RE: Exposure Value
RE: Exposure Value
- Subject: RE: Exposure Value
- From: Roger Breton via colorsync-users <email@hidden>
- Date: Fri, 21 Jun 2019 23:14:59 -0400
Dear Wayne,
I plan on going back to very basic "stuff" tomorrow, among others, like
shooting a plain vanilla ColorChecker 24 chart. This type of target ought to
easily reveal any differences between the RAW file, as developed with
CameraRAW, and the JPEG "in-camera" rendering. It's going to be very easy to
compare both processed pixels values.
This is still a mystery to me.
I probably did something silly on my camera ...
Best / Roger
-----Original Message-----
From: Wayne Bretl <email@hidden>
Sent: Friday, June 21, 2019 11:10 PM
To: email@hidden; 'Iliah Borg' <email@hidden>;
email@hidden
Subject: RE: Exposure Value
Dear Roger,
Without further evidence, it is still a mystery why your original raw-file-only
mode pictures show apparent underexposure, and the ones taken after choosing
raw + jpeg do not. Is it possible you changed another mode setting, like going
from Program to Full Automatic, or some other setting (matrix, spot, center
average) that changed the exposure metering mode in your camera? Some modes may
center the dominant exposure sensing on the in-focus point, for example. Some
cameras also have an automatic tone curve adjustment (possibly enabled or
disabled according to the exposure mode and/or in a menu somewhere) that
attempts to recognize and compensate for various types of images. This is an
opportunity for the camera to do things you don't like, as well as possibly
rescuing some shots for the unknowledgeable person using full automatic mode.
Another thing to note is that while the data in the raw file is supposed to be
a mostly untouched rendering of the light falling on the sensor, the rendering
output of the raw processor definitely is not so simple. The rendering is not
even simply an application of a gamma curve to match the standard jpg 1/2.2
power. That sort of simple process was used in very early digital cameras
adapted from video chips, such as the Sony Mavica, and was universally disliked
compared to film. Now, raw processors always apply an S curve, mimicking film,
because this provides the best looking result: an increase in contrast (at
least for the mid tones) to compensate color appearance effects of the human
visual system, and a gradual toe and shoulder to accommodate highlights and
shadows without severe clipping. This is one example of the kind of
behind-the-curtain processes, the details of which are buried in the camera
profile and never revealed to the user by Adobe camera raw.
Regards,
Wayne
-----Original Message-----
From: colorsync-users
<colorsync-users-bounces+waynebretl=email@hidden> On Behalf Of Roger
Breton via colorsync-users
Sent: Friday, June 21, 2019 7:15 PM
To: 'Iliah Borg' <email@hidden>
Cc: email@hidden
Subject: RE: Exposure Value
Dear Ilah,
Thank you so much for applying your vast technical expertise to my humble
problem. I realize, again, that there is so much to learn in digital
photography, I feel ashamed to have neglected digging into this fascinating
part of image processing for so long (It took me more than a few CPU cycles of
my brain to figure Exposure Value, today)...
For sure, I need to invest in a "new" gray card, I confess that this one has
seen its share of "tales from the trenches".
Having said that, I'm relieved to read in your analysis that my camera "light
meter" isn't playing tricks on me or need to go back to the manufacturer under
warranty.
I *will* reshoot, for sure, as you suggested. It's an exercise worth every
effort and probably investigate using one of your suggested software..
For now, I am not sure what to conclude from my experience? Tomorrow, I will
probably experiment with shooting A) RAW alone and shooting B) RAW + JPEG. In
"theory", there should not be any difference between the two modes of shooting
but if there should be one, then this is going to prove interesting.
Have you had a chance to take a look at my Excel sheet at all? To see whether I
correctly applied the metering equations from measured Luminance and
Illuminance?
I think the calculations are sound.
One last thing I have not tried and I *will* try is, to shoot with my antique
D100, in RAW, to compare with shooting the same scene with the D810, in RAW, to
test any potential differences between the two systems, in terms of exposure.
One thing I wonder, as I was thinking about the experimental setup, something
completely absurd (?), was, what that, could there be any difference in the
underlying capture of the pixels between the two cameras?
Suppose, on the D810, at 1/200s, F/5.6, ISO400, the histogram looks skewed to
the left, suggesting some degree of under-exposure (see my earlier links)?
Suppose further, on the D100, at the same 1/200s, F/5.6, ISO400 settings, the
histogram looks "normal", with pixels distributed throughout the tonal range,
from 0 to 255, then what?
Until I'm convinced that there is nothing wrong with this camera, I figure I
need to further test it, until I'm fully satisfied that there is nothing
"wrong" with it.
Best regards / Roger
-----Original Message-----
From: Iliah Borg <email@hidden>
Sent: Friday, June 21, 2019 6:33 PM
To: email@hidden
Cc: email@hidden
Subject: Re: Exposure Value
Dear Roger,
Based on your gray card NEF:
You are in 14-bit mode, maximum is 2^14-1-black level (600) = 15873; average on
the card (and the card seems to have some rather strong imperfections) is 1181
(you can check with RawDigger, trial version is enough). That puts the exposure
at log2(1181 / 15871) = 3.75 EV below clipping; the camera is calibrated to
produce midtone 3.35 EV below clipping, that means the exposure is 0.4 EV below
the midtone.
Nothing dramatic so far, provided that the card has a lot of wear and tear on
it, it isn't shot out of focus, and not at an angle as it should be (see
http://www.zonephoto.it/images/pdf/kodak-grey-card1903061.pdf ). The result is
mathematically inconclusive - but from my experience I would be expecting
something like the numbers above, given how the shot was taken, and would say
that metering is calibrated OK.
Of course, if you could re-shoot we will be able to come to a more definite
conclusion.
On Jun 21, 2019, at 6:03 PM, Roger Breton via colorsync-users wrote:
> I will get to the Gray Card NEF in a few minutes, Ilah, but I thought I would
> try turning on Image quality = NEF + JPEG in the camera shooting menu?
>
> And I got very different results?
> See this link:
> https://1drv.ms/u/s!AkD78CVR1NBqkLVK68wxOJ6dhvv2nQ
>
> On the left hand side is the JPEG opened in Photoshop as usual, and on the
> right hand side is the RAW, opened in CameraRAW. The only thing that I could
> say is the Histograms are very different, this time, extending well into the
> highlights...
>
> Thanks for you guys patience and help.
> I'm trying to contact Nikon tech support...
>
> Best / Roger
>
>
> -----Original Message-----
> From: colorsync-users
> <colorsync-users-bounces+graxx=email@hidden> On Behalf Of
> Roger Breton via colorsync-users
> Sent: Friday, June 21, 2019 5:44 PM
> To: 'Wayne Bretl' <email@hidden>; email@hidden
> Subject: RE: Exposure Value
>
> I just tried with ViewNX2 and it is still too "dark"?
> So it is not coming from CameraRAW -- good news 😊
>
> ViewNX2 link is here:
> https://1drv.ms/u/s!AkD78CVR1NBqkLVIL-cOCJBYIX0LuA
>
> CameraRAW link is here:
> https://1drv.ms/u/s!AkD78CVR1NBqkLVJgBzv-yK7ygFD7g
>
> The histograms are not lying, this is under-exposed.
>
> Best / Roger
>
>
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
> This email sent to email@hidden
>
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> colorsync-users mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
> This email sent to email@hidden
--
Best regards,
Iliah Borg
LibRaw, LLC
www.libraw.org
www.rawdigger.com
www.fastrawviewer.com
_______________________________________________
Do not post admin requests to the list. They will be ignored.
colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden