Finally, a D300 Near UniWB JPEG

Started Apr 3, 2008 | Discussions
gollywop
gollywop Veteran Member • Posts: 8,284
I also have NeutralFlat and Neg2.2Gamma Curves

These are easily made in Picture Control Utililty. The Negative 2.2 Gamma curve does give great information about the highlights, but it also creates an LCD image that is typically too dark to be useful. It's too bad one can't have separate curves for the histograms and the LCDs.

-- hide signature --

gollywop

-----------

Iliah Borg Forum Pro • Posts: 25,841
Re: I also have NeutralFlat and Neg2.2Gamma Curves

Maybe it is even simpler - too bad one can't have just raw histogram (linear gamma, nothing like WB or other controls applied, eV marks on the histogram) and normal thumbnails with all picture controls applied. One day.

-- hide signature --

no text

gollywop
gollywop Veteran Member • Posts: 8,284
Amen (nt)
-- hide signature --

gollywop

-----------

lovEU Veteran Member • Posts: 3,135
Why so less interchange of ideas?

Why is it such difficult to get all parties to one place? No real conference organized by some of the big players like IEC, ICC or so? That would be an interesting event, I guess - RAW developers like you, discussing with color scientists (like Scott) and tecs from the major camera producers. There seems to exist some need for it. Btw, I'm sure there are many conferences etc. But obviously none of them is designed especially to give impact for some very basic but important questions concerning R&D of digital imaging - strange, isn't it? Maybe one should suggest to the organizers of the Photokina to provide a forum for such discussions (scientists, developers and other pros exclusively)?

Iliah Borg wrote:

Maybe it is even simpler - too bad one can't have just raw histogram
(linear gamma, nothing like WB or other controls applied, eV marks on
the histogram) and normal thumbnails with all picture controls
applied. One day.

Scott Geffert wrote:

Does anyone from Nikon of Canon read these forums? It would be great
to get this type of discussion to the engineers.

[ http://forums.dpreview.com/forums/read.asp?forum=1021&message=27496312]
--
regards, eric

Mark den Hartog Veteran Member • Posts: 3,679
and now make it print correct

you will highly unlikely get this to print correct.
you will have (minor) color shifts..

or am I missing something?

gollywop
gollywop Veteran Member • Posts: 8,284
Not clear what your comment means.

The "this" in your comment is a rather ambiguous reference. Do you mean the use of the Near UniWB? or do you mean the greenish tint? or what?

In any event, it does indeed appear that you are missing something. When shooting RAW, the WB as set in the camera has no effect on the RAW data, but it does affect the histogram information. The Near UniWB, then, allows one better to assess exposure, and then the proper WB (established, say, with a gray card or Expodisc) is set when converting the RAW data. It is that latter WB that determines what happens with the printer -- assuming proper color management throughout the rest of the process.

-- hide signature --

gollywop

-----------

Raul Veteran Member • Posts: 8,439
reverse tonal transform

how is such a curve prepared and how loaded?

regards

GaborSch wrote:

Iliah Borg wrote:

Second biggest issue after white balance is tonal mapping to gamma
2.2 and this can be avoided too loading reverse tonal transform as a
custom curve thus ending up with a linear histogram. If it is done,
the statement below is true

The mapping changes the distribution on the scale,

And prevents correct evaluation of the real mass of the affected
highlights and shadows. That is, it is really important to know the
distribution of the tones across the scale.

 Raul's gear list:Raul's gear list
Nikon Df Nikon D5 Nikon D850
petroag Forum Member • Posts: 54
Re: Not clear what your comment means.

If you use uniwb, how can you use the expodisc to correct for wb in post processing?

gollywop
gollywop Veteran Member • Posts: 8,284
Re: Not clear what your comment means.

Well, I've never really used Expodisc, but I suppose, in principle, you could use the Expodisc to set WB, and then take a test shot. Then switch back to UniWB for the real shots. You can then use the "As Shot" WB information from the test shot to set WB in PP for the others -- at least, you could do this with ACR; I don't know how one might proceed with NX.

-- hide signature --

gollywop

-----------

gollywop
gollywop Veteran Member • Posts: 8,284
Re: reverse tonal transform

Raul:

Open Picture Control Utility and start with a Neutral (predefined) case. Drop the Sharpening to 0 -- keep everything else at 0 --, and click "Use Custom Curve." Now pull the center of the curve down with a single point to a position as close as you can get it to Input 128, Output 56. Now click "New," and name your custom curve. Finally, export it to your CF card. It will be put in a folder named NIKON, subfolder CUSTOMPC, file PICCONxx.NCP. Then follow the instructions in the manual for loading a custom curve.

But beware, the resulting tone curve will produce an LCD image that is very, very dark, and perhaps, as a result, pretty useless. Still, one can use it to take a test shot just to see what the histograms look like under your shooting conditions.

-- hide signature --

gollywop

-----------

GaborSch Veteran Member • Posts: 7,203
There is no reason to switch back and forth

if you have a shot of anything white or grey, you can pick WB on that shot during raw processing and carry that on the other shots.

It depends on the actual raw process how this can be done. For example ACR displays the temperature and tint value after having picked WB, and you can enter those values for other images. Alternatively, you can select all images and when you pick WB on that white spot, that setting will be applied to all selected images.

gollywop
gollywop Veteran Member • Posts: 8,284
Re: There is no reason to switch back and forth

Hi Gabor. Quite right, but I never said you had to switch back and forth -- just that you take one test shot and then apply its WB to the others, just as you suggest. You would, of course, have to take new test shots for each change in lighting conditions.

-- hide signature --

gollywop

-----------

GaborSch Veteran Member • Posts: 7,203
Misunderstanding

gollywop wrote:

Hi Gabor. Quite right, but I never said you had to switch back and
forth -- just that you take one test shot and then apply its WB to
the others, just as you suggest

You posted above

use the Expodisc to set WB, and then take a test shot. Then switch back to UniWB for the real shots

What I said that the Expodisk (or anything else for WB picking) shot too can be done with UniWB.

gollywop
gollywop Veteran Member • Posts: 8,284
Ah, yes.

Quite right; gray is gray. It's my inexperience with the Expodisc that set me off. I've, of course, no trouble shooting a gray card with the UniWB, and, yes, it must be the same with the Expodisc.

Thanks for keeping me straight. And thanks also, by the way for your input into these matters. I've found them quite helpful.

-- hide signature --

gollywop

-----------

gollywop
gollywop Veteran Member • Posts: 8,284
Second Thoughts

Gabor, now you've got me confused. The post to which I was answering asked how to use the Expodisc with the UniWB. And I think I acquiesced too easily to your post that switching was not necessarity.

As I understand it, the Expodisc sets the WB in the camera before the shot is taken. This, of course, is quite different from using a gray card, which provides one with the information needed to set the WB after the shot is taken.

Thus, if one uses the Expodisc, the camera's WB is set, and one cannot use the UniWB. And this was the issue that petroag was raising and to which I was responding.

With a gray card, of course, one can shoot the gray card with the UniWB and then set the proper WB later. But with the Expodisc, to use UniWB, one must first shoot a scene with the Expodisc-determined WB, and then switch to UniWB for the shots of interest. The shot taken with the Expodisc WB can then be used in PP to set the WB of the UniWB shots. But a switch must be made -- not a switch "back and forth," but at least one switch per change in lighting conditions. Yes?

-- hide signature --

gollywop

-----------

GaborSch Veteran Member • Posts: 7,203
How Expodisc works

Honestly, I mixed up Expodisc with Whitedisk, namey there is such a thing as well, used like a grey card; I was thinking of that.

However, now I read about Expodisc, and I don't see any difference in their usage from this perspective . With Whitedisk you capture the reflected light, with Expodisk the emitted light. The difference is in which direction you make the shot, but the result is a gray shot; with the Expodisk, the entire image is supposed to be grey.

So, there is no difference: you record the raw image through the Expodisk; the actual WB setting is irrelevant. Later you can use that greyish image for picking WB.

At least this is my understanding after having read a few reviews of the Expodisk.

petroag Forum Member • Posts: 54
Re: How Expodisc works

When using uniwb the photo does appear green. I would think this is a result of the bayer filter pattern. 50% more green so you see green in the exposure. Now if I wanted to expose the red and blue channels to the right without blowing out the green channel, wouldn't I want to use a green filter such as CC30M? I then could get all channels exposed to the right more efficiently and then correct for WB in post processing.

gollywop
gollywop Veteran Member • Posts: 8,284
Re: How Expodisc works

Ah. Well, as I read it, you use the Expodisc manually to set the camera's custom WB; that is, you would use it to set d-0 in the D300, which you would then use in your shoot. Of course, if you use this d-0 WB, you cannot use UniWB. This is quite different from shooting a gray card using the UniWB and then re-establishing that WB in PP.

So, again to petroag, I think you'd have to set the camera's WB with the Expodisc, shoot a test shot with that WB, switch to UniWB for your real shoot, and then copy the WB of the test shot to the others in PP.

-- hide signature --

gollywop

-----------

gollywop
gollywop Veteran Member • Posts: 8,284
Re: How Expodisc works

No, petroag. The greenish tint of the UniWB is due to the fact that it is compensating for the camera's processor's conversion of the camera's native white (which is not white) to a balanced white. As a result, the camera's histograms show this altered information rather than the RAW data's actual information, which is what is seen when using the UniWB. See the discussion between me and Iliah Borg at

http://forums.dpreview.com/forums/read.asp?forum=1039&message=27408206

for more detail.

-- hide signature --

gollywop

-----------

GaborSch Veteran Member • Posts: 7,203
Filter for DR

petroag wrote:

When using uniwb the photo does appear green. I would think this is
a result of the bayer filter pattern. 50% more green so you see
green in the exposure

No, it has nothing to do with that. You don't see raw pixels (except with Rawnalyze). If a white object appears greenish with UniWB, that indicates, that the captured light contains more green than other wavelengths. This has nothing to do with the number of pixels, but with

1. the ilumination,

2. the microfilter over the pixel sites.

Now if I wanted to expose the red and blue
channels to the right without blowing out the green channel, wouldn't
I want to use a green filter such as CC30M? I then could get all
channels exposed to the right more efficiently and then correct for
WB in post processing.

CC30M is a magenta filter. That would be the right one, though the 30 may be too strong. If you analyze shots with Rawnalyze, you can see, how much stronger the green comes in different settings.

I already ordered a B+W CC20M, and it turned out, that it has been taken out of production. Though I had some concerns anyway: these filters are not meant for digital photography, they are not coated like really good UV or polar filters are. Digital is much more sensitive to internal reflections and stray light than film.

Someone had another idea, useful only indoor: put a "filter" (it can be even cellophane) at the front of your light source. But that is not necessarily magenta, it depends on the light source; incandescent is typically very warm, so you would want to filter out the red. A cyan-like "filter" would be useful in such case.

Keyboard shortcuts:
FForum MMy threads