12 vs 14 Bit color
Feb 14, 2018 14:15:23 #
I have a Nikon D850 and have read lots of material/opinions on the subject. I have not been able to find any "evidence based" material clearly demonstrating the superiority of 14 bit over 12 bit. Does such information exist? Where can I find it?
Thank you.
Thank you.
Feb 14, 2018 14:27:38 #
I would suggest you do a google search for
why is 14 bit color better than 12 bit color
There are a number of pages that will provide details. I'm not trying to be rude and tell you to look it up. It's just that there are so many links provided, I'd be doing a lot of copying and pasting to provide them here.
--Bob
why is 14 bit color better than 12 bit color
There are a number of pages that will provide details. I'm not trying to be rude and tell you to look it up. It's just that there are so many links provided, I'd be doing a lot of copying and pasting to provide them here.
--Bob
brjomd wrote:
I have a Nikon D850 and have read lots of material/opinions on the subject. I have not been able to find any "evidence based" material clearly demonstrating the superiority of 14 bit over 12 bit. Does such information exist? Where can I find it?
Thank you.
Thank you.
Feb 14, 2018 14:51:34 #
Feb 14, 2018 18:13:33 #
I am going to do my best to answer this question in practical workaday terms. Here is my approach to this and many other technical issues in digital technology. When someone asks for advice from a pro, the may expect a very detailed answere the drills down deeply into the technology on almost a "molecular" level and in some cases I can and will do that. In actual practice, however, many professionals , including myself, oftentimes boil things down to procedures, methodologies and routine acceptable protocols.
Having experienced many years in the film era, I was not necessarily involved with film manufacturing technology in terms of the physics and complex chemistry but would simply familiarize myself with each film's basic characteristics, grain structure, dynamic range potential, speed, chromatic sensitivity and color palette where applicable and choose the products accordingly. In digital imaging, I find that many photographer are more deeply immersed in very finite technical details where pixel counting is just the tip of the iceberg. When I made the transition into digital photography, as an old timer in the business,I decided to study the nitty-gritty as much as I could absorb it all and then create a set of compartmentalized procedures and methods so that i could replace each of my film based techniques with its equivalent or near-equivalent in the "new" technology. This is usually the kind of advice I have to offer, that is nuts and bolts- what to do! In my own situation at the time the rapid transition required quick fixes so that I could maintain production.
The basic difference between 12 and 14 bit has to do with nuances in shades of color. A 12 bit RAW image can store up to 68 billion different shades of color where a 14 bit image file stores 4 trillion. The human eye can differentiate between 2.5 and 16.8 million colors. So...12 bits would be sufficient? probably, in some cases but I like to overkill.
The rest of what I Have to offer is anecdotal and just based on my experience and what I was TOLD to do by more experienced and learned technicians. In the a bigger file there is more correction potential for underexposed files or if there there is a lack of shadow detail. Generally, the larger file will produce better color accuracy and quality especially where high degree of enlargement is required and in high definition lithography, big refection prints and other usages where resolution and acutence is more critical.
So...in my shop, I shoot everything in RAW 14 bit in that most of the images we are producing will be for display advertising, high definition print advertising, large screen video display and large format portraits.
I have seen smaller prints made in both ways and in maller formats, to the naked eye, there is hardly and difference providing the file is non-problematic as to exposure and range. The only advantage of 12 bit files is that they take up less space on you memory cards and they may be more appropriate for shooting in rapid sequence.
In most of my work, I have full control over lighting, lighting ratio, dynamic range and exposure so I am not particularly concerned about the potential radical post-processing corrections. I don't do very much work that would require extremely rapid shooting sequences. On long shoots I carry multiple cards. If I had to, I could shoot 8 bit Jpegs and still get 16.8 billion shades.*
I hope this helps. I am sure that others here will be to better elaborate on this question. Using the routine, I have had successful results even where the end use of an image is in a billboard display. I get no complaints from publisher, ad agents, publisher and lithographers. So far- so good!
Some of theses number may be debatable, however, the seem to add up in terms of actual results.
Having experienced many years in the film era, I was not necessarily involved with film manufacturing technology in terms of the physics and complex chemistry but would simply familiarize myself with each film's basic characteristics, grain structure, dynamic range potential, speed, chromatic sensitivity and color palette where applicable and choose the products accordingly. In digital imaging, I find that many photographer are more deeply immersed in very finite technical details where pixel counting is just the tip of the iceberg. When I made the transition into digital photography, as an old timer in the business,I decided to study the nitty-gritty as much as I could absorb it all and then create a set of compartmentalized procedures and methods so that i could replace each of my film based techniques with its equivalent or near-equivalent in the "new" technology. This is usually the kind of advice I have to offer, that is nuts and bolts- what to do! In my own situation at the time the rapid transition required quick fixes so that I could maintain production.
The basic difference between 12 and 14 bit has to do with nuances in shades of color. A 12 bit RAW image can store up to 68 billion different shades of color where a 14 bit image file stores 4 trillion. The human eye can differentiate between 2.5 and 16.8 million colors. So...12 bits would be sufficient? probably, in some cases but I like to overkill.
The rest of what I Have to offer is anecdotal and just based on my experience and what I was TOLD to do by more experienced and learned technicians. In the a bigger file there is more correction potential for underexposed files or if there there is a lack of shadow detail. Generally, the larger file will produce better color accuracy and quality especially where high degree of enlargement is required and in high definition lithography, big refection prints and other usages where resolution and acutence is more critical.
So...in my shop, I shoot everything in RAW 14 bit in that most of the images we are producing will be for display advertising, high definition print advertising, large screen video display and large format portraits.
I have seen smaller prints made in both ways and in maller formats, to the naked eye, there is hardly and difference providing the file is non-problematic as to exposure and range. The only advantage of 12 bit files is that they take up less space on you memory cards and they may be more appropriate for shooting in rapid sequence.
In most of my work, I have full control over lighting, lighting ratio, dynamic range and exposure so I am not particularly concerned about the potential radical post-processing corrections. I don't do very much work that would require extremely rapid shooting sequences. On long shoots I carry multiple cards. If I had to, I could shoot 8 bit Jpegs and still get 16.8 billion shades.*
I hope this helps. I am sure that others here will be to better elaborate on this question. Using the routine, I have had successful results even where the end use of an image is in a billboard display. I get no complaints from publisher, ad agents, publisher and lithographers. So far- so good!
Some of theses number may be debatable, however, the seem to add up in terms of actual results.
Feb 14, 2018 20:33:02 #
brjomd wrote:
Does such information exist? Where can I find it?
Found this with a quick search, from David Busch's D850 guide. You save space with a smaller image, but loose some details in the highlights.
Feb 15, 2018 07:49:46 #
brjomd wrote:
I have a Nikon D850 and have read lots of material/opinions on the subject. I have not been able to find any "evidence based" material clearly demonstrating the superiority of 14 bit over 12 bit. Does such information exist? Where can I find it?
Thank you.
Thank you.
Putting aside the theoretical advantage of shooting 14 bit over 12 bit, there does not seem to be much of a practical advantage for the majority of images. I suspect that extremely wide gamut images, or those with extremely wide dynamic range may benefit slightly from the higher accuracy, but I haven't really seen it in my images. That being said, I shoot 14 bit lossless compressed.
You might find this interesting.
https://photographylife.com/14-bit-vs-12-bit-raw
Feb 15, 2018 10:37:00 #
It is true that many photographers never view their images in large format media such as large prints screen displays. In a 5X7 or 8X10 print or in a smaller monitor image, it would be difficult if not impossible to discern the difference between a 14 and 16 bit file.
In my own case, I create the 16 bit RAW files routinely as a precautionary measure. A client may need a product shot for a simple 8 1/2 X 11 sell sheet and then decides to use same image for transit advertising and requires a print for display on the side of bus. I might decide to use any of my images for a display or portfolio print or transparency so I always want the potential to be there. I don't always know exactly what the final print or publication media is going to be- it could end up being reproduced with a fairly coarse screen on newsprint or in high resolution lithography on a clay coated or "glossy" stock.
Perhaps it is overkill but I also want to maxamize the nuances in color balance that I could work with, especially in portraiture. I made all of my color prints in house for many years back in the analog days and like to make very finite adjustments to color balance and skin tones. I am not pretending to be able to "SEE billions of shades of color" but I am critical and want to be able to retain quality at any degree of enlargement and in any number of reproduction processes. In the event of a bad exposure or a file with a dynamic range issue, I want to maxamize the potential for post processing corrections.
I seldom shoot at 7 FPS or worry about conserving space on memory cards, so I don't see any reason to create smaller files.
So many photographers are concerned about using the most acute lenses- they worry about diffraction, lens aberrations and quality loss due to filters and there is so much conversation on sensor sizes and pixel counts and everyone wants a rock solid tripod and lens stabilization so I find it is kinda anticlimactic to "risk" any quality loss in the reproduction process.
Murphy's law prevails- Sometimes the best shot I made all month is something I knocked off on my smartphone camera on the way home from work- looks OK on the phone...
In my own case, I create the 16 bit RAW files routinely as a precautionary measure. A client may need a product shot for a simple 8 1/2 X 11 sell sheet and then decides to use same image for transit advertising and requires a print for display on the side of bus. I might decide to use any of my images for a display or portfolio print or transparency so I always want the potential to be there. I don't always know exactly what the final print or publication media is going to be- it could end up being reproduced with a fairly coarse screen on newsprint or in high resolution lithography on a clay coated or "glossy" stock.
Perhaps it is overkill but I also want to maxamize the nuances in color balance that I could work with, especially in portraiture. I made all of my color prints in house for many years back in the analog days and like to make very finite adjustments to color balance and skin tones. I am not pretending to be able to "SEE billions of shades of color" but I am critical and want to be able to retain quality at any degree of enlargement and in any number of reproduction processes. In the event of a bad exposure or a file with a dynamic range issue, I want to maxamize the potential for post processing corrections.
I seldom shoot at 7 FPS or worry about conserving space on memory cards, so I don't see any reason to create smaller files.
So many photographers are concerned about using the most acute lenses- they worry about diffraction, lens aberrations and quality loss due to filters and there is so much conversation on sensor sizes and pixel counts and everyone wants a rock solid tripod and lens stabilization so I find it is kinda anticlimactic to "risk" any quality loss in the reproduction process.
Murphy's law prevails- Sometimes the best shot I made all month is something I knocked off on my smartphone camera on the way home from work- looks OK on the phone...
Feb 15, 2018 11:02:06 #
amfoto1
Loc: San Jose, Calif. USA
There's so little difference in image quality, one might argue in favor of 12 bit.
In terms of space savings, compressed 12 bit files are approx. 2/3 the size of uncompressed 14 bit... a "space savings" thanks to 33% smaller size files. If you compare uncompressed 12 bit with uncompressed 14 bit, the difference is less: about a 15% space saving. But, with the relatively low cost of memory and huge sizes of drive storage space that are available at low cost now, there's little reason to there's little reason to NOT use 14 bit.
Nikon may be the only manufacturer still offering choice of 12 bit or 14 bit. I think some Nikon models, especially older ones, can shoot at a little faster continuous frame rate and/or buffer a few more images in 12 bit mode. At least in more recent models, there's not very much difference in camera performance. For example, Nikon states the D500 is able to shoot 14 bit, uncompressed NEF files at "up to" it's max 10 fps and buffer up to 200 images. This is probably due to a combination of sensor and processor performance, as well as a large internal memory capacity and the speed of data transfer to the new types of memory cards (XQD in this case). Nikon's slight hedging in their statement is probably because the camera may still need to slow for autofocus and/or metering to work properly.
When working with either 12 bit or 14 bit RAW files in post processing programs, it's interpolated as 16 bit by the software. If you shoot JPEGs only, it might matter even less since those are reduced to 8 bit when they're processed in the camera.
FYI, Canon's last 12 bit cameras in each of their series were: 30D (APS-C, 2006) and original 5D (full frame, 2005), 1D Mark II (APS-H, 2005), Rebel XT/350D (APS-C, 2005) and 1Ds Mark II (full frame, 2004). All subsequent Canon DSLRs have been 14 bit.
And, BTW, many medium format digital camera capture RAW images in full 16 bit.
In terms of space savings, compressed 12 bit files are approx. 2/3 the size of uncompressed 14 bit... a "space savings" thanks to 33% smaller size files. If you compare uncompressed 12 bit with uncompressed 14 bit, the difference is less: about a 15% space saving. But, with the relatively low cost of memory and huge sizes of drive storage space that are available at low cost now, there's little reason to there's little reason to NOT use 14 bit.
Nikon may be the only manufacturer still offering choice of 12 bit or 14 bit. I think some Nikon models, especially older ones, can shoot at a little faster continuous frame rate and/or buffer a few more images in 12 bit mode. At least in more recent models, there's not very much difference in camera performance. For example, Nikon states the D500 is able to shoot 14 bit, uncompressed NEF files at "up to" it's max 10 fps and buffer up to 200 images. This is probably due to a combination of sensor and processor performance, as well as a large internal memory capacity and the speed of data transfer to the new types of memory cards (XQD in this case). Nikon's slight hedging in their statement is probably because the camera may still need to slow for autofocus and/or metering to work properly.
When working with either 12 bit or 14 bit RAW files in post processing programs, it's interpolated as 16 bit by the software. If you shoot JPEGs only, it might matter even less since those are reduced to 8 bit when they're processed in the camera.
FYI, Canon's last 12 bit cameras in each of their series were: 30D (APS-C, 2006) and original 5D (full frame, 2005), 1D Mark II (APS-H, 2005), Rebel XT/350D (APS-C, 2005) and 1Ds Mark II (full frame, 2004). All subsequent Canon DSLRs have been 14 bit.
And, BTW, many medium format digital camera capture RAW images in full 16 bit.
Feb 16, 2018 10:34:29 #
Thank all of you for taking the time to answer my question. It still appears that there is no ABSOLUTE, evidence based answer.
As it is apparent, by my choice of cameras, I want the best image I can possibly extract from my D850. File size is not an issue since I'm using a 45.7MP sensor. If there is only a theoretical advantage of 14 bit, I'm going to continue to use it.
Happy shooting hogs.
As it is apparent, by my choice of cameras, I want the best image I can possibly extract from my D850. File size is not an issue since I'm using a 45.7MP sensor. If there is only a theoretical advantage of 14 bit, I'm going to continue to use it.
Happy shooting hogs.
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.