Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
What's with this "ERADR" and "EBTR" stuff?
Page 1 of 3 next> last>>
Nov 23, 2014 11:46:46   #
Uuglypher Loc: South Dakota (East River)
 
What's with this "ERADR" and "EBTR" stuff?

It turns out that most, if not all, consumer/"Pro-sumer"/professional digital cameras are supplied with some amount of dynamic range in excess of that implied by the dynamic range of the JPEG-adjusted histogram and the JPEG thumbnail image provided by the cameras software. Even if you are capturing RAW data, the histogram and thumbnail are specifically attuned to the Dynamic Range ( number of "stops"/E.V.S) allocated to JPEG files extracted from the RAW data collected with each exposure.

There is, beyond the JPEG file's dynamic range, some additional "headroom" of dynamic range there to be used by RAW capture. It may be as little as 1/3 of a stop, up to two full stops ...or more...depending on your particular camera. I say "your paticular camera" because in several instances in which more than one camera of a given camera model has been tested the " Extra RAW-Accessible Dynamic Range" (ERADR) has differed by as much as 2/3 to one and 1/3 stops. Thus, each camera must be tested for its unique amount of ERADR.

Why test for ERADR? Its a crap-shoot!, If yor camera has only 1/3 stop of ERADR, you'll wondrr "Why did I waste my time?" But if your camera can provide you with one to two full stops of additional ERADR, Wouldn't you like to use it? For example if DXOmark rates you new camera's sensor at 12 E.V.S ( landscape) but you find it can provide 13 and 2/3 or 14 full stops, wouldn't you like to use its full complement of dynamic range? An additional two full stops of exposure? Why leave that sitting unused,on the table (or in your sensor?)

It's not brain science...or rocket surgery...to figure out your camera's ERADR.
Figure your camera model provides 11 stops of DR for a JPEG file and the right end of the histogram is where you start clipping highlights.

Set your camera to manual exposure, manual focus, and pick a chosen WB (not AUTO WB).

Choose a scene with a DR that falls within your camera's stated DR. and, for greatest accuracy, do it on a somewhat dull, overcast day and make sure the scene includes some highlights with faintly discernible detail ( zone VIII if you are a zone System user) .

With your camera on a tripod, set your exposure so the right end of your histogram (the "light pile") just barely touches the right end of the histogram frame. That's Exposing to the Right (ETTR).

Then take make a series of additional exposures increasing by 1/3 stop of shutter speed (1/3 slower shutter) until you've shot, say, three additional full stops of exposure ( nine exposures - each 1/3 stop brighter than the previous one). This is Exposing Beyond the Right (EBTR)...beyond the right of the right end of the ol' JPEG histogram frame.

Download the exposures into your computer and open in Adobe Camera RAW,
The thumbnails of these exposures will look overexposed and washed out because if they were JPEG files, they would be. But these are RAW captures taking advantage of the Extra RAW-Accessible Dynamic Range.

You will show these to be properly exposed RAW data by moving the "exposure" slider to the left and tonally normalizing the image. Open, and tonally normalize each image intil you finally find an exposure in which the highlights actually are clipped of their detail. Those exposures not showing clipping tell you your camera's ERADR.

Five additional exposures without clipping? One and two-thirds stop of ERADR.
two additional exposures without clipping? 2/3 st of ERADR
Three additional exposures without clipping? one full stop of ERADR.
Six additional exposures without clipping? two full stops of ERADR.

See? Easy! ( I would say "easy peasy"..but that just sounds unduly weird ')

So, when to use that ERADR? When you really need that extra DR!

On a bright, sunny day, there's no need unless you are after a supper fast shutter and a super small aperture and a low ISO.

But EBTR comes into its own under less-than-ideal lighting. Find your ETTR exposure, then add your camera's ERADR, and ,"there y'go," or "Bob's your uncle!" or "That's it", or "you done it"....or whatever you like to say at such a successful juncture of events.


It's all based upon and derived from what I consider to be the basic Canon of Digital Exposure Knowledge:

Reichmann, Michael, essay in Luminous Landscape, 2003
"Expose Right"
http://www.luminous-landscape.com/tutorials/expose-right.shtml
...and a follow-up in 2011:"Optimizing Exposure"
http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml

Fraser, Bruce,
"RAW Capture, Linear Gamma, and Exposure"Adobe White Paper, 2004,
http://www.adobe.com/digitalimag/pdfs/linear_gamma.pdf

Fraser, Bruce, and Jeff Schewe, "Real World Camera RAW.
Peachpit Press, 20O9,

Schewe, Jeff, "The Digital Negative", Peachpit Press, 2013

Schewe, Jeff, "The Digital Print" , Peachpit Press, 2014

The bottom line is the need to realize that the basic principles of proper film/jpeg exposure and proper RAW data exposure are two completely different concepts.

Vive la difference!

I got quickly sold on the logic and visible proof of ETTR and then as quickly began experimenting with EBTR which I've now happily and productively used and taught EBTR for almost a decade.

Try it!
You'll like it!
If you don't, what's wasted....oh....that's right ...the cost of all those wasted exposures!!!!

Best regards,
Dave Graham

Reply
Nov 23, 2014 20:50:23   #
lamiaceae Loc: San Luis Obispo County, CA
 
Uuglypher wrote:
What's with this "ERADR" and "EBTR" stuff?

It turns out that most, if not all, consumer/"Pro-sumer"/professional digital cameras are supplied with some amount of dynamic range in excess of that implied by the dynamic range of the JPEG-adjusted histogram and the JPEG thumbnail image provided by the cameras software. Even if you are capturing RAW data, the histogram and thumbnail are specifically attuned to the Dynamic Range ( number of "stops"/E.V.S) allocated to JPEG files extracted from the RAW data collected with each exposure.

There is, beyond the JPEG file's dynamic range, some additional "headroom" of dynamic range there to be used by RAW capture. It may be as little as 1/3 of a stop, up to two full stops ...or more...depending on your particular camera. I say "your paticular camera" because in several instances in which more than one camera of a given camera model has been tested the " Extra RAW-Accessible Dynamic Range" (ERADR) has differed by as much as 2/3 to one and 1/3 stops. Thus, each camera must be tested for its unique amount of ERADR.

Why test for ERADR? Its a crap-shoot!, If yor camera has only 1/3 stop of ERADR, you'll wondrr "Why did I waste my time?" But if your camera can provide you with one to two full stops of additional ERADR, Wouldn't you like to use it? For example if DXOmark rates you new camera's sensor at 12 E.V.S ( landscape) but you find it can provide 13 and 2/3 or 14 full stops, wouldn't you like to use its full complement of dynamic range? An additional two full stops of exposure? Why leave that sitting unused,on the table (or in your sensor?)

It's not brain science...or rocket surgery...to figure out your camera's ERADR.
Figure your camera model provides 11 stops of DR for a JPEG file and the right end of the histogram is where you start clipping highlights.

Set your camera to manual exposure, manual focus, and pick a chosen WB (not AUTO WB).

Choose a scene with a DR that falls within your camera's stated DR. and, for greatest accuracy, do it on a somewhat dull, overcast day and make sure the scene includes some highlights with faintly discernible detail ( zone VIII if you are a zone System user) .

With your camera on a tripod, set your exposure so the right end of your histogram (the "light pile") just barely touches the right end of the histogram frame. That's Exposing to the Right (ETTR).

Then take make a series of additional exposures increasing by 1/3 stop of shutter speed (1/3 slower shutter) until you've shot, say, three additional full stops of exposure ( nine exposures - each 1/3 stop brighter than the previous one). This is Exposing Beyond the Right (EBTR)...beyond the right of the right end of the ol' JPEG histogram frame.

Download the exposures into your computer and open in Adobe Camera RAW,
The thumbnails of these exposures will look overexposed and washed out because if they were JPEG files, they would be. But these are RAW captures taking advantage of the Extra RAW-Accessible Dynamic Range.

You will show these to be properly exposed RAW data by moving the "exposure" slider to the left and tonally normalizing the image. Open, and tonally normalize each image intil you finally find an exposure in which the highlights actually are clipped of their detail. Those exposures not showing clipping tell you your camera's ERADR.

Five additional exposures without clipping? One and two-thirds stop of ERADR.
two additional exposures without clipping? 2/3 st of ERADR
Three additional exposures without clipping? one full stop of ERADR.
Six additional exposures without clipping? two full stops of ERADR.

See? Easy! ( I would say "easy peasy"..but that just sounds unduly weird ')

So, when to use that ERADR? When you really need that extra DR!

On a bright, sunny day, there's no need unless you are after a supper fast shutter and a super small aperture and a low ISO.

But EBTR comes into its own under less-than-ideal lighting. Find your ETTR exposure, then add your camera's ERADR, and ,"there y'go," or "Bob's your uncle!" or "That's it", or "you done it"....or whatever you like to say at such a successful juncture of events.


It's all based upon and derived from what I consider to be the basic Canon of Digital Exposure Knowledge:

Reichmann, Michael, essay in Luminous Landscape, 2003
"Expose Right"
http://www.luminous-landscape.com/tutorials/expose-right.shtml
...and a follow-up in 2011:"Optimizing Exposure"
http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml

Fraser, Bruce,
"RAW Capture, Linear Gamma, and Exposure"Adobe White Paper, 2004,
http://www.adobe.com/digitalimag/pdfs/linear_gamma.pdf

Fraser, Bruce, and Jeff Schewe, "Real World Camera RAW.
Peachpit Press, 20O9,

Schewe, Jeff, "The Digital Negative", Peachpit Press, 2013

Schewe, Jeff, "The Digital Print" , Peachpit Press, 2014

The bottom line is the need to realize that the basic principles of proper film/jpeg exposure and proper RAW data exposure are two completely different concepts.

Vive la difference!

I got quickly sold on the logic and visible proof of ETTR and then as quickly began experimenting with EBTR which I've now happily and productively used and taught EBTR for almost a decade.

Try it!
You'll like it!
If you don't, what's wasted....oh....that's right ...the cost of all those wasted exposures!!!!

Best regards,
Dave Graham
What's with this "ERADR" and "EBTR&... (show quote)


But JPGs are all 8-bit, never 16-bit or 32-bit like a TIF, DNG, or other uncompressed file from a 14-bit to 24-bit Raw file.

Reply
Nov 23, 2014 21:49:08   #
DavidPine Loc: Fredericksburg, TX
 
Great posting for those, like me, who have been trying to understand "exposure beyond the right". Thank you Dave
Uuglypher wrote:
What's with this "ERADR" and "EBTR" stuff?

It turns out that most, if not all, consumer/"Pro-sumer"/professional digital cameras are supplied with some amount of dynamic range in excess of that implied by the dynamic range of the JPEG-adjusted histogram and the JPEG thumbnail image provided by the cameras software. Even if you are capturing RAW data, the histogram and thumbnail are specifically attuned to the Dynamic Range ( number of "stops"/E.V.S) allocated to JPEG files extracted from the RAW data collected with each exposure.

There is, beyond the JPEG file's dynamic range, some additional "headroom" of dynamic range there to be used by RAW capture. It may be as little as 1/3 of a stop, up to two full stops ...or more...depending on your particular camera. I say "your paticular camera" because in several instances in which more than one camera of a given camera model has been tested the " Extra RAW-Accessible Dynamic Range" (ERADR) has differed by as much as 2/3 to one and 1/3 stops. Thus, each camera must be tested for its unique amount of ERADR.

Why test for ERADR? Its a crap-shoot!, If yor camera has only 1/3 stop of ERADR, you'll wondrr "Why did I waste my time?" But if your camera can provide you with one to two full stops of additional ERADR, Wouldn't you like to use it? For example if DXOmark rates you new camera's sensor at 12 E.V.S ( landscape) but you find it can provide 13 and 2/3 or 14 full stops, wouldn't you like to use its full complement of dynamic range? An additional two full stops of exposure? Why leave that sitting unused,on the table (or in your sensor?)

It's not brain science...or rocket surgery...to figure out your camera's ERADR.
Figure your camera model provides 11 stops of DR for a JPEG file and the right end of the histogram is where you start clipping highlights.

Set your camera to manual exposure, manual focus, and pick a chosen WB (not AUTO WB).

Choose a scene with a DR that falls within your camera's stated DR. and, for greatest accuracy, do it on a somewhat dull, overcast day and make sure the scene includes some highlights with faintly discernible detail ( zone VIII if you are a zone System user) .

With your camera on a tripod, set your exposure so the right end of your histogram (the "light pile") just barely touches the right end of the histogram frame. That's Exposing to the Right (ETTR).

Then take make a series of additional exposures increasing by 1/3 stop of shutter speed (1/3 slower shutter) until you've shot, say, three additional full stops of exposure ( nine exposures - each 1/3 stop brighter than the previous one). This is Exposing Beyond the Right (EBTR)...beyond the right of the right end of the ol' JPEG histogram frame.

Download the exposures into your computer and open in Adobe Camera RAW,
The thumbnails of these exposures will look overexposed and washed out because if they were JPEG files, they would be. But these are RAW captures taking advantage of the Extra RAW-Accessible Dynamic Range.

You will show these to be properly exposed RAW data by moving the "exposure" slider to the left and tonally normalizing the image. Open, and tonally normalize each image intil you finally find an exposure in which the highlights actually are clipped of their detail. Those exposures not showing clipping tell you your camera's ERADR.

Five additional exposures without clipping? One and two-thirds stop of ERADR.
two additional exposures without clipping? 2/3 st of ERADR
Three additional exposures without clipping? one full stop of ERADR.
Six additional exposures without clipping? two full stops of ERADR.

See? Easy! ( I would say "easy peasy"..but that just sounds unduly weird ')

So, when to use that ERADR? When you really need that extra DR!

On a bright, sunny day, there's no need unless you are after a supper fast shutter and a super small aperture and a low ISO.

But EBTR comes into its own under less-than-ideal lighting. Find your ETTR exposure, then add your camera's ERADR, and ,"there y'go," or "Bob's your uncle!" or "That's it", or "you done it"....or whatever you like to say at such a successful juncture of events.


It's all based upon and derived from what I consider to be the basic Canon of Digital Exposure Knowledge:

Reichmann, Michael, essay in Luminous Landscape, 2003
"Expose Right"
http://www.luminous-landscape.com/tutorials/expose-right.shtml
...and a follow-up in 2011:"Optimizing Exposure"
http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml

Fraser, Bruce,
"RAW Capture, Linear Gamma, and Exposure"Adobe White Paper, 2004,
http://www.adobe.com/digitalimag/pdfs/linear_gamma.pdf

Fraser, Bruce, and Jeff Schewe, "Real World Camera RAW.
Peachpit Press, 20O9,

Schewe, Jeff, "The Digital Negative", Peachpit Press, 2013

Schewe, Jeff, "The Digital Print" , Peachpit Press, 2014

The bottom line is the need to realize that the basic principles of proper film/jpeg exposure and proper RAW data exposure are two completely different concepts.

Vive la difference!

I got quickly sold on the logic and visible proof of ETTR and then as quickly began experimenting with EBTR which I've now happily and productively used and taught EBTR for almost a decade.

Try it!
You'll like it!
If you don't, what's wasted....oh....that's right ...the cost of all those wasted exposures!!!!

Best regards,
Dave Graham
What's with this "ERADR" and "EBTR&... (show quote)

Reply
 
 
Nov 23, 2014 23:19:38   #
Uuglypher Loc: South Dakota (East River)
 
DavidPine wrote:
Great posting for those, like me, who have been trying to understand "exposure beyond the right". Thank you Dave


quote=DavidPine]Great posting for those, like me, who have been trying to understand "exposure beyond the right". Thank you Dave[/quote]

You are very welcome, Dave.
i feel fortunate in having,quite early, developed interest in the realm of using all the dynamic range available for RAW data capture, and must ascribe that to the writings and work of Michael Reichmann, Bruce Fraser, Jeff Schewe, and the team that Adobe tasked with the continuing development of Adobe Camera Raw.
EBTR has been the single greatest influence in aiding my rational development as a digital photographer with firm grounding in my more than half century experience as a film photographer. EBTR has been the single greatest tool I've found that can reliably and consistently improve image data quality, and thereby image quality.

If questions occur as you pursue EBTR, don't hesitate to ask.

Best regards, Dave,

Dave Graham

Reply
Nov 23, 2014 23:28:17   #
Uuglypher Loc: South Dakota (East River)
 
lamiaceae wrote:
But JPGs are all 8-bit, never 16-bit or 32-bit like a TIF, DNG, or other uncompressed file from a 14-bit to 24-bit Raw file.


Hi, lamiaceae,
I have to admit that although your statement is true, I do not understand its relevance to the topic at hand, the use of EBTR to permit utilization of the heretofore unutilized Extra RAW-Accessible Dynamic Range (EBTR) that has been lurking, unused and unheralded, off beyond tje right side of the JPEG-adjusted histogram in your, and my, cameras.

Dave Graham

Reply
Nov 24, 2014 00:37:36   #
blackest Loc: Ireland
 
Uuglypher wrote:
Hi, lamiaceae,
I have to admit that although your statement is true, I do not understand its relevance to the topic at hand, the use of EBTR to permit utilization of the heretofore unutilized Extra RAW-Accessible Dynamic Range (EBTR) that has been lurking, unused and unheralded, off beyond tje right side of the JPEG-adjusted histogram in your, and my, cameras.

Dave Graham


This is a bit murky and will probably need correcting but off the top of my head each f stop is a doubling of the light and with raw files the number of bits more or less corresponds to the doubling or halving of the light. (you wouldnt actually rely on 1 bit to decide a tone)
1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192
what should be fairly clear is that there are a lot more bit values that equate to a bright zone as compared with a dark zone. Camera sensors basically count photons but there is also noise and that noise is counted a long with the photons. This is most noticeable in the dark parts of an image (since such a small range of values represents a dark tone) and in the brightest tones the noise is much smaller than the photons counted.

so expose to the right essentially means putting the pixel counts into the bigger buckets which gives the noise less effect and then you shift the tones back to where they should be when you process the raw file.

jpeg files are 8 bit but they are not like raw files the pixel values don't relate linearly the way they do in a raw file. An S curve has been applied i think its what is called a gamma correction. anyway this essentially mimics our eyes response to light and it looks fine to us, (film managed to do this without informing us of the math). it does mean though that a jpeg cannot and doesn't represent every shade of every colour however it doesn't matter because our eyes can't tell the difference. But it then does lead into the confusion of different colour spaces.

Anyway the 8 bit per colour of jpegs has nothing to do with expose to the right. if your like me you probably thought 8bit jpegs were just truncating the numbers like calling 1.25 1 but no its not that simple.

does that help or just confuse?

Reply
Nov 24, 2014 01:28:58   #
Uuglypher Loc: South Dakota (East River)
 
blackest wrote:
This is a bit murky and will probably need correcting but off the top of my head each f stop is a doubling of the light and with raw files the number of bits more or less corresponds to the doubling or halving of the light. (you wouldnt actually rely on 1 bit to decide a tone)
1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192
what should be fairly clear is that there are a lot more bit values that equate to a bright zone as compared with a dark zone. Camera sensors basically count photons but there is also noise and that noise is counted a long with the photons. This is most noticeable in the dark parts of an image (since such a small range of values represents a dark tone) and in the brightest tones the noise is much smaller than the photons counted.

so expose to the right essentially means putting the pixel counts into the bigger buckets which gives the noise less effect and then you shift the tones back to where they should be when you process the raw file.

jpeg files are 8 bit but they are not like raw files the pixel values don't relate linearly the way they do in a raw file. An S curve has been applied i think its what is called a gamma correction. anyway this essentially mimics our eyes response to light and it looks fine to us, (film managed to do this without informing us of the math). it does mean though that a jpeg cannot and doesn't represent every shade of every colour however it doesn't matter because our eyes can't tell the difference. But it then does lead into the confusion of different colour spaces.

Anyway the 8 bit per colour of jpegs has nothing to do with expose to the right. if your like me you probably thought 8bit jpegs were just truncating the numbers like calling 1.25 1 but no its not that simple.

does that help or just confuse?
This is a bit murky and will probably need correct... (show quote)


Partly, a bit of each; let's consider some misunderstandings that might arose from this central point which requires some clarification:

"So expose to the right essentially means putting the pixel counts into the bigger buckets which gives the noise less effect and then you shift the tones back to where they should be when you process the raw file."

ETTR, and even beyond the right with EBTR, doesn't change the pixels into "bigger buckets". It simply moves the "light pile" as far as possible to the right, filling the photocells (pixels) and their electronwells as far as possible away from the left end of the histogram frame that represents "the hometown of noise" (low S:N ratios) and moving even the darkest values of the image up into the realms where greater numbers of grays (more shadow detail) can be captured. Not only the brightest values, but the darkest as well are moved as far to the right as possible, where greater numbers of gray values ( and cusps of detail) can be captured and giving every image data point a higher signal:noise ratio.

When the image is tonally normalized in ACR by linear processing, the S:N ratio of all those data points and all the cusps of detail in that full tonal spectrum are retained. Image data quality is dependent on where the data is originally collected, not where it winds up with tonal normalization ( if normalized by linear processing in the Raw converter/ACR.

We must never forget that if tonal normalization is not performed by linear processing in the Raw converter (ACR) but by non-linear processing under the "image"menu in Photoshop, much of the image data throughout the brightness range will be lost along with the high detail rendition of the otherwise complete complement of data points.

I hope this helps clarify this aspect of the benefits to be derived from collecting our image data points as far to the right as possible and taking as great advantage as possible of the Extra RAW-Accessible Dynamic Range (ERADR) of each of your cameras.

Thanks, Blackest, for your question.

I'll discuss some points of confusion relative to the statement on Bit environments in a subsequent response.

Best regards,
Dave Graham

Reply
 
 
Nov 24, 2014 08:54:08   #
Uuglypher Loc: South Dakota (East River)
 
[quote=blackest]This is a bit murky and will probably need correcting but off the top of my head each f stop is a doubling of the light and with raw files the number of bits more or less corresponds to the doubling or halving of the light. (you wouldnt actually rely on 1 bit to decide a tone)
1 2 4 8 16 32 64 128 256 512 1024 2048 4096 8192

I agree, blackest; let's clear up some of the murkiness.

The bit-depth of any image file determines the number of tones of grat it can register from black to white. Without getting in the mathematics of it all:

one bit: black and white
two bits: black, =4 tones
three bits:black, =8 tones
....
.
eight bits = 256 tones (JPEG files)
....
twelve bits= 4,096 tones (Raw data from some cameras)
....
fourteen bits= 16,384 tones (Raw data from other cameras.)

The number of tones of gray capable of being registered is important because the border each adjacent pair of grays is where therir difference can be accentuated by brightening the lighter amd farkening the farker. Thus the border of two adjacent grays is a potential cusp of detail. The more finely scaled the grays in the image, the more potential detail can be captured.


Bits are not to be confused with Dynamic Range.

Dunamic Range is the number of stops of Exposure that can be regitered by your camera's sensor, whether you capture a JPEG image file or RAW image data.
Since each stop registers twice the light of the next brighter stop, the brightest stop used in making the exposure registers half the image data points in the entire image which is why it is important to capture the image data of any image as far "to the Right" as possible.

Best regards,
Dave Graham

Reply
Nov 24, 2014 11:49:37   #
kymarto Loc: Portland OR and Milan Italy
 
I have to say that with the clean shadows of my D800 I much prefer not to expose for the brightest highlights plus alpha extra headroom. This screws up the tone curve mightily in the high key areas, even if not blown out. I'll always expose normally and pull up shadows a bit. Much much more pleasing...

Reply
Nov 24, 2014 13:43:00   #
Uuglypher Loc: South Dakota (East River)
 
Hi, kymarto,
I appreciate your input in regard to the high end curve problems in using the max of bright end DR in your D800.
That is one of the models for which I've been able to test several examples for ERADR and have found as little as 2/3 stop to a maximum (so far) of two and 1/3 stop ERADR.
Have you, by any fotuitous chance, tested your particular D800 to determine its ERADR? In some cameras, as the maximum DR is approached the shoulder curve can be less-than optimally predictable., which is why testing each individual camera is necessary before embarking with security on the path of EBTR.

I ask because it would be helpful to know if you ran into the problem you describe in a D800 with minimal ERADR, in other words one with a DR actually close to that stated for that model, or if it actually was one with, say, close to an extra two full stops or more of ERADR.

Many thanks for your taking the time to comment on your observation. I hope you can understand my interest in your D800's actual ERADR.

Dave Graham

Reply
Nov 24, 2014 13:52:04   #
Uuglypher Loc: South Dakota (East River)
 
ERRATUM NOTICE
ERRATUM NOTICE

Egad, please accept my apology for the following misstatement (that's also called a serious, egregious error!)

.
"Since each stop registers twice the light of the next brighter stop,...,
SHOULD READ:

"Since each stop registers twice the light of the adjacent LESS-BRIGHT stop,..., "

Please accept my apologies for the confusion that must have caused.
My embarassment about this knows no bounds.

David Graham

Reply
 
 
Nov 24, 2014 13:56:42   #
jcboy3
 
Uuglypher wrote:
So, when to use that ERADR? When you really need that extra DR!

On a bright, sunny day, there's no need unless you are after a supper fast shutter and a super small aperture and a low ISO.


I think you mean slow shutter, large aperture (small aperture number) or high ISO. You would use a slow shutter to keep it within flash sync speed for fill flash in shadows or blur motion. You would use a large aperture for narrow depth of field (DOF). And I don't have any clue why one would punch up the ISO on a bright day.

Otherwise, a useful description.

Reply
Nov 24, 2014 15:49:32   #
Uuglypher Loc: South Dakota (East River)
 
Uuglypher wrote:
So, when to use that ERADR? When you really need that extra DR!

On a bright, sunny day, there's no need unless you are after a supper fast shutter and a super small aperture and a low ISO.


I think you mean slow shutter, large aperture (small aperture number) or high ISO. You would use a slow shutter to keep it within flash sync speed for fill flash in shadows or blur motion. You would use a large aperture for narrow depth of field (DOF). And I don't have any clue why one would punch up the ISO on a bright day.

Otherwise, a useful description.

jcboy3
xxxxxxxx
I appreciate ypu response, jcboy3,

My point with that less-than-well-stated statement was that even on a bright day there are those times when I wish for more dynamic range to be able to make better creative use of the exposure limiters, a smaller aperture and/or a faster shutter WITHOUT THE NEED for an unduly high ISO, yet still be able to pull the light pile as far to the right as possible.
Those extra stops of exposure may come in handy even on a bright day if shadow detail n Zone III or Zone II are important. I see more available dynamic range as money in the exposure bank. I'm always happy for anything that will let me buy an exta stop or two of exposure, because it always expands creative options.

I did, early on, recognize the need to mentally envision a RAW data capture histogram frame that extends a sufficient degree to the right of the end of the old JPEG-adjusted frame to accomodate the extra ynamic range available for RAW capture.

Those with concerns dictated by use of fill flash will have other perceptions, but as a landscape photogrpher, flash concerns don't enter into my thinking.

Thanks, jcboy3, for calling the need for more clarity to my attention.

And, as kymarto mentioned above, some peculiaities in the upper end of the presumed straight line response (in the area of the "shoulder" of the DH curve) may occur, requiring adjustments be made in detrmining useful extra dynamic range in each individual camera.

Again, "your mileage may vary".

Dave Graham

Reply
Nov 24, 2014 18:48:29   #
kymarto Loc: Portland OR and Milan Italy
 
I do a lot of HDR, so I have scores of exposure series 1 stop apart. IIRC the D800E has about 1.5 stops headroom in raw. I can check more closely and get back to you. I do know that when I choose to try to recover DR in a high contrast scene, I almost always get more pleasing results pulling up shadows rather than pulling down highlights.

I also use the Canon 5D ii and iii, and those cameras are a different story due to the very high shadow noise at low ISOs. There you are better off not to raise shadows too much--also with the Nikon at higher ISOS.

Reply
Nov 24, 2014 19:45:44   #
Uuglypher Loc: South Dakota (East River)
 
kymarto wrote:
I do a lot of HDR, so I have scores of exposure series 1 stop apart. IIRC the D800E has about 1.5 stops headroom in raw. I can check more closely and get back to you. I do know that when I choose to try to recover DR in a high contrast scene, I almost always get more pleasing results pulling up shadows rather than pulling down highlights.

I also use the Canon 5D ii and iii, and those cameras are a different story due to the very high shadow noise at low ISOs. There you are better off not to raise shadows too much--also with the Nikon at higher ISOS.
I do a lot of HDR, so I have scores of exposure se... (show quote)


Thanks, kymarto,
If, when youcheck the D800 more closely, you could check with 1/3 stop intervals, that would be helpful. Same with the 5Dii and 5D iii if possible.
When you "...try to recover DR in a high contrast scene" are you trying to tonally normalize (within the high contrast context) in ACR with linear processing, or in PS with non-linear processing?

Thanks,
Dave

Reply
Page 1 of 3 next> last>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.