Lightroom/ACR: Do custom camera profiles based on other profiles need to be redone?

  • 1
  • Question
  • Updated 7 years ago
I know if I have a custom profile based on Nikon v3 profiles, they need to be redone for v4, but what about a custom profile based on:

* Adobe Standard
* ACR X.X
* Same version camera matching profile

Do they ever need to be re-done?
R
Photo of Rob Cole

Rob Cole

  • 4831 Posts
  • 384 Reply Likes

Posted 7 years ago

  • 1
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
These don't change as far as I know.

But even v3 versus v4, you don't have to redo the profiles based on them if you like them. Once, a custom profile is done — it's baked, it does not have any kind of dependency on the parent profile.
Photo of Rob Cole

Rob Cole

  • 4831 Posts
  • 384 Reply Likes
Thanks Dorin, but

|> "These don't change as far as I know."

How far do you know? (seriously)

I realize that a custom v3 profile will still work, just like a non-custom v3 profile will still work, I just meant that if you want to have a v4-compatible profile, you have to re-do it. I 'spose the same would hold true for ACR X.X???

But what if Adobe improves the Adobe Standard profile (or fixes a bug for one of my models)? - These are not versioned (are they?), so how would I know if I have to redo the custom profiles based on Adobe Standard? Is the Adobe Standard profile also embedded (baked) into a custom profile?, or is it just a name/reference/link, such that the newest version would automatically be used...
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
Well, just bit-compared two Adobe Standard for D300: one from Lightroom 2.2 and another from 4.1RC2. They matched.
Photo of Rob Cole

Rob Cole

  • 4831 Posts
  • 384 Reply Likes
I'm confused - wasn't Adobe Standard improved between Lr2.2 and 4.1RC2? I thought it was necessary to eliminate the hue shift in highlight recovery. Perhaps next step is to see if Adobe Standard is embedded in custom profiles.
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
No, the profiles have not been changed, the maths had.

I was curious too about the details of the fix. Search the forums high and low, but was only able to find a brief response from and employee that the bug is now fixed (even if screwed people adjustments that were done to compensate for the hue shifts). Nothing in release notes of 3.4 (the version that fixed it) either.

My guess that the order of some operations have been changed, akin to "invariant profiles" made by Sandy-MC's dcp_tool. That is, the hue twists built into the profiles have been moved to a stage before global Exposure, rather than after. I might be wrong, though...
Photo of Rob Cole

Rob Cole

  • 4831 Posts
  • 384 Reply Likes
Well, it actually makes more sense to me that the hue shift was fixed in the software, rather than the profiles, although it seemed the more "common knowledge" was that it was the profiles. Anyway, I think I'll make a *new* batch of custom profiles based on v4 camera matchers, and leave the Adobe Standard ones as is - should be good for another year or three anyway. Thx for all. -R
Photo of Eric Chan

Eric Chan, Camera Raw Engineer

  • 617 Posts
  • 121 Reply Likes
The color processing pipeline has not changed, though a couple of options were added to DNG 1.4 (spec on the way). The so-called hue shift/twist aspect is a deliberate aspect of some of the profiles, and in fact are the ingredient that some (many?) photographers really like -- even though they may not say so in those terms.

For example, we recently had a discussion of hue-preserving tone curves and traditional Photoshop-style RGB curves (which can now be done with the per-channel point curve in PV 2012) ... some users chimed in saying they greatly preferred the Photoshop-style RGB curves. It turns out those curves have hue-shift properties (e.g., concave-up curve portions get pushed towards additive primaries like red, and the concave-down curve parts get pushed towards subtractive primaries like yellow). Different strokes for different folks ...
Photo of Rob Cole

Rob Cole

  • 4831 Posts
  • 384 Reply Likes
Thanks Eric.

I like the twisted profiles. but whatever you did to eliminate the hue shifts in Lightroom 3 Recovery was good.

And of course PV2012 rocks! :-)

Cheers,
Rob
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
Eric, I think we are talking about, perhaps related, but different hue shift/twists.

I guess you mean the hue twists built into DNG profiles, where hue changes with the luminosity (LookTable, if I'm not mistaken). These twists are responsible for the "pleasant color" everyone is liking. Yes, even if they don't understand it.

Rob, on the other hand, is talking about the the undesired hue shifts that used to happen when applying tonal adjustments, mainly Recovery. As I understood it from your explanations at the time, the unwanted shifts were a side effect of having the LookTables with twists in profiles. This issue has been certainly fixed in Lightroom 3.4, as confirmed here by Becky.

So, the question is: "How was this fixed, if the profiles, afaik, have not been modified (e.g. stripped of LookTables)?"