Sign in with
Sign up | Sign in
Your question

Correct exposure for digital ?

Last response: in Digital Camera
Share
Anonymous
January 29, 2005 8:55:27 PM

Archived from groups: rec.photo.digital (More info?)

Okay, a "newbie" question.

With film I don't have a problem working out what the "correct"
exposure is. I just look at the negative. If I want to be more
accurate I might use a grey scale and/or something like a
densitometer.

How do I work out the "optimum" exposure for digital ? Can someone
here explain the best method that they know of and/or refer me to a
web site for the answer please ? TIA.

Regards, John.
Anonymous
January 29, 2005 8:55:28 PM

Archived from groups: rec.photo.digital (More info?)

John Fitzsimons wrote:
> Okay, a "newbie" question.
>
> With film I don't have a problem working out what the "correct"
> exposure is. I just look at the negative. If I want to be more
> accurate I might use a grey scale and/or something like a
> densitometer.
>
> How do I work out the "optimum" exposure for digital ? Can someone
> here explain the best method that they know of and/or refer me to a
> web site for the answer please ? TIA.
>
> Regards, John.
>

As far as exposure goes, whatever works for film will work for digital.
Sunny 16 Rule
Incident Light Meter
Reflective Light Meter
Grey Card
Other....
In addition, some digital cameras present you with a Histogram of the
scene. You can then adjust f-stop, shutter speed, and/or ISO # to get
the histogram looking right.
Bob Williams
Anonymous
January 29, 2005 8:55:28 PM

Archived from groups: rec.photo.digital (More info?)

John Fitzsimons wrote:
> Okay, a "newbie" question.
>
> With film I don't have a problem working out what the "correct"
> exposure is. I just look at the negative. If I want to be more
> accurate I might use a grey scale and/or something like a
> densitometer.
>
> How do I work out the "optimum" exposure for digital ? Can someone
> here explain the best method that they know of and/or refer me to a
> web site for the answer please ? TIA.
>
> Regards, John.

You work out the exposure the same way you did, if you did it right,
with film. You look at the results.

If you want to view the images on the screen, you view them there. If
you want to look at the result on a print, you look at them there (using the
printing system you will use for your final product). It is the result that
counts not the path you take to get there.

All the other stuff is nothing more than the tools to help get you to
the end. The tools may be wrong. The end product evaluation is never
wrong.


--
Joseph Meehan

26 + 6 = 1 It's Irish Math
Related resources
January 29, 2005 8:55:28 PM

Archived from groups: rec.photo.digital (More info?)

John Fitzsimons wrote:
> Okay, a "newbie" question.
>
> With film I don't have a problem working out what the "correct"
> exposure is. I just look at the negative. If I want to be more
> accurate I might use a grey scale and/or something like a
> densitometer.
>
> How do I work out the "optimum" exposure for digital ? Can someone
> here explain the best method that they know of and/or refer me to a
> web site for the answer please ? TIA.
>
> Regards, John.
>
Use your usual method. Then, to be more confident that you'll get it right, most digicams can be set to take a set of three
shots in fast sequence, each with a slightly different exposure setting. It costs nothing to do this, since you don't have
to pay to buy film and get it developed, and you only print the best of the three.
Anonymous
January 29, 2005 8:55:29 PM

Archived from groups: rec.photo.digital (More info?)

On Sat, 29 Jan 2005 02:07:21 -0800, Bob Williams wrote:

> John Fitzsimons wrote:
>> Okay, a "newbie" question.
>>
>> With film I don't have a problem working out what the "correct"
>> exposure is. I just look at the negative. If I want to be more
>> accurate I might use a grey scale and/or something like a
>> densitometer.
>>
>> How do I work out the "optimum" exposure for digital ? Can someone
>> here explain the best method that they know of and/or refer me to a
>> web site for the answer please ? TIA.
>>
>> Regards, John.
>>
>
> As far as exposure goes, whatever works for film will work for digital.
> Sunny 16 Rule
> Incident Light Meter
> Reflective Light Meter
> Grey Card
> Other....

Not *quite* true, reciprocity being an example. But in normal lighting I
agree. Additionally it should be treated like reversal film - expose for the
highlights, not the shadows.

> In addition, some digital cameras present you with a Histogram of the
> scene. You can then adjust f-stop, shutter speed, and/or ISO # to get
> the histogram looking right.

This is the most important exposure tool to master. This is a useful
article: http://www.luminous-landscape.com/tutorials/expose-righ...

--
John Bean

In all large corporations, there is a pervasive fear that someone, somewhere
is having fun with a computer on company time. Networks help alleviate that
fear (John C. Dvorak)
Anonymous
January 29, 2005 8:55:29 PM

Archived from groups: rec.photo.digital (More info?)

Joseph Meehan wrote:
> John Fitzsimons wrote:
>
>>Okay, a "newbie" question.
>>
>>With film I don't have a problem working out what the "correct"
>>exposure is. I just look at the negative. If I want to be more
>>accurate I might use a grey scale and/or something like a
>>densitometer.
>>
>>How do I work out the "optimum" exposure for digital ? Can someone
>>here explain the best method that they know of and/or refer me to a
>>web site for the answer please ? TIA.
>>
>>Regards, John.
>
>
> You work out the exposure the same way you did, if you did it right,
> with film. You look at the results.
>
> If you want to view the images on the screen, you view them there. If
> you want to look at the result on a print, you look at them there (using the
> printing system you will use for your final product). It is the result that
> counts not the path you take to get there.
>
> All the other stuff is nothing more than the tools to help get you to
> the end. The tools may be wrong. The end product evaluation is never
> wrong.
>
>
The problem is with digital you have to be sure you do not
saturate the highlights. Look at your histograms when you
take a picture to be sure you have a good distribution
of intensities and the upper end is not clipped.
On Canon cameras (all the ones I've seen) blink
areas in the image on the LCD screen if the image
data are saturated. If so, take another image
with exposure compensation.

Roger
http://www.clarkvision.com
Anonymous
January 29, 2005 8:55:29 PM

Archived from groups: rec.photo.digital (More info?)

In message <NKKKd.63723$re1.61375@fe2.columbus.rr.com>,
"Joseph Meehan" <sligojoe_Spamno@hotmail.com> wrote:

> If you want to view the images on the screen, you view them there. If
>you want to look at the result on a print, you look at them there (using the
>printing system you will use for your final product). It is the result that
>counts not the path you take to get there.

I disagree. The capture is everything. What you do to get it to any
medium is post-processing.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
January 29, 2005 8:55:30 PM

Archived from groups: rec.photo.digital (More info?)

John Bean wrote:
>
> Not *quite* true, reciprocity being an example. But in normal lighting I
> agree. Additionally it should be treated like reversal film - expose for the
> highlights, not the shadows.
>
>
>>In addition, some digital cameras present you with a Histogram of the
>>scene. You can then adjust f-stop, shutter speed, and/or ISO # to get
>>the histogram looking right.
>
>
> This is the most important exposure tool to master. This is a useful
> article: http://www.luminous-landscape.com/tutorials/expose-righ...


Hmmm... from that link:

"With raw capture it is not as bad as with JPEG capture, since with raw
what you care about is clipping in the native camera color space, rather
than in the working RGB space (since conversion to working space happens
after the raw converter's tone adjustments). Lots of colors clip in sRGB
or even Adobe RGB that don't clip in camera native space."

Shooting landscapes in winter with a bright cloudy sky & shady ground is
VERY difficult. I tried exposing to avoid flashing highlights and the
ground was black. I didn't realize with RAW I can leave a bit of
flashing highlights. Even with that I need to merge two exposures for
many of these extreme contrast shots.
Anonymous
January 29, 2005 8:55:30 PM

Archived from groups: rec.photo.digital (More info?)

In message <41FB8EEA.5080701@qwest.net>,
"Roger N. Clark (change username to rnclark)" <username@qwest.net>
wrote:

>The problem is with digital you have to be sure you do not
>saturate the highlights. Look at your histograms when you
>take a picture to be sure you have a good distribution
>of intensities and the upper end is not clipped.
>On Canon cameras (all the ones I've seen) blink
>areas in the image on the LCD screen if the image
>data are saturated. If so, take another image
>with exposure compensation.

That's a good, safe default method for someone who doesn't know the
camera, but as one will learn, if they really look into things, the
histogram may clip when the RAW data is still 2 stops short of clipping,
and visa-versa.

The histogram is weighted luminance on most cameras, so clipped blue or
red data may not show up as clipped, if the subject is highly saturated,
even in JPEG mode. The other extreme, such as if the Canon 20D is set
to a cool-light white balance, the red channel in the RAW data can have
more than 2 stops more headroom than what the JPEG clips.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
January 29, 2005 8:55:31 PM

Archived from groups: rec.photo.digital (More info?)

On Sat, 29 Jan 2005 09:19:34 -0800, paul wrote:

> John Bean wrote:
>>
>> Not *quite* true, reciprocity being an example. But in normal lighting I
>> agree. Additionally it should be treated like reversal film - expose for the
>> highlights, not the shadows.
>>
>>
>>>In addition, some digital cameras present you with a Histogram of the
>>>scene. You can then adjust f-stop, shutter speed, and/or ISO # to get
>>>the histogram looking right.
>>
>>
>> This is the most important exposure tool to master. This is a useful
>> article: http://www.luminous-landscape.com/tutorials/expose-righ...
>
>
> Hmmm... from that link:
>
> "With raw capture it is not as bad as with JPEG capture, since with raw
> what you care about is clipping in the native camera color space, rather
> than in the working RGB space (since conversion to working space happens
> after the raw converter's tone adjustments). Lots of colors clip in sRGB
> or even Adobe RGB that don't clip in camera native space."
>
> Shooting landscapes in winter with a bright cloudy sky & shady ground is
> VERY difficult. I tried exposing to avoid flashing highlights and the
> ground was black. I didn't realize with RAW I can leave a bit of
> flashing highlights. Even with that I need to merge two exposures for
> many of these extreme contrast shots.

Not necessarily, colour clipping - falling outside the current gamut - is
not the same thing as highlight clipping, but both are reduced dramatically
by not converting to JPEG in the camera. A side effect of the conversion is
to downsample the depth to 8-bit, making shadow recovery very difficult
without causing posterisation. If using raw there is an advantage to be
gained by recovering the highlights in the raw converter, converting to a
16-bit colour space of your choice and recovering the shadows by
manipulating the low tones. Editing in 16-bit mode from a raw conversion
offers opportunities to reveal much more image data from a single exposure.

--
John Bean

I always keep a supply of stimulant handy in case I see a snake - which I
also keep handy (W. C. Fields)
Anonymous
January 29, 2005 10:10:22 PM

Archived from groups: rec.photo.digital (More info?)

JPS@no.komm wrote:
> In message <NKKKd.63723$re1.61375@fe2.columbus.rr.com>,
> "Joseph Meehan" <sligojoe_Spamno@hotmail.com> wrote:
>
>> If you want to view the images on the screen, you view them
>> there. If you want to look at the result on a print, you look at
>> them there (using the printing system you will use for your final
>> product). It is the result that counts not the path you take to get
>> there.
>
> I disagree. The capture is everything. What you do to get it to any
> medium is post-processing.

Capture is the first step and if you don't get it right, you can't
change it later. However the real measure of the capture is the final
result.

>
> <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
> John P Sheehy <JPS@no.komm>
>> <<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

--
Joseph Meehan

26 + 6 = 1 It's Irish Math
Anonymous
January 29, 2005 10:44:44 PM

Archived from groups: rec.photo.digital (More info?)

In message <ef6dnXBxErc1WGbcRVn-3Q@speakeasy.net>,
paul <paul@not.net> wrote:

>Hmmm... from that link:
>
>"With raw capture it is not as bad as with JPEG capture, since with raw
>what you care about is clipping in the native camera color space, rather
>than in the working RGB space (since conversion to working space happens
>after the raw converter's tone adjustments). Lots of colors clip in sRGB
>or even Adobe RGB that don't clip in camera native space."
>
>Shooting landscapes in winter with a bright cloudy sky & shady ground is
>VERY difficult. I tried exposing to avoid flashing highlights and the
>ground was black. I didn't realize with RAW I can leave a bit of
>flashing highlights. Even with that I need to merge two exposures for
>many of these extreme contrast shots.

.... and how much would it cost to show an RGB histogram for the RAW data
on the LCD of any camera? Nothing, except a couple hours of some
programmer's time. These companies make me sick. Their high-end
cameras are better enough in so many ways, to justify their higher cost;
there is no reason to purposely retard the $800-$2000 DSLRs.

Bottom line: I am not going to pay $4000 or $8000 to get a proper RAW
histogram; I will buy such cameras based on resolution, ruggedness,
speed, etc. The manufacturers screw us for buying the $1500 cameras.

Imagine this; you have a RAW RGB histogram (0 through 4095, times three
channels) on the back of the camera. You are working under studio
conditions, and can control the color of the light sources or have a
wide array of colored camera filters. You take your shot at ISO 100.
You see that the green highlights have a stop of headroom, the red
channel has two stops, and the blue channel has a stop-and a half. You
put on a magenta filter, increase exposure by 2.5 stops, and shoot
again. The histogram is showing a max of 4070-4090 in each channel.

What do you have? You have the best exposure you can hope to get, and
about ISO 40 or so, to boot, with slighty over a bit more bit depth in
the green channel, almost two in the blue, and over two in the red
channel, with signal-to-noise ratios to match.

Current digital photography is in the dark ages.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
January 30, 2005 2:15:29 AM

Archived from groups: rec.photo.digital (More info?)

In message <htjnqxj9vm6d.dlg@waterfoot.net>,
John Bean <waterfoot@gmail.com> wrote:

>Editing in 16-bit mode from a raw conversion
>offers opportunities to reveal much more image data from a single exposure.

There is nothing lacking in dynamic range on a sensitive, low-noise
DSLR, when you actually *use* the available dynamic range. I spend a
lot of time looking at RAW data, and often a shot taken with the
camera's auto-exposure doesn't register more than 400 out of 4095 in any
color channel! That is a very inefficient way of doing exposure.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
January 31, 2005 12:04:24 PM

Archived from groups: rec.photo.digital (More info?)

On Sat, 29 Jan 2005 11:25:43 +0000, John Bean <waterfoot@gmail.com>
wrote:

>On Sat, 29 Jan 2005 02:07:21 -0800, Bob Williams wrote:

>> John Fitzsimons wrote:

Thanks everyone for your contributions. :-)

< snip >

>>> How do I work out the "optimum" exposure for digital ?

< snip >

>> As far as exposure goes, whatever works for film will work for digital.
>> Sunny 16 Rule
>> Incident Light Meter
>> Reflective Light Meter
>> Grey Card
>> Other....

Not sure how one uses a grey card with a digital camera. Care to
explain that a little ?

>Not *quite* true, reciprocity being an example. But in normal lighting I
>agree. Additionally it should be treated like reversal film - expose for the
>highlights, not the shadows.

Okay.

>> In addition, some digital cameras present you with a Histogram of the
>> scene. You can then adjust f-stop, shutter speed, and/or ISO # to get
>> the histogram looking right.

Yes, but wouldn't where the histogram is depend on whether one had a
high key subject or a low key subject (lighting) ? It could be easily
moved left or right.

One could certainly visually alter things BUT I was hoping that there
was a less "subjective" method of doing things.

>This is the most important exposure tool to master. This is a useful
>article: http://www.luminous-landscape.com/tutorials/expose-righ...

I agree. I have seen it before but appreciate the reminder. But it
still depends on visual assessment. In film use there is nothing
"subjective" in assessing the density of a grey scale by a
densitometer.

I guess, in light of the above link, what I would like to do is to
photograph a scene inside, or outside, that has exactly the 5 stop
range. Then if info was lost either end of the histogram I could
increase/decrease exposure accordingly. Perhaps a stepped grey
scale could be used like that ?

Regards, John.
Anonymous
January 31, 2005 12:04:25 PM

Archived from groups: rec.photo.digital (More info?)

In message <7dmqv0tnqmuaumevb9n77najjfucnts33g@4ax.com>,
John Fitzsimons <johnf@net2000.com.au> wrote:

>Yes, but wouldn't where the histogram is depend on whether one had a
>high key subject or a low key subject (lighting) ? It could be easily
>moved left or right.

There isn't one way a histogram should be; it depends on the effect you
want, and this is especially true if you are not going to have much to
do with post-processing (such as when you take the camera's JPGs to a
print service). You really need to know what shape the histogram will
be before you even see it, and actually be interested in *where* the
shape falls.

If you're shooting RAW and doing your own post-processing, then you
don't need to keep a dark subject low in the exposure, etc. What you
want then is to "expose to the right", making sure that only unwanted
details get clipped (within the limits of ISO, aperture, and shutter
speed). Unfortunately, most cameras only tell you what gets clipped in
a JPEG, not in the RAW data.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
January 31, 2005 3:24:35 PM

Archived from groups: rec.photo.digital (More info?)

JPS@no.komm wrote:
>
> Unfortunately, most cameras only tell you what gets clipped in
> a JPEG, not in the RAW data.

You know, I tried overexposing a tad on my D70 yesterday and the blown
highlights were indeed blown even in RAW. I don't know, maybe a slight
bit of leeway, I didn't do a real scientific study but I'm inclined to
continue to correct to avoid blinking highlights. I have realized though
that a second lighter shot is needed as the noise in shadows can be
horrible. It's a lot of work merging but sometimes necessary.
Anonymous
January 31, 2005 4:28:31 PM

Archived from groups: rec.photo.digital (More info?)

"paul" <paul@not.net> wrote in message
news:fbydnUSB7KuYCWPcRVn-gg@speakeasy.net...
> JPS@no.komm wrote:
> >
> > Unfortunately, most cameras only tell you what gets clipped in
> > a JPEG, not in the RAW data.
>
> You know, I tried overexposing a tad on my D70 yesterday and the blown
> highlights were indeed blown even in RAW. I don't know, maybe a slight
> bit of leeway, I didn't do a real scientific study but I'm inclined to
> continue to correct to avoid blinking highlights. I have realized though
> that a second lighter shot is needed as the noise in shadows can be
> horrible. It's a lot of work merging but sometimes necessary.

This brings up a feature I'd really like to see implemented in the next
generation of digital cameras: a generalized bracketing function. I'd like
to be able to spot-meter off several elements in a scene, re-frame and
re-focus the shot, press the shutter once and have the camera take all the
queued-up exposures sequentially. Sometimes it is absolutely necessary to
combine shots to get the picture you want and such a function would make my
wde-dynamic-range landscape photography a lot less tedious. Even better,
since this feature requires changes only to the camera's software, it could
be implemented tomorrow if manufacturers chose to do so.

Funny, even though we're several generations into the digital camera
revolution, many people still think solely in terms of the film paradigm.
The era of Photoshop-type post-processing is here and digital cameras should
operate accordingly.
January 31, 2005 5:32:29 PM

Archived from groups: rec.photo.digital (More info?)

Paul H. wrote:

> "paul" <paul@not.net> wrote in message
> news:fbydnUSB7KuYCWPcRVn-gg@speakeasy.net...
>
>>JPS@no.komm wrote:
>> >
>>
>>>Unfortunately, most cameras only tell you what gets clipped in
>>>a JPEG, not in the RAW data.
>>
>>You know, I tried overexposing a tad on my D70 yesterday and the blown
>>highlights were indeed blown even in RAW. I don't know, maybe a slight
>>bit of leeway, I didn't do a real scientific study but I'm inclined to
>>continue to correct to avoid blinking highlights. I have realized though
>>that a second lighter shot is needed as the noise in shadows can be
>>horrible. It's a lot of work merging but sometimes necessary.
>
>
> This brings up a feature I'd really like to see implemented in the next
> generation of digital cameras: a generalized bracketing function. I'd like
> to be able to spot-meter off several elements in a scene, re-frame and
> re-focus the shot, press the shutter once and have the camera take all the
> queued-up exposures sequentially.


That'd be nice.

I set my D70 so the half pressed shutter locks focus only and the
exposure lock button is needed to lock exposure. Lots of reframing &
button pushing needed to get what you describe though it's possible.

Another thing someone discussed is the (possibility?) to set the command
wheels so that in A mode, the large wheel adjusts aperture & the small
wheel adjusts exposure compensation, then in S mode the large wheel
adjusts speed and the small wheel is still exposure compensation.

Actually in manual mode, if you watch the meter in the viewfinder this
gives the same effect and the auto A/S modes can be discarded while
still using the camera's metering. Just adjust both dials manually for
the +/- desired all while looking through the viewfinder. You know, this
works very nicely: set the aperture with thumb, set the speed in front
with finger to sero out the meter then spin up & or down as desired and
your finger is right next to the shutter. I think this is the way to go.
Exposure compensation button is a pain, you have to hold two buttons at
once to adjust that way.

I really do miss the LCD preview on my P&S though. It was so easy to
judge exposure intuitively! While we are talking dream modes, I'd like
to turn on the EVF with a click to preview exposure complete with
blinking highlights after composing with the optical viewfinder <grin>.



> Sometimes it is absolutely necessary to
> combine shots to get the picture you want and such a function would make my
> wde-dynamic-range landscape photography a lot less tedious. Even better,
> since this feature requires changes only to the camera's software, it could
> be implemented tomorrow if manufacturers chose to do so.
>
> Funny, even though we're several generations into the digital camera
> revolution, many people still think solely in terms of the film paradigm.
> The era of Photoshop-type post-processing is here and digital cameras should
> operate accordingly.
Anonymous
February 1, 2005 2:10:17 AM

Archived from groups: rec.photo.digital (More info?)

In message <ZoOdnYMME_NtP2PcRVn-1w@comcast.com>,
"Paul H." <xxpaulhtck@zzcomcast.yycom> wrote:

>Funny, even though we're several generations into the digital camera
>revolution, many people still think solely in terms of the film paradigm.
>The era of Photoshop-type post-processing is here and digital cameras should
>operate accordingly.

The software is still in the dark ages, though. There is very little
vision in this field.

You should be able to take two RAW shots 4 stops apart, and graft them
together as a single 16-bit RAW, fading the data from one to the other
in the overlapping 8-bit range. You could do a lot more with this than
you can with blending two converted exposures together, which kills the
contrast in the non-overlapping ranges, and keeps useless artifacts.

--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
February 1, 2005 2:10:18 AM

Archived from groups: rec.photo.digital (More info?)

JPS@no.komm wrote:

> In message <ZoOdnYMME_NtP2PcRVn-1w@comcast.com>,
> "Paul H." <xxpaulhtck@zzcomcast.yycom> wrote:
>
>
>>Funny, even though we're several generations into the digital camera
>>revolution, many people still think solely in terms of the film paradigm.
>>The era of Photoshop-type post-processing is here and digital cameras should
>>operate accordingly.
>
>
> The software is still in the dark ages, though. There is very little
> vision in this field.
>
> You should be able to take two RAW shots 4 stops apart, and graft them
> together as a single 16-bit RAW, fading the data from one to the other
> in the overlapping 8-bit range. You could do a lot more with this than
> you can with blending two converted exposures together, which kills the
> contrast in the non-overlapping ranges, and keeps useless artifacts.


I've had the best results simply pasting the two exposures on layers and
manually erasing out the offending exposure area with a fuzzy eraser in
PS. Sometimes it is easy and sometimes it's a bear. Sometimes a magic
wand selection can be feathered to delete with. There was a blending
program posted recently and the results were very unnatural looking but
maybe they were pushing it too far. I've pushed it too far myself and it
can definitely look unnatural.
Anonymous
February 1, 2005 5:21:01 AM

Archived from groups: rec.photo.digital (More info?)

In message <auqdnXfXh7yWVWPcRVn-sw@speakeasy.net>,
paul <paul@not.net> wrote:

>I've had the best results simply pasting the two exposures on layers and
>manually erasing out the offending exposure area with a fuzzy eraser in
>PS. Sometimes it is easy and sometimes it's a bear. Sometimes a magic
>wand selection can be feathered to delete with. There was a blending
>program posted recently and the results were very unnatural looking but
>maybe they were pushing it too far. I've pushed it too far myself and it
>can definitely look unnatural.

What is needed is better curve-control tools. The ones in photoshop are
awkward, IMO. There is nothing intuitive about the response of the
Curves tool, and adjustment is extremely coarse.

What I would need, when trying to put a high-dynamic-range capture into
lesser space, would be moveable markers for "zones" underneath a 0-255
scale, like so:



| | | | | |
0 51 102 153 204 255

|| | | | | | | | | | |||


Where the right-most marker on the bottom line represents maximum
luminance, and each line going to the left represents a stop, or
half-stop. You could lock a range together so that they move in unison.
I think you could get much more natural curves this way. Also, this
could just represent a low-pass transfer curve, and you could have an
adjustment for each marker that indicates a cut or boost in
high-frequency contrast at that point (like the Shadow/Highlight tool of
Photoshop, but for multiple moving bands).
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
February 1, 2005 2:57:43 PM

Archived from groups: rec.photo.digital (More info?)

<JPS@no.komm> wrote:
>
> What is needed is better curve-control tools. The ones in photoshop are
> awkward, IMO. There is nothing intuitive about the response of the
> Curves tool, and adjustment is extremely coarse.

You might try Picture Window Pro's Brightness Curve tool. It shows both
input and output histograms and allows you to define your own control
points.

http://www.normankoren.com/PWP_intro.html

http://www.dl-c.com/Temp/

David J. Littleboy
Tokyo, Japan
Anonymous
February 1, 2005 2:57:44 PM

Archived from groups: rec.photo.digital (More info?)

In message <ctmr7i$cli$1@nnrp.gol.com>,
"David J. Littleboy" <davidjl@gol.com> wrote:

><JPS@no.komm> wrote:

>> What is needed is better curve-control tools. The ones in photoshop are
>> awkward, IMO. There is nothing intuitive about the response of the
>> Curves tool, and adjustment is extremely coarse.

>You might try Picture Window Pro's Brightness Curve tool. It shows both
>input and output histograms and allows you to define your own control
>points.

Thanks, but I'd have to buy it, unless a new version is out since I
tried it last.

These demos with 15-day time-outs are really annoying; you try them once
for 5 minutes, and the next time you go to look at them, they have
expired.

Maybe it's never been on my laptop.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
February 2, 2005 4:11:52 AM

Archived from groups: rec.photo.digital (More info?)

<JPS@no.komm> wrote in message
news:58otv0d2s693inbsklnlvi8qogblbavi4b@4ax.com...
> In message <auqdnXfXh7yWVWPcRVn-sw@speakeasy.net>,
SNIP
> What is needed is better curve-control tools. The ones in
> photoshop are awkward, IMO. There is nothing intuitive about
> the response of the Curves tool, and adjustment is extremely
> coarse.
>
> What I would need, when trying to put a high-dynamic-range
> capture into lesser space, would be moveable markers for
> "zones" underneath a 0-255 scale, like so:

As mentioned by David J. Littleboy, Picture Windows Pro allows to do
some of that. However, more intelligent/automatic tonemapping
approaches have been researched, e.g. by Greg Ward
(http://radsite.lbl.gov/radiance/papers/lbnl39882/tonema...). His
histogram adjustment, in its basic implementation, is quite easy to
implement if Raw data capture is available. Lots of other interesting
stuff by Greg Ward can be found at:
http://www.anyhere.com/gward/papers.html .
Another useful study is found at:
http://www.cs.wright.edu/people/faculty/agoshtas/hdr.ht... .

Bart
Anonymous
February 2, 2005 4:26:56 AM

Archived from groups: rec.photo.digital (More info?)

<JPS@no.komm> wrote in message
news:kgetv09ekfced0nnhhampjqu4i9g26oeuj@4ax.com...
> In message <ZoOdnYMME_NtP2PcRVn-1w@comcast.com>,
SNIP
> The software is still in the dark ages, though. There is very
> little vision in this field.

The vision is there, it's just not widespead in the photographic
community yet.

> You should be able to take two RAW shots 4 stops apart,
> and graft them together as a single 16-bit RAW, fading the
> data from one to the other in the overlapping 8-bit range.
> You could do a lot more with this than you can with blending
> two converted exposures together, which kills the contrast in
> the non-overlapping ranges, and keeps useless artifacts.

I agree, HDR imaging is catching on *very* slowly, despite "older"
research in that direction for either stationary subjects:
http://www.cg.tuwien.ac.at/research/theses/matkovic/
http://www.cs.huji.ac.il/~danix/hdr/results.html
http://www.cs.huji.ac.il/~danix/hdr/pages/belgium.html
or for post-processing tonemapping in (preferably) 16bit/channel
images:
http://www.cs.huji.ac.il/~danix/hdr/enhancement.html (esp. the swan
pictures).

You can use a program like <http://www.ict.usc.edu/graphics/HDRShop/&gt;
(version 1 is a free download) to create an HDR image, and a plugin to
import it in an editor like Photoshop, but that would still only allow
to select a luminance range from the total HDR scene. There is an
HDRshop specifc plugin available for tonemapping available at:
http://www.gregdowning.com/HDRI/tonemap/Reinhard/ (I don't
particularly like the highlight mapping though). Also,
http://www.openexr.com/ may turn out a more useful general purpose
file format (not necessarily because it is better, but it may gain
wider acceptance). EXR file creation/output would need e.g. HDRshop
Version 2, but I find the US$ 650 single user licence fee too
prohibitive for a trial, while still needing a tonemapping function
for displayable (as 8-bit/channel) output.

Issues related to bracketed exposures are covered in:
http://www.anyhere.com/gward/papers/jgtpap2.pdf .

Bart
!