• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Touted dynamic range of 1DsIII, 5DII and Nikon D3, D3x or D700: practical benefits?

Asher Kelman

OPF Owner/Editor-in-Chief
We all have faced far too brightly lit scenes with deep shadows where it was tough to grab the entire range of tones. Now we have new remarkable cameras by Nikon, Canon and Sony boasting higher dynamic range.

Sometimes one cannot use flash and bracketing is not feasible ( e.g. for moving objects). In your own work, have you found that your new camera handles wide tonal range much better? Can you now photograph such scenes without loss of detail you had to fight to get previously?

Asher
 

Doug Kerr

Well-known member
Hi, Asher,

We all have faced far too brightly lit scenes with deep shadows where it was tough to grab the entire range of tones. Now we have new remarkable cameras by Nikon, Canon and Sony boasting higher dynamic range.
How do the manufacturers of these cameras express their "improved" dynamic range? Do they state it in an objective form (numerically), inviting comparison with the same parameter for "other" cameras, or do they just say (for example) "greater", "improved", etc.?

In any case, your inquiry is most apt, and I will certainly be looking forward to the reports of our colleagues.

Best regards,

Doug
 

Alain Briot

pro member
I don't use Canon 1DsMk3 or other recent DSLR so I can't really discuss them, however I have noticed an increased ability to capture a wide range of contrast with the Phase One digital backs that I currently use.

I also noticed a larger color palette as well as finer transitions between colors with these same backs.

This is why I use digital backs. The higher resolution is a bonus but can be achieved with DSLRs through stitching multiple captures of the same scene. The wider dynamic range and the larger number of colors captured (which translates into finer gradations within each color and across the entire palette) cannot.
 

Daniel Buck

New member
I've not used one of the latest models, but on my 1Ds2 I quite often notice a bit of noise in the shadows (even when I'm not trying to pull more detail from them) In this respect, better dynamic range would be nice!
 

Ken Tanaka

pro member
"Practical" is a relative value. Many of the most highly revered photographs in history are black-and-white and actually feature very limited "dynamic range". Subject, composition, timing are generally all far more powerful qualities for photographs. But because they're not salable or quantifiable they're not promotionally touted and wouldn't hit home with the legions of technically-oriented buyers representing the "target market", many of whom spend much more time typing than photographing.

Ranting preamble aside ;-) .... yes, the Canon cameras have made some strides that are welcome. But the situation you portray -- dark shadows in bright scenes -- is actually not the primary daily practical advantage of increased dynamic range sensors. Being able to preserve chrominance details is more practical advantage, at least to me, than wrestling with high contrast ratio scenes. I benefit much more by being able to record, say, five tones of blue in one scene than five tones of gray. That's where these cameras have really made progress.

One last observation. During the past years RAW image files have become de rigeur for the dedicated amateur photographer. "No RAW, no sale" has become their chant. That's generally a fitting perspective for higher-end cameras. But it can sometimes be counterproductive mindless dogma with lower-end cameras, as I've recently learned. A great percentage of the developmental expense for lower echelon cameras is devoted toward in-camera image processing. Programming teams beaver-away in the wee hours week after week writing signal processing algorithms to preserve detail, manage noise, record accurate color, and, yes, resurrect details from black graveyards and blown fields. Nobody knows a camera's sensor better than the folks working INSIDE that camera. But most of the benefits of that knowledge are discarded when you select a RAW image. I recently discovered this with my Leica D-Lux 4 (Panasonic LX3), and then again with my Canon G10.

I realize that there will be plenty of readers who will roll their eyes when reading this. But I encourage them to experiment for themselves, secretly if they must. I think some may be rather shocked to discover that pre-cooked image meals can be quite good, perhaps even better than home cooked. And remember, all that counts is the final print and there's no such thing as a 16-bit print (unless you're referring to its price).
 

Asher Kelman

OPF Owner/Editor-in-Chief
.....yes, the Canon cameras have made some strides that are welcome. But the situation you portray -- dark shadows in bright scenes -- is actually not the primary daily practical advantage of increased dynamic range sensors. Being able to preserve chrominance details is more practical advantage, at least to me, than wrestling with high contrast ratio scenes. I benefit much more by being able to record, say, five tones of blue in one scene than five tones of gray. That's where these cameras have really made progress.
I could add blown out highlights. A typical example would be stage lights in concert halls that come directly from above. So the shoulders, arms, fingers and parts of instruments hit directly are easily blown out.

The advanced nature of in certain camera processing in mass produced cameras. This writing of yours is a delightful challenge to photographers to wake up and smell the roses! (This deserves and will make a great new thread).

One last observation. During the past years RAW image files have become de rigeur for the dedicated amateur photographer. "No RAW, no sale" has become their chant. That's generally a fitting perspective for higher-end cameras. But it can sometimes be counterproductive mindless dogma with lower-end cameras, as I've recently learned. A great percentage of the developmental expense for lower echelon cameras is devoted toward in-camera image processing. Programming teams beaver-away in the wee hours week after week writing signal processing algorithms to preserve detail, manage noise, record accurate color, and, yes, resurrect details from black graveyards and blown fields. Nobody knows a camera's sensor better than the folks working INSIDE that camera. But most of the benefits of that knowledge are discarded when you select a RAW image. I recently discovered this with my Leica D-Lux 4 (Panasonic LX3), and then again with my Canon G10.

I realize that there will be plenty of readers who will roll their eyes when reading this. But I encourage them to experiment for themselves, secretly if they must. I think some may be rather shocked to discover that pre-cooked image meals can be quite good, perhaps even better than home cooked. And remember, all that counts is the final print and there's no such thing as a 16-bit print (unless you're referring to its price).

Here your points are well made! Still, some of the capabilities such as bringing out detail from shadows and limiting over exposure of highlights are in a separate chip from companies like Texas Instruments and one might still get the benefits in RAW. Kodak and HP use such chips.

Asher
 

Will Thompson

Well Known Member
Alain, Ken, and Asher. I believe that it is not the dynamic range but the usable dynamic range due to increased bit depth that is of benefit. Increased tonal gradation is another good byproduct of increased bit depth, The more the better when it comes to bit depth, well as long as not going to extremes. In the past the only reason for reduced bit depth has been speed and cost both due to the larger file size required. The ideal situation would be equal bit distribution over the exposure range say 4096 or more at every level. Now that would really be HIGH DEFINITION!
 

Alain Briot

pro member
"Alain, Ken, and Asher. I believe that it is not the dynamic range but the usable dynamic range due to increased bit depth that is of benefit."

I don't see what is the difference between "dynamic range" and "usable dynamic range." To me what matters is that I get more colors, lower contrast, etc. I leave the technical aspects to the engineers!
 

Will Thompson

Well Known Member
Alain, Try (on a 12 bit camera) increasing the exposure 3 stops on a raw file that is shot 3 stops under exposed and you will see what I am referring to. The color is all muddy and the file falls apart. This is not the case with 14 and 16 bit raw files. Every extra bit adds around 1 stop more useable range on the low end before the file falls apart due to lack of tonal gradation!
 

Doug Kerr

Well-known member
Hi, Will,

Alain, Try (on a 12 bit camera) increasing the exposure 3 stops on a raw file that is shot 3 stops under exposed and you will see what I am referring to.
If the shot was "underexposed" by 3 stops, one cannot later "increase the exposure" on the raw file.

Are you referring to "pushing" the development of the raw file by 3 stops? If so, I think it is confusing to refer to that as "increasing the exposure".

Perhaps I am hampered by concentrating on the actual meaning(s) of exposure.

Best regards,

Doug
 

John Sheehy

New member
Alain, Try (on a 12 bit camera) increasing the exposure 3 stops on a raw file that is shot 3 stops under exposed and you will see what I am referring to.

That should work very well with the D3x (in 12-bit mode) and the A900 (12-bit).

The color is all muddy and the file falls apart. This is not the case with 14 and 16 bit raw files. Every extra bit adds around 1 stop more useable range on the low end before the file falls apart due to lack of tonal gradation!

I think you might be confusing bit depth with some other coincidences. In totally controlled situations, where I have full control of all the steps of conversion and precision, I can not find any visible losses when 14 bit RAW data is quantized to 12 bit. The only camera which may be an exception is the Nikon D3x, which has less read noise in 14-bit mode (presumably because the sensor is read more than once and a composite 14-bit image from 12-bit digitizations is created). The D3x can probably use about 13 bits for its "14-bit" capture effectively. And that's state of the art.

Most cameras are worthy of no more than 12 bits. No Canon is. Canon ISO 100 RAWs can be quantized to even 10 bit with no losses in normal exposures and conversions. Almost any camera can get away with 6 bits for super-high-ISOs. There is just far too much noise for great RAW precision to be of value. The noise dithers the image making posterization impossible.

One reason why 14- and 16-bit RAWs may seem superior automatically is that the converters may treat them with more precision. IOW, a 12-bit RAW could be loaded into a RAW converter with "00" tacked onto the end of the digital value for 14-bit data which is fake, but guarantees *better handling* by the converter, much in the same way that it is best to edit 8-bit TIFFs or JPEGs in 16-bit mode. This is an oversight on the part of the converter writers; they should always use lots of precision for conversion; more than the RAW contains.

Also, many lines of cameras have had less analog noise over time, as the bit depth is increased. What one assumes to be due to increased bit depth may easily be attributed to reduced electronic noise.
 

Asher Kelman

OPF Owner/Editor-in-Chief
I shot today in the fabulous Walt Disney Concert Hall. The lighting is a deceptively good to look at people, but not really at all well lit for photography! The light is mostly from above! This creates a high key top heavy effect. With the 5D this blows out shoulders and hands, the 5DII however, handles it well.

Shooting was only during practice sessions, and then for the actual performance just before each piece started and then at then end for the orchestra being applauded. I would have loved to have had more pixels but the 5D2 with the 24 mm f1.4 was able to grab all the info enough to look good and get each person recognizable. The stages in Classical Music locations seem to be lit predominately for a razzle dazzle effect on the audience. The light comes more from above and the face then are shadowed but surprisingly it looks right! That's because those shadows, when the heads are all small, packed across the stage, provide clues to the features. However, the faces are not lit will for any of us to be satisfied. I'm sure shadow highlight tool in photoshop will be a help.

This is where extra BIT depth with usable data included would be helpful. I immediately noticed that the 5DII had better ability to deal with the big jump between over exposed and the shadow areas. Still, I'd like more.

I tried a Phase One on an Alpa but the back some amazing gap of at least several to 10 seconds to finish writing to memory! So Phase One may have the extra BIT depth but that particular back is too slow!

Asher
 

Ken Tanaka

pro member
Sounds like you had some fun, Asher.

FWIW, much professional photography and videography of theatrical events is done during staged rehearsals that are often lit specifically for photography. Back when I was involved in filmmaking I filmed promotional excerpts from plays to be aired on PBS affiliates. I nearly always used lighting schemes very different from performance plans. I know that this is still the case for promotional work.

So don't be too hard on yourself or your cameras when they struggle to keep pace with live theatrical lighting.
 

Asher Kelman

OPF Owner/Editor-in-Chief
Sounds like you had some fun, Asher.
Ken,

It's so much fun and a pleasure to experience the work of devoted faculty and conservatory students. I feel privileged.

FWIW, much professional photography and videography of theatrical events is done during staged rehearsals that are often lit specifically for photography. Back when I was involved in filmmaking I filmed promotional excerpts from plays to be aired on PBS affiliates. I nearly always used lighting schemes very different from performance plans. I know that this is still the case for promotional work.
Ken,

Today we're so fortunate to have very smart cameras with extended light sensitivity and dynamic range. That and Photoshop highlight shadow tool and noise Ninja, although some work, is the road to Nirvana at present.

I tried one Phase one back with ore dynamic range but write speeds were so slow I would be faster with film on a Mamiya 7II!

I'm hoping to get them to put lights coming from the back of the concert hall! A dream for the moment!

So don't be too hard on yourself or your cameras when they struggle to keep pace with live theatrical lighting.

I was surprised how a set up that looks so defined and impressive to the eye, can actually made up of over emphasized lines and shadows. Seemingly the human eye has an adjustable dynamic range for every part of the image. I remember seeing how stage makeup which looks "over-the-top" on close inspection, appears quite normal under the lights.

Anyway as you stated, it is fun and I plan to make the most of it!

Asher
 
Last edited:

Jack_Flesher

New member
We all have faced far too brightly lit scenes with deep shadows where it was tough to grab the entire range of tones. Now we have new remarkable cameras by Nikon, Canon and Sony boasting higher dynamic range.

First off, they tout this added DR by using a custom setting that usually pumps up the low end by a stop, the net effect being protecting the top end from blowing while adding shadow detail. Note that with this "expanded" exposure mode enabled, the highest ISO available is usually also reduced by one stop, ostensibly because the low end is now pumped up one and already at the highest effective ISO...

That said, I do happen to own a 1Ds3 and a MF digital back --- and I can confirm that the MFDB has at least one full stop if not 1-1/2 additional real, usable stops of DR. The difference is readily visible in a typical landscape shot with bright clouds in it...

Cheers,
 

Asher Kelman

OPF Owner/Editor-in-Chief
That said, I do happen to own a 1Ds3 and a MF digital back --- and I can confirm that the MFDB has at least one full stop if not 1-1/2 additional real, usable stops of DR. The difference is readily visible in a typical landscape shot with bright clouds in it...
Jack,

I'm glad to hear that it's of practical value. I've hear good things by John on the math of the nikon D3X and it seems that this too has further headroom than the 1DsIII. I was a little disconcerted by the slow writing speed of a Phase One back on an alpa I saw recently. What back do you have and how fast does it write the files?

Asher
 

Asher Kelman

OPF Owner/Editor-in-Chief
That said, I do happen to own a 1Ds3 and a MF digital back --- and I can confirm that the MFDB has at least one full stop if not 1-1/2 additional real, usable stops of DR. The difference is readily visible in a typical landscape shot with bright clouds in it...
Jack,

I'm glad to hear that it's of practical value. I've heard good things by John Sheehy on the math of the nikon D3X and it seems that this too has further headroom than the 1DsIII. I was a little disconcerted by the slow writing speed of a Phase One back on an alpa I saw recently. What back do you have and how fast does it write the files?

Asher
 

Jack_Flesher

New member
Jack,

I'm glad to hear that it's of practical value. I've heard good things by John Sheehy on the math of the nikon D3X and it seems that this too has further headroom than the 1DsIII. I was a little disconcerted by the slow writing speed of a Phase One back on an alpa I saw recently. What back do you have and how fast does it write the files?

Asher

Hi Asher,


To your first comment, I am not a Nikon shooter, but I think when you are in the D3x's 14-bit capture mode is where you gain the added DR, and then the capture rate slows significantly -- like down to around MF DB status. I'm sure somebody with a D3x can confirm the exact spec...

My MF DB is the P45+, and relative to a DSLR its framerate is s l o w w w. That said, it can produce one frame about every 1.5 seconds or so, and that is WAYYYYY faster than my 4x5 could do! The reality is that for the way I use it -- primarily landscape -- it is more than adequate since I am usually waiting for the light anyway. Alternatively it is usually adequate in the studio too, since it is still faster than most pro packs can recycle at mid-power settings. As any MF DB shooter will tell you, the big benefits over DSLR's are very broad DR, more accurate color and very smooth tonality, all combined with impressive detail.

recentcf029892.jpg


florida_jennifercf029061.jpg
 

Mike Bailey

pro member
I shot today in the fabulous Walt Disney Concert Hall. The lighting is a deceptively good to look at people, but not really at all well lit for photography! The light is mostly from above! This creates a high key top heavy effect. With the 5D this blows out shoulders and hands, the 5DII however, handles it well.

Asher,

Would you be willing to post a couple of full-size cropped examples of the above?

Mike
 
Top