Home · Register · Search · View Winners · Software · Hosting · Software · Join Upload & Sell

Moderated by: Fred Miranda
Username   Password

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Nikon Forum | Join Upload & Sell

1       2       3      
4
       5       end
  

Archive 2012 · D800: 12-bit RAW vs. 14-bit RAW
  
 
locomacdaddy
Offline
• •
Upload & Sell: Off
p.4 #1 · p.4 #1 · D800: 12-bit RAW vs. 14-bit RAW


Thanks for the examples and explanation Brian. Once again, Scott loves to pontificate his 'opinion' in post after post- instead of opening his mind to the possibility that he may NOT know everything about photography. Brian post tons of examples and provides input while others simply argue their point ad nauseam.

If we cannot see the difference in print or screen, perhaps there is basis to look at shooting 12 bit for some occasions/situations. Faster downloads/less storage etc., all benefits for the wedding/event/casual shooter. Then again if one spends all there time reading what others say instead of testing and validating for oneself...you can choose to that route too.

Edited on Apr 11, 2012 at 10:00 PM · View previous versions



Apr 11, 2012 at 10:00 PM
mshi
Offline
• • • •
Upload & Sell: Off
p.4 #2 · p.4 #2 · D800: 12-bit RAW vs. 14-bit RAW


taob wrote:
1-bit image? You mean a monochrome bitmap that only uses black or white pixels? Aside from people who want the look of the original Macintosh, who would be interested in that?



Try to convert your image to Bitmap mode in Photoshop. It offers a particular artistic look. Photography technicians are normally keen on technical aspects while artists go after looks. That's why Leonardo Da Vinci shared his observations of three classes of folks, which you even pointed out another class he possibly missed to point out about 500 years ago.

Here is a quick PS Job in less than 10 seconds using Bitmap color mode.








Apr 11, 2012 at 10:00 PM
afm901
Offline
• •
Upload & Sell: Off
p.4 #3 · p.4 #3 · D800: 12-bit RAW vs. 14-bit RAW


Brian,

Not all of us display our work on computer screens only and limit ourselves to JPEG. If that is all I did and knew that is the only place I would display my photographs, I may be inclined to use 12-bit with lossy compression.

The fact that I don't know how I may choose to display my photographs causes me to want to capture as much information as possible. I don't know how much detail I might want to pull out of the shadows. I don't know if or how I might print my photographs.

Again, just because you can't easily see the difference does not mean it is not there. It may be very difficult to show on a computer screen the subtle changes in tonality of the sky in a photograph. When it is printed on a high end professional printer using a RAW file as the source file, those subtle changes, though not obvious, will cause the printed photograph to look better than it otherwise would.

Scott



Apr 11, 2012 at 10:24 PM
crewshin
Offline
• •
Upload & Sell: Off
p.4 #4 · p.4 #4 · D800: 12-bit RAW vs. 14-bit RAW


taob wrote:
In theory, yes. But here in the real world, the practical difference is often zero. Your monitor cannot show 65536 discrete levels (maybe 1024 at best). You cannot print 65536 discrete levels. In fact, your eyes and brain cannot distinguish anywhere close to 65536 discrete levels at the same time.


taob: You are completely mis-understanding the use of higher bit depths.

It's not that you get better looking images just from switching from 12 to 14 or 16 bit and calling it a day. If that was the case... then your posts would make perfect sense.

But that's not the case. The real advantage of shooting higher bit depths is the flexibility of tweaking the image in post. Just as most monitors are only 8bit... this means that you can only see 256 colors at the same time. That's not to say that those are the only colors in the image... push, pull, slide, adjust to your hearts content and make it look bitchin'. Then save out to an 8bit format for web use or whatever you want. This essentially bakes the image into that look.

Remember...

8bit = 256
12bit = 4096
14bit = 16384
16bit = 65536

That's a HUGE difference when it comes to pushing values in post.

I'm not entirely sure what point you were trying to make by showing the same image on the web with lower bit depths. It doesn't really prove a thing.

To make this extremely simple for the OP:

If you are the type of photographer that does very little post work in something like Lightroom... then you are probably good with whatever you choose and can probably save a few MB from using 8bit jpg or 12bit raw.

If you are the type of photographer that does push things around in post... then you will gain HUGE benefits from using 14bit lossless... yes... even over 12bit lossless. That's a 12288 color difference between 12 and 14bit.



Apr 11, 2012 at 10:27 PM
taob
Offline
• • •
Upload & Sell: Off
p.4 #5 · p.4 #5 · D800: 12-bit RAW vs. 14-bit RAW


mshi wrote:
Here is a quick PS Job in less than 10 seconds using Bitmap color mode.


So-o-o-o-o... what does this have to do with NEF compression modes?



Apr 11, 2012 at 11:07 PM
taob
Offline
• • •
Upload & Sell: Off
p.4 #6 · p.4 #6 · D800: 12-bit RAW vs. 14-bit RAW


afm901 wrote:
The fact that I don't know how I may choose to display my photographs causes me to want to capture as much information as possible. I don't know how much detail I might want to pull out of the shadows. I don't know if or how I might print my photographs.


I'm getting the sense that ignorance is a big part of your life.

Again, just because you can't easily see the difference does not mean it is not there.

It's not just me. You cannot see the difference either. If you can, you can easily prove it to everyone else here. Take that 3x3 image challenge. I expect you to give a 100% correct answer on your first try.

When it is printed on a high end professional printer using a RAW file as the source file, those subtle changes, though not obvious, will cause the printed photograph to look better than it otherwise would.

I have tried this too. Again, for a properly-exposed image (your stipulation, not mine), it makes no difference. Even for modest adjustments, there is no visible difference. When you start getting into extreme adjustments, then yes, you may see smoother transitions or less obvious banding. That is the point of my comparisons: to see at what point the difference becomes visible. And even if the difference can be seen by the human eye, it may not necessarily be objectionable (e.g., look at the dithering examples I posted, which is essentially how all inkjet printers simulate multitudes of colours from only a handful of different inks).

As I said earlier, you are free to do what you want. I don't really care. I'm just putting up examples to guide other people towards making an informed decision about their workflow. How you interpret the facts is entirely up to you. But at least start with the facts.



Apr 11, 2012 at 11:32 PM
taob
Offline
• • •
Upload & Sell: Off
p.4 #7 · p.4 #7 · D800: 12-bit RAW vs. 14-bit RAW


crewshin wrote:
taob: You are completely mis-understanding the use of higher bit depths. [...] The real advantage of shooting higher bit depths is the flexibility of tweaking the image in post.


Ummmmm... did you completely miss my other posts showing highlight recovery of an image overexposed by 2 stops? Or cranking up both the contrast and saturation to +100 for more vibrant colours? Or rescuing an image that was underexposed by 6 stops? If that's not "tweaking the image in post", I don't know what is. Pretty much all I've been talking about is the effect of 12/14-bit and lossy/lossless when pushing and pulling an image post-capture. You may want to go back and re-read the rest of the thread.

I'm not entirely sure what point you were trying to make by showing the same image on the web with lower bit depths. It doesn't really prove a thing.

It demonstrates that if a 5- or 6-bit image still looks quite good, even viewed at 100% magnification, then capturing that image at 12-bit, 14-bit or 16-bit won't make a lick of difference. Because of dithering and noise (present in all captured images), we can reduce the number of discrete levels by quite a bit before the image quality becomes unacceptably poor.

If you are the type of photographer that does push things around in post... then you will gain HUGE benefits from using 14bit lossless... yes... even over 12bit lossless. That's a 12288 color difference between 12 and 14bit.

You'll have to demonstrate these "HUGE benefits". Go capture a scene at 12 bits and then again at 14 bits, and show us what you do to it to show this huge benefit. Then we can all decide for ourselves if those post-processing actions are relevant to the way we shoot.



Apr 11, 2012 at 11:42 PM
crewshin
Offline
• •
Upload & Sell: Off
p.4 #8 · p.4 #8 · D800: 12-bit RAW vs. 14-bit RAW


crewshin wrote:


taob wrote:
It demonstrates that if a 5- or 6-bit image still looks quite good, even viewed at 100% magnification, then capturing that image at 12-bit, 14-bit or 16-bit won't make a lick of difference. Because of dithering and noise (present in all captured images), we can reduce the number of discrete levels by quite a bit before the image quality becomes unacceptably poor.

Haha, but that demonstrates absolutely nothing in regards to the topic at hand. Tweak the photo after each reduction and then see how far those bit depths get you.

Prove "huge" with a photo? I don't need to. Look at the math. 12288 MORE levels of precision. 12bit has only 4096 colors total. Argue your point to death... but that's massive no matter how you look at it.

I'm not really going to waste time responding to anything else. No offense meant. I just think the internet is an interesting medium.

Honestly this is what it comes down to, period:

If you are the type of photographer that does very little post work in something like Lightroom... then you are probably good with the lower quality methods (lossy/jpg) and can probably save a few MB by using 8bit jpg or 12bit raw.

If you are the type of photographer that does push things around in post... then you will gain HUGE benefits from using 14bit lossless... yes... even over 12bit lossless. That's a 12288 precision difference between 12 and 14bit.



Take care

Gene Crucean
www.genecrucean.com



Apr 11, 2012 at 11:58 PM
wfektar
Offline

Upload & Sell: Off
p.4 #9 · p.4 #9 · D800: 12-bit RAW vs. 14-bit RAW


taob: Great demonstration. Couple points to add. First, Nikon (or Sony or whoever) knocked one out of the park with the D800 sensor. Second, this is key:

The quality from a good 12-bit file far exceeds that of a poor 14-bit file that needs rescuing in post.



Apr 12, 2012 at 12:13 AM
afm901
Offline
• •
Upload & Sell: Off
p.4 #10 · p.4 #10 · D800: 12-bit RAW vs. 14-bit RAW


Gene,

Love your logic. I just looked at your web site and find that I enjoy your work too!

Scott



Apr 12, 2012 at 12:13 AM
 

Search in Used Dept. 



mshi
Offline
• • • •
Upload & Sell: Off
p.4 #11 · p.4 #11 · D800: 12-bit RAW vs. 14-bit RAW


taob wrote:
So-o-o-o-o... what does this have to do with NEF compression modes?



It was in response to the question you raised. "1-bit image? You mean a monochrome bitmap that only uses black or white pixels? Aside from people who want the look of the original Macintosh, who would be interested in that?"



Apr 12, 2012 at 12:50 AM
slrl0ver
Offline
• •
Upload & Sell: On
p.4 #12 · p.4 #12 · D800: 12-bit RAW vs. 14-bit RAW


taob: Thanks for more examples.

mshi: Taob's tossing of data was done on a per-channel basis (R,G,B), I don't think the point was to simulate a monochrome (1-bit pixel image) but show that even with a significant loss of the lower bits (again, per channel), the majority of the content still comes through with very few discernible differences. His "6-bits" example is equal to 18-bits total data, or 2^18 total number of colors.

crewshin:

I think you need to check your math. The sensor captures data in 14-bits *per* channel of which there are three: R, G, B. The total number of colors (theoretical after Bayer interpolation) is 3x14-bit, a very large number. 2^36 (3x12-bit depth) is also a very large number >>> the 12288 you mentioned.

Secondly, I think you're confusing the image editing program's internal depth vs. the capture data. Most modern programs probably work with either 16-bits per channel, or now floating-point data. So even if the capture itself is "only" 8-bit, the manipulation process shouldn't truncate anything -- if it did, then the process wouldn't be sufficient for higher bit-depth capture.

This is a good resource if you're curious: http://www.normankoren.com/digital_tonality.html

- slrl0ver





Apr 12, 2012 at 06:15 AM
afm901
Offline
• •
Upload & Sell: Off
p.4 #13 · p.4 #13 · D800: 12-bit RAW vs. 14-bit RAW


A 12-bit file has 4096 colors per channel or 12288 colors for a three channel RGB file
A 14-bit file has 16384 colors per channel or 49152 colors for a three channel RGB file

A 14-bit file can have 36864 more colors than a 12-bit image. That's a difference of exactly 3 times the total number of colors that a 12-bit file has from the start.

Look here:

https://nikoneurope-en.custhelp.com/app/answers/detail/a_id/23851/~/what-are-the-advantages-of-14-bit-recording-over-12-bit%3F

I know, you guys will say it is just marketing hype.

Look here again to see examples that can be seen:

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

http://www.earthboundlight.com/phototips/14-bit-raw-12-bit-part-two.html

The bottom line is that if you might want or need to pull details out of shadows, shoot 14-bit. If you are going to manipulate the RAW data in post processing, even if all your final output is JPEG, it's better to shoot in 14-bit.

Scott



Apr 12, 2012 at 07:31 AM
slrl0ver
Offline
• •
Upload & Sell: On
p.4 #14 · p.4 #14 · D800: 12-bit RAW vs. 14-bit RAW


afm901,

Sorry, but your math is wrong.

The sensor data is 12 or 14-bits per channel: Red Channel, Green Channel, Blue channel. Since the sensor is a Bayer pattern sensor, each RAW sensel can only measure one of those colors. De-mosaicing computes the equivalent missing colors to form a pixel.

Color (for this process) is the combination of Red, Green and Blue values, so it cannot just be 2^12 or 2^14 colors. If you look at your desktop settings you'll notice it says 24-bit or 32-bit. That's 8-bit for Red, 8-bits for Green, 8-bit for Blue, and 8-bits for 'alpha' (transparency). 3x8=24 and 2^24 represents total number of different colors your monitor can theoretically display. That's why you see it written as '24bpp' or 24-bits per pixel.

- slrl0ver



Apr 12, 2012 at 09:24 AM
ausemmao
Offline
• •
Upload & Sell: Off
p.4 #15 · p.4 #15 · D800: 12-bit RAW vs. 14-bit RAW


afm901 wrote:
A 12-bit file has 4096 colors per channel or 12288 colors for a three channel RGB file
A 14-bit file has 16384 colors per channel or 49152 colors for a three channel RGB file

A 14-bit file can have 36864 more colors than a 12-bit image. That's a difference of exactly 3 times the total number of colors that a 12-bit file has from the start.

Look here:

https://nikoneurope-en.custhelp.com/app/answers/detail/a_id/23851/~/what-are-the-advantages-of-14-bit-recording-over-12-bit%3F

I know, you guys will say it is just marketing hype.

Look here again to see examples that can be seen:

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

http://www.earthboundlight.com/phototips/14-bit-raw-12-bit-part-two.html

The bottom line is that if you might want or need to pull details out of
...Show more

A 12 bit rgb file has somewhere around 68 billion possible colours, not 12000
A 14 bit rgb file has somewhere around 4.4 trillion colours.

If the internal workings of image software used fixed point or integer operations, then those gradations would be useful across the range. Given they use 16 bit floating point, it's moot.

"Use 14 bit for better shadow detail"...that's one of the things taob showed in the images not half a page above your post. It's as if you see the words and pictures but don't read them...



Apr 12, 2012 at 10:54 AM
afm901
Offline
• •
Upload & Sell: Off
p.4 #16 · p.4 #16 · D800: 12-bit RAW vs. 14-bit RAW


ausemmao,

You right...it was late...I copied from another site.....should have given credit too....

Moral of the story? Don't trust someone else's math late at night.

Scott



Apr 12, 2012 at 03:03 PM
ausemmao
Offline
• •
Upload & Sell: Off
p.4 #17 · p.4 #17 · D800: 12-bit RAW vs. 14-bit RAW


afm901 wrote:
ausemmao,

You right...it was late...I copied from another site.....should have given credit too....

Moral of the story? Don't trust someone else's math late at night.

Scott


Late at night I barely trust my own maths, let alone someone else's



Apr 12, 2012 at 03:44 PM
mshi
Offline
• • • •
Upload & Sell: Off
p.4 #18 · p.4 #18 · D800: 12-bit RAW vs. 14-bit RAW


Normally artists don't need any understanding of math involved here, unless you're camera technicians or operators. A visit to HSL color paradigm will definitely beef up your confidence in higher bit-depth.


Apr 12, 2012 at 03:54 PM
ausemmao
Offline
• •
Upload & Sell: Off
p.4 #19 · p.4 #19 · D800: 12-bit RAW vs. 14-bit RAW


mshi wrote:
Normally artists don't need any understanding of math involved here, unless you're camera technicians or operators. A visit to HSL color paradigm will definitely beef up your confidence in higher bit-depth.

Certainly true.
But with regard to the question asked, the maths is most definitely relevant. Noone has said that higher bit depth does not provide benefit. It does. What has been said is that the benefit of higher bit depth is only relvant for shadows, and that as compression throws away data (not the same as information) from the highlight parts of the photo, it will not impact image quality in a way that you will be able to see.

If you disagree with that, feel free to demonstrate the difference. There has been aple demonstration of what is asserted above.



Apr 12, 2012 at 04:03 PM
mshi
Offline
• • • •
Upload & Sell: Off
p.4 #20 · p.4 #20 · D800: 12-bit RAW vs. 14-bit RAW


Only for shadows? Not entirely true because after all it's an equal opportunity provider. Take a look at some samples showing differences in both both Highlights and Shadows. Visual folks do visual way.

shadows






highlights












Apr 12, 2012 at 04:07 PM
1       2       3      
4
       5       end




FM Forums | Nikon Forum | Join Upload & Sell

1       2       3      
4
       5       end
    
 

You are not logged in. Login or Register

Username   Password    Reset password