Pages: 1 [2]   Go Down

Author Topic: Another camera profiling issue (DNGPE)  (Read 11644 times)

Peter_DL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 544
Re: Another camera profiling issue (DNGPE)
« Reply #20 on: March 15, 2015, 07:44:51 am »

In the first statement I read "replaced by" to mean replaced, not modified by application of the Chart Wizard.
The second statement seems to contradict (or perhaps clarify) this; indicating the effect of the edit is to modify the color table.

…pending question,
yet I did not find a good way to clearly prove this in either direction.

--
Logged

Tim Lookingbill

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2436
Re: Another camera profiling issue (DNGPE)
« Reply #21 on: March 15, 2015, 04:56:46 pm »

…pending question,
yet I did not find a good way to clearly prove this in either direction.

--

Can anyone show some A/B examples of the effects of these tables on real world images (not on color patches) in order to connect what's discussed here with something that's understandable and useful?
Logged

Peter_DL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 544
Re: Another camera profiling issue (DNGPE)
« Reply #22 on: March 15, 2015, 06:29:04 pm »


Sorry Tim, - here is what I think about the substance and the usefulness of your posts:

Logged

Peter_DL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 544
Re: Another camera profiling issue (DNGPE)
« Reply #23 on: March 16, 2015, 04:48:27 pm »

In the first statement I read "replaced by" to mean replaced, not modified by application of the Chart Wizard.
The second statement seems to contradict (or perhaps clarify) this; indicating the effect of the edit is to modify the color table.

In the meantime I did some tests by creating a 'false profile' based on the Adobe Standard profile but without LookTable and in particular with some 'undesired' hue shifts in the HueSatDelta table.  Upon using this false profile as the Base Profile to run the Chart Wizard, the hue/sat.-adjustments shown in the DNGPE Color Tab are quite strong – which means that the given HueSatDelta table with its 'undesired' hue shifts is taken into account and that the Chart Wizard tries to correct it. In other words, the HueSatDelta table as given with the Base Profile is not replaced in a way that its previous content would be discarded and ignored, actually, it lays the ground for the hue/sat.-adjustments done by the Chart Wizard.

Also, the final custom profile is not identical to a reference profile which was created the same way but without the 'undesired' hue shifts in between. The 'undesired' hue shifts were done with "real world colors" from a normal photo (DNG file) and not exactly with the ColorChecker colors, so that the 'undesired' hue shifts effect the ColorChecker colors only partially, and vice versa with the hue/sat.-adjustments by the Chart Wizard. In other words, even after running the Chart Wizard there is residual content left in the HueSatDelta table from the previously inserted 'undesired' hue shifts.

Aside from the insight as such, this may also explain Eric's warning, not to >> apply the Chart Wizard iteratively <<.
It's better to start from scratch again with the Adobe Standard profile (to avoid any residual content in the HueSatDelta table which may not get perfectly overwritten by a second session and run of the Chart Wizard).

That said, another good advice was already given to the OP
(at least I agree to it):

I am less sure about the usefulness of DNG camera profiles these days. … I feel that all the talk about cameras Adobe uses for profiling being vastly different from the ones we use, hence necessitating custom profiles is a more or less untrue. … That was a rabbit hole I went down into for too long.

In case you don't like the default Adobe Standard profile and the Camera Matching profiles,
just run the Chart Wizard, then to go to Edit > Clear Color Adjustments, and export the resulting custom profile
(let's call it "core profile").

This core profile is independent from the quality of the ColorChecker capture, i.e. the light source, the exposure level of the chart image, the evenness of illumination, and finally all the nice variables which otherwise can influence results from custom profiling.

In case you still want (or need) to custom-profile,
this core profile provides at least another good reference regarding what you really gain.

Peter

--
« Last Edit: March 16, 2015, 04:51:46 pm by Peter_DL »
Logged
Pages: 1 [2]   Go Up