic with the value scale starting in highlights the more values will be available for the shadows detail. if you mix that with the exposure to the right method you'll get an even better SNR, less noise.

i just saw the news about the hp's new dreamcolor 30-bit lcd (10 bit per channel) and since it covers adobergb 100% you might actually see the difference on everyday pictures. the bad news is that is expensive, the good news is that a <$200 card like ati 4850 has 30-bit display capability so. more things to wish for...

I agree with your comments. The noticeable improvement will be most appreciated as providing a more robust file to work with in post processing, which in the end should equate to a slightly better print.

Regards,

Don Kesler

http://www.donaldkesler.com

Through the judicious use of adjustment layers, no pixels were actually harmed in the processing of my shots..

But from reading all that technical analysis, it still seems that no one actually sees the difference in a real photograph, of real subjects, through the human eye.

The downside of shooting in 14bit, in addition to 25% reduction in storage space is the slower recording speed, and a reduction in battery life, much of which is consumed in the storage/transfer process.

Which leads to another subject, which is that it seems that (haven't done any tests to determine this)the D700 is going through the standard battery much more quickly than the D300, which had improved over the D200. I wonder whether anyone has tested the battery life compared to other cameras.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

Hi Larry,

I agree that my battery is draining faster than normal, but I'm also spending a lot of time setting up menus and gawking at histograms, experimenting with the FN button, learning to use the commander mode with my SB's, and showing people how gorgeous the LCD screen is In another couple of weeks I expect this to settle down quite a bit and battery life should be lots better. If not I'll just buy more batteries and get on with life. It is what it is.

As far as 14bit vs, 12, I'm not talking about spitting out two files of the same subject in ACR or NX and just looking at them. I'm saying that the file is more robust for serious editing. This debate is slightly similar to the arguements that raged for years about 8 bit vs 16 bit, or RAW vs JPEG. Finally enough people have gotten the message to where it seems there are only a few cinders left burning.

Regards,

Don Kesler

http://www.donaldkesler.com

Through the judicious use of adjustment layers, no pixels were actually harmed in the processing of my shots..

Well...no one has yet said they can SEE any difference. I do not think this debate is similar to the RAW vs JPEG debate. There there are clear differences in performance that are easily seen or experienced. When editing a RAW file or a JPEG file it is not too hard to see the advantage of RAW, particularly with the NX software that allows you to alter in-camera settings.

Saying that 14bit is "more robust for serious editing" I again find to be a theoretical statement. A photograph is something you see, with your eyes, not with computer code. Still no one seems to be able to say "yes, LOOK at this photo here, you can SEE that the 14bit file is more...(something)."

So far, the argument goes something like, this musical instrument, or speaker, can make sounds that are 20% higher pitched than some other speaker or instrument, but of course you can't hear that pitch, only a dog can or some electronic instrument." OK, fine. Nice technical achievement, but so what? Should I pay more for this speaker that produces a sound I can't hear? I think not. Music is about hearing, like photography is about seeing.

You say,"enough people have gotten the message...only a few cinders left burning." Is this true. What message have they gotten? I guess I am a cinder still flaming. Questioning really.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

You seem to be trying awfully hard to convince those of us who do appreciate the extra headroom of the additional FX bits that we've all been brainwashed. Why do you find that necessary?

Could it just be that maybe some of us can, occasionally--though certainly not always--make real use of 14 bits because we can see a difference in our work process, but that your work process simply doesn't stress the image file enough to require anything more than 12 bits?

Personally, I think the jpeg analogy is not all that farfetched. For many people, including a lot of seasoned full-time pros, jpeg is an adequate file format for all of their photographic needs, since in MOST circumstances the human eye won't detect the compression artifacts that DO, in fact, EXIST. For others, particularly those who do heavy post-processing or who want to make medium-format-sized enlargements from a 35mm-based digital camera, RAW is a necessity.

Similarly, if you don't need 14 bits for your style of photography, good for you. If you CAN use it, it's nice to have it available. It's also nice that Nikon provided the option of selecting it or ignoring it, which I assume you would prefer to do.

Strokes for folks, courses for horses...

Bruce Jones
Writer, rider, shooter

Bruce,

I am not trying to convince anyone of anything. I am asking the question, which no one has answered, can you SEE a difference in 14 and 12 bit photos with the D700?

In your first sentence you refer to the "extra headroom of the additional FX bits..." This is not about FX or DX. I have done side by side shots with the D300 and D700 and I can see the advantage of FX, particularly in low light conditions. The issue is the 12 or 14 bit issue. And, again, I have no trouble seeing the advantage of RAW over JPEG. I don't think my question should be confused with other questions for which the answers are clear, at least to me.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

"I am asking the question, which no one has answered, can you SEE a difference in 14 and 12 bit photos with the D700?"

Sorry, I thought several of us HAD already answered that. Yes, "in some circumstances--though certainly not all," I can, and several of us gave you examples of circumstances in which the extra room might manifest itself. Although I'm working with a D3 rather than a D700, since the sensor and processor are the same I assumed my experience would be of some use in the discussion. Since the other respondents to your question ARE using D700's, it would seem that my assumption was warranted.

MY point was that if YOU can't see the difference in YOUR work then the answer to your question--for you--is that the difference between 12-bit and 14-bit is irrelevant. For those of us who do use and appreciate the extra bits, the difference is meaningful. Once that's been established--and it appeared to me that it had been--then continuing to press the same question after getting an answer you don't approve amounts to trying to convince those with a different answer to agree with a conclusion we haven't independently formed.

BTW, if the camera captured even more bits, I'd use those too. At some point, the extra data are strictly there for processing headroom, not because you can see them straight out of the camera in a raw file. After multiple levels of processing even you might be able to see the difference in a 24x36 print, which size I'm occasionally asked to produce. I recently shot a family portrait and the client specifically requested a 20x30 heirloom print. Since the D3 was still new I took the D2X along as a backup and shot the pic with both cameras. Outdoor portrait with plenty of light, so old-DX noise wasn't a problem. In the enlargement, you can easily detect a slight improvement in the smoothness of the background sky and in the level of detail rendered in the background vegetation. Some of that, no doubt, was due to generational improvements in the signal processing and the fatter pixels in the FX format, but some of it appears to be the result of extra sampling data. Even side by side, I question whether or not the client would have seen a meaningful difference, but that wasn't YOUR question.

My backyard experiments with the D3, shot both in 12-bit and 14-bit, confirm those results. I personally don't see a difference (others with better, younger eyes might) until the file is significantly enlarged, and even then only in some specific areas of the image. You'll never see that in any examples posted electronically because the file sizes necessary to reveal the difference are extreme. Is 14-bit a universal necessity for me? No. Is it occasionally useful? Absolutely.

Sorry about the confusion with FX vs. DX. That was my shorthand way of pointing to 14-bit data, not realizing that the D300 apparently also has 14-bit capture available. I have a D200 and a D2X that serve my DX needs just fine, so I haven't paid much attention to the 300. I understand that FX/DX isn't synonymous with 14-bit vs. 12-bit; I just didn't know there was a DX camera with 14-bit capabilities.

Enjoy every "bit" of that D700, regardless of how many you choose to employ.


Bruce Jones
Writer, rider, shooter

Thank you. That is a helpful explanation. I am really not trying to be argumentative and certainly not trying to convince anyone of anything.

My thought, which may not be worthwhile, was that perhaps with the larger pixel size of the FX sensor, someone else might see the difference between the 12 and 14bit photos, or perhaps that difference is only seen in certain lighting conditions, etc. I was trying to narrow down under what conditions it is worth the "cost" of the additional storage space to shoot in 14bit.

Your answer certainly answers the question from your experience. thank you.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

was that perhaps with the larger pixel size of the FX sensor, someone else might see the difference between the 12 and 14bit photos

I am probably sticking my 2 cents where it may not be wanted but let me make a few comments since this is how I see it, so correct me if I am wrong

1) 12 vs 14 bits has nothing to do with more dynamic range. That is theory, and this chart, from Nikonian bclaff, based on actual data from bodies like a D3 show this to be the case. So differences are unlikely to be noticable in most case


2) the extra 2 bits should not have any "direct" advantage on prints. 12 bits gives us 4096 shades of grey, 8 bits gives us 256, but a printer offers less than that. In one printing seminar I went to, sponsored by Epson, it is closer to 100 bits so proper PP is critical to preserve print detail

3) Benefits of 14 bits come in the form of more information that can be stretched without posterization or excessive noise in areas where extra detail needs to be pulled out --- like shadow detail 'recovery' You know how you can take a 8-bit jpeg file of a low contrast capture and bring the levels sliders together and see posterization? Well, if it is a 12-bit RAW you can do more of this before the image posterizes. Extend this argument to 14-bits and you have more information to separate, when needed and if needed, minimizing potential problems/artifacts

4) Indirectly, there is an advantage for printing certain captures, since this additional information can be separated to provide the added detail/local contrast for the print. The entire concept of levels and curves is, as Bob Johnson put it, 'stealing contrast' from one area to another to get detail and tonal balance as desired.

Edited to add I think where a 14-bit capture may be a real advantage over 12-bits is if you plan on tonal-mapping a single image to extend the detail in highlight, midtone, and shadow areas. The additional info should yield a slightly cleaner, more detailed, result. This isn't a 14-bit D700 capture, but it is is a single file tone mapped to get the most detail out. http://www.pbase.com/alvalentino/image/98464847 This would be one of the few types of images I personally would consider 14-bit captures.

Albert J Valentino
Nikonian Team Member
Vantage Point Images
Mastery of Composition is the Key to Great Photography

The prevailing reasoning seems to be whether the difference is visible in print. However, for those of us shooting for other purposes, such as stock, the end result might not be print.

I have set my D700 for 14-bit capture. I never shoot jpegs, a legacy from also shooting with a Kodak 14nx dslr and a Mamiya ZD, both of which have rubbish jpegs. I assume - I think correctly - that there will be occasions where 14-bit raw will offer advantages, otherwise Nikon would not have provided that facility. Life is, however, too short and storage too cheap to spend hours checking one bit depth depth against the other.

QDFB

I think that is a good summary point. My take aways from this discussion are:

a) In virtually all cases, a 12 and 14 bit shot of the same scene, displayed on your computer screen will reveal no noticeable difference, even when displayed at 300%.

b) The above does not mean that there are no differences in the photo file. These differences may appear in prints when processed in certain ways and printed in certain ways (I am not sure what all of those are).

c) There is no point shooting in 14bit if your primary purpose is to use the photos as a computer file or on the internet.

b) If there is a likelihood that you will do significant post processing and you plan to print in large format, it is best to shoot in 14bit.

d) The cost of storage, as you point out, is now so cheap that when in doubt, shoot in 14bit and don't worry about it. I have one 16 gig and two 8gig cards. Together they should hold at least 1200 14bit photos. The likelihood of shooting more than that before downloading, at least for me, is non-existent.

e) No one reported that they had found any difference between the D300 and D700/D3 in how 14bit files may exceed 12bit files. I suppose it could still be that shooting in 14bit may be more advantageous with the FX chip than the DX chip, something that I would suspect and was the main reason I asked the question to begin with. It remains a question.

Larry Miller, Annapolis, MD
D700/D800
http://www.pbase.com/lmmiller9
http://lmmillerphotography.smugmug.com/

Hmmm.

a) essentially true for images not PP since DR would be identical for 14 vs 12 bits.

c) NO, not necessarily. 14 bits has more information that can be pulled out and separated with a cleaner, non-posterized result. So PP version 'can' have more detail and that detail should be visible regardless of the viewing medium. Think about capturing the sun in an image and with 14 bits having smoother tonal ramps due to more information between the tones for smoother separation

e) I might argue that all things being equal, like same ISO and exposure settings for a Dx vs Fx capture of a contrasty scene, that there would be more need to use a 14 bit setting on Dx because the Fx sensor will inherently have more DR and thus the Dx sensor would need more help to pull out/separate detail without noise or notcable posterization/tonal ramps

Albert J Valentino
Nikonian Team Member
Vantage Point Images
Mastery of Composition is the Key to Great Photography

Very few computer monitors reach 11 bits so you won't see a difference by direct viewing.

Connect by HDMI to a high grade plasma display if you need to view 12 bits of brightness.

The brightness difference in the last 4 values is measureable only by equipment you can't afford.

An interesting article by Allan Weitz on 14 versus 12 bit viewing.

http://www.bhphotovideo.com/c/find/newsLetter/Bang-for-Your-Bits.jsp

Hi !

Well, I think that the difference between 12 and 14 bits does not appear immediately on the screen or even on a wide print. It seems to be useful specially when treating highlights/shadows in NEFs as a 14 bits file contains more information than a 12 one.

EYES WIDE SHUTter...
Christian
Belgium (EU)