Converting Lightroom Presets to Profiles – Issue

[Jonathan Posner]

Lightroom Version Number

9.1 (Camera Raw 12.1)

Operating System

1. macOS 10.13 High Sierra

Have just come across a major issue here and so wonder if anybody has any ideas or suggestions (though I suspect this needs to be dealt with by Adobe. Julianne . . .?)

First, the context:

I happily use Profiles, especially the excellent Lutify.me packs, because they're fast and easy to apply as well as being strength-adjustable. So they've largely taken over from my favourite VSCO or RNI presets (also adjustable with Opal, but much more fiddly). However, I'm still attached to several of my VSCO Film presets and have been thrilled to discover that they can be very quickly and easily converted into Profiles by doing it in Photoshop's ACR. So far so good.

Second, the problem:

Both the Lutify profiles (in fact all profiles), AND all the VSCO presets have been correctly designed to exclude Exposure and Colour Temperature. This is of course vital for a meaningful workflow for someone like me who prefers to get the basic image corrections done before applying and comparing various preset or profile effects using Virtual Copies.
However, now I discover that the ACR conversion engine transfers ALL the Basic panel settings so therefore one would have to go back to re-adjust esposure and colour balance each time because the Profile you've created is technically only right for the base image that you'd chosen to do the conversion with in the first place.

I think this is a terrible thing that needs fixing. I suppose one could do a workaround with CT by making every Profile with a Daylight balance, for example, but the Exposure settings are going to vary between image to image and as I say I think it's a poor intrusion into the Lightroom workflow because one will always have to edit Exposure twice. Actually, I'm astonished that Adobe doesn't seem to have thought this through properly by simply modifying the tickbox panel.

Maybe I'm missing something here, so ideas and advice welcome!

 

[Victoria Bampton]

Hi Jonathan. I can't stop and test this right now, but the resulting profiles are just xmp, so I suspect you can simply open them in a text editor and remove any sliders you want to exclude. If you get to test it before I do, let me know if that does the trick!

 

[Jonathan Posner]

Hi Victoria, thanks for commenting. I think the xmp route is a little beyond (even!) me so if and when you have time would be grateful for any further help.

Meanwhile, I feel even more out of my depth in terms of understanding what might be going on.

As part of my original testing I deliberately ramped down the colour temperature of an image:

ONE: Original image
TWO: Original Temperature readout
THREE: Original image, 'Tungsten'
FOUR: 'Tungsten' Temperature readout
FIVE: New Profile created from 'Tungsten' image
***FIVE: But now in Basic panel, new Temperature readout has 'reverted' to original!

I'm sure similar baffling anomalies would show up in the Exposure department but by now I've lost the will to live

 

[Victoria Bampton]

I'm not quite following your testing . What were you trying to prove or disprove there?

 

[Jonathan Posner]

I don't understand why, if colour temperature is a setting that is included when making the profile, it hasn't transferred over to any new images with the profile applied (image SIX). It's added yet another variable into the process and it's very confusing. Once again, it makes it so difficult to make a 'reliable' profile where I don't have to fiddle with each new image after applying it – surely that was the point of profiles in the first place?

 

[Victoria Bampton]

Email me the profile Jonathan, let me take a look at the contents. That’ll save some guesswork.

 

[Jonathan Posner]

Thanks for taking a look, Victoria. If it's simply a matter of deleting some code (as you suggested at the beginning) I'd be grateful if you could show me how to do it. Specifically, Colour Temp + Tint, and Exposure (assuming that's all.) I could then hopefully delete these lines for every subsequent preset I convert.

 

[Paul McFarlane]

Hi Jonathan

Two observations:

1. There are no lines for color temp, tint or exposure in the profile, so they won't be affected when you apply them (as an example, for Exposure there would be an entry like 'Exposure2012="0.25"'
2. These are the old lrtemplate files not the converted xmp files?

 

[Jonathan Posner]

Hi Paul,

Thank you for these observations, much appreciated. Re your numbered points:

1. Yet this seems to be completely contrary to my (tested) experience. If you again look at my Image FIVE, you'll notice that the exagerated blue temperature shift that I'd applied to the original image from which I then made the Profile has been carried over into the new profile (ie so if I now apply this profile to subsequent images they too are "over-blue".) And I'd also made the point that (Image SIX) the temperature readout of the newly-profiled image has reverted to the normal temperature of the captured image, which is obviously nonsense in this case. Adobe can't have it both ways!

2. I'd only included the original lrt file as a reference for how VSCO had created the original preset, in case that was helpful; I can confirm that the active preset from which I made the Profile is indeed an xmp file.

I'm so sorry if all this looks like dancing on the head of a pin but I wouldn't be going on about it if it weren't important to me. Obviously, Photoshop needs an image in order to enact the preset conversion but the image itself shouldn't have any bearing on the outcome for ever more. My contention (and already partly demonstrated) is that if I were to use 10 different images with the same preset applied for the conversion I'd end up with 10 different profiles. Lightroom is, if nothing else, based on a system of measureable adjustments that can be consistently applied so I'm at a loss as to why this particular process is leading to such vague outcomes. Preset designers had sorted this out more than a decade ago when I was using Lightroom 4!

 

[Paul McFarlane]

Hi Jonathan

It's hard to tell you the lines to delete if it isn't a representative xmp file; it's easy (as you may appreciate) to simply delete the color, tint, exposure from the xmp; if you sis an adjustment to HSL then that too can be removed. But you said the color temperature had been applied; maybe I'm missing the point!

Whatever way, we need to see the contents of the actual file in use (xmp)

 

[Jonathan Posner]

Hi Paul,

Thanks again for your prompt response.

I believe I sent the xmp profile in an earlier post. As i'm now getting a bit out of my depth do let me know if you require something else in xmp form (and if so, how to extract it etc.)

 

[Paul McFarlane]

Hi Jonathan

My apologies, I only looked at one (the .lrtemplate one) not the second which is indeed the converted xmp.

However, in the xmp, there are no lines for temperature, tint or exposure so when you apply them they will stay as they are. I also notice the ACR version is being set to 6.7 - as in released in 2012. Do these show as profiles in italics by any chance?

 

[Jonathan Posner]

Hi Paul, no worries (and thank you for your quick response.)

No, the particular one I sent you was not in Italics (though some other of my VSCO and RNI presets are). Your observations about the absence of lines for exposure, temperature and tint are of course perfectly logical (and in line with Victoria's first comments way back). And yet . . . I made another profile conversion from a neighbouring VSCO preset where I overexposed the 'base' image in Photoshop. Sure enough, when I applied the resultant profile to a normally exposed image in Lightroom the image became . . . overexposed. How can this be? (Of course, it 's the same with the blue temperature-shift example I've been quoting up to now.)

Maybe we'll never get to the bottom of it but apart from the maddening and uncharacteristic imprecision I'd so love to be vindicated when Adobe next release an update. I hope they read all your blogs, by the way.

 

[Paul McFarlane]

Did you check if the preset that exhibits this behavior has Exposure or other values in it that adjust that though?

 

[Jonathan Posner]

Yes, it's one I use all the time. But have just double-checked. If you look at my last screenshot I made two Fuji Provia 100F++, both from the same image. The first at normal exposure (suffix-2) and the second deliberately overexposed. The Profiles correspond. (And for the avoidance of doubt, all the sliders in the below examples are in exactly the same positions as on the original edited image before applying the profiles.)

 

[Victoria Bampton]

Sure enough, when I applied the resultant profile to a normally exposed image in Lightroom the image became . . . overexposed. How can this be? (Of course, it 's the same with the blue temperature-shift example I've been quoting up to now.)

I might be missing the point completely, but... tone curve? Have you opened up the original preset to see what it changes?

 

[Bernard]

I might be missing the point completely, but... tone curve? Have you opened up the original preset to see what it changes?

I think you are completely right :
ParametricHighlights="+25" ,.......etc
and furthermore HSL : Hue, saturation and luminance are changed for almost all colors.
Bernard

 

[Jonathan Posner]

Thank you everybody for bearing with me and for continuing to help.

I know I'm probably not very bright, but I do still feel that that I simply can't grasp any explanation or hypothesis offered so far.

Yes, of course, along with HSL, there is a minimal adjustment to the tone curve (for the preset in my examples above). But mostly a mild flattening out of the end points. I don't understand why this should effect my profile's exposure if it doesn't affect the exposure (detrimentally) of any image I apply the preset to in Lightroom. As I said, I deliberately bumped up the exposure on my test image before applying the preset so of course my image remained overexposed in Lightroom. The issue, once again, is why has Photoshop delivered me a Profile that overexposes all images that it is applied to. Why does Exposure have anything to do with creating the Profile? As per my four example screenshots above, why hasn't the exposure slider changed between the normally-exposed and overexposed versions?

About to head off down the rabbit hole.