Luminous Landscape Forum

Raw & Post Processing, Printing => Adobe Lightroom Q&A => Topic started by: dreed on July 10, 2011, 04:50:37 pm

Title: Darkening, brightening and information loss
Post by: dreed on July 10, 2011, 04:50:37 pm
With the "Fill Light" slider in LR, it is tempting to push it to the left (darken when over exposed to the right) and then pull it out again to the right using "Fill Light".

To my way of thinking, this would result in the lightened areas at the end having less colour resolution than they started with because the darkening transformation takes away data that cannot be replaced - if a lossy transformation such as multiple (by a fraction) or divide are used with integers.

Is this correct?
Or does LR use floating point math internally for all of those pixels (the 97.9% red comes to mind) with originally integer values, allowing them to be scaled up and down without any loss of information?
Title: Re: Darkening, brightening and information loss
Post by: Jonathan Ratzlaff on July 10, 2011, 09:20:26 pm
Lightroom by definition is non destructive so anything you do has no permanent effect on the original.  That isone of the main reasons most people use it myself included
Title: Re: Darkening, brightening and information loss
Post by: Scott Hargis on July 11, 2011, 12:34:14 am
The Fill Light slider's default position is zero, so there's no moving it to the left unless you've already made an adjustment to the right. And since LR is non-destructive, the back-and-forth isn't going to change the image at all. You'll end up where you started.
Title: Re: Darkening, brightening and information loss
Post by: Jeremy Roussak on July 11, 2011, 03:30:29 am
The Fill Light slider's default position is zero, so there's no moving it to the left unless you've already made an adjustment to the right. And since LR is non-destructive, the back-and-forth isn't going to change the image at all. You'll end up where you started.
I expect he meant pull the Exposure slider to the left, then restore the dark areas with Fill Light.

Jeremy
Title: Re: Darkening, brightening and information loss
Post by: stamper on July 11, 2011, 04:52:59 am
I know that Raw is described as non destructive but in reality that can't be correct? If so then there wouldn't be a need for HDR or programmes to extend dynamic range?
Title: Re: Darkening, brightening and information loss
Post by: dreed on July 11, 2011, 05:24:26 am
Let me make the question clearer...

A pixel in your pixel has an RGB value, of say, (10,10,10).

If I then darken it by 66%, I might end up with (6,6,6) or (7,7,7) - depending if the integer math rounds up or truncates.

If I then brighten it by 33% I've either got (9,9,9) or (10,10,10).

Thus I'm curious to know if LR uses floating point math for all of its internal operations on pixels before they are presented or integer. This is important because the use of integer means that all brightening operations should be done first because there's a potential loss of information from darkening.

The problem is arguably less relevant for brighter colours.
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 11, 2011, 06:21:22 am
From what I know Lightroom (and ACR) will not do the calculations in the order you have made the adjustments. As long as all you adjustments are done on top of a RAW file (and we are not talking about round trips to TIFF or anything like this) then Lightroom can reorder the calculations to get the best and most correct result. I don't know what precision Lightroom uses internally for the calculations but would expect no precision loss (that would be too silly ;) ). And be sure there will no integer representation between the various operations as this would cause severe imprecision.

As long as your pictures looks good to you, you can safely use fill light :) But I would advise to not only use fill light to adjust. If you have an overexposed picture you can use a combination of reduction in exposure and recovery or even the tone curve to drag down the light level in the over exposed area as far as you can recover the details. Then the picture typically will become too dark and you can then use the brightness slider to compensate and to some degree the fill light slider. You can also use the fill light and the blacks slider to iteratively come to the look you like for what remains after you have recovered for the over exposure. Alternatively use the graduated filter if it is a sky that is overexposed or too bright.

Title: Re: Darkening, brightening and information loss
Post by: Costas on July 11, 2011, 06:35:12 am
Deleted - answered above
Title: Re: Darkening, brightening and information loss
Post by: sandymc on July 11, 2011, 08:13:25 am
Let me make the question clearer...

A pixel in your pixel has an RGB value, of say, (10,10,10).

If I then darken it by 66%, I might end up with (6,6,6) or (7,7,7) - depending if the integer math rounds up or truncates.

If I then brighten it by 33% I've either got (9,9,9) or (10,10,10).

Thus I'm curious to know if LR uses floating point math for all of its internal operations on pixels before they are presented or integer. This is important because the use of integer means that all brightening operations should be done first because there's a potential loss of information from darkening.

The problem is arguably less relevant for brighter colours.

Short answer is that LR always keeps the the original 10,10,10 pixel stored, regardless of what adjustments you make. That's what makes it "non-destructive". To display, print or whatever, it starts with that original pixel value, and does all the adjustments on that to get to an output value. So the kind of math doesn't matter, and you can go back and forth on adjustments as many times as you want.

Sandy
Title: Re: Darkening, brightening and information loss
Post by: dreed on July 11, 2011, 08:34:38 am
Short answer is that LR always keeps the the original 10,10,10 pixel stored, regardless of what adjustments you make. That's what makes it "non-destructive".

That the (10,10,10) is kept is unrelated to my question.

Quote
To display, print or whatever, it starts with that original pixel value, and does all the adjustments on that to get to an output value. So the kind of math doesn't matter, and you can go back and forth on adjustments as many times as you want.

It does matter because the order of operations and what type of operations are performed impact the end result. For example, if you do integer math (no fractions) with 20/6*6 then you end up with 18. If you do floating point, then you end up with 20 again (or a very close approximation thereof.)

Now someone else has eluded to the possibility that whilst LR presents a sequence of operations in your develop list, the actual transformation performed on the picture is a compressed one where LR works out what the final transformation matrix should be and applies that to the picture rather than apply a sequence of individual transformations. That's actually quite an important detail.
Title: Re: Darkening, brightening and information loss
Post by: eliedinur on July 11, 2011, 09:04:47 am
Surely in 14 bit, or even 12, quantization errors are trivial?

Quote
Now someone else has eluded to the possibility that whilst LR presents a sequence of operations in your develop list, the actual transformation performed on the picture is a compressed one where LR works out what the final transformation matrix should be and applies that to the picture rather than apply a sequence of individual transformations. That's actually quite an important detail.
Either way, internal workflow cannot be altered by the user.
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 11, 2011, 09:20:25 am
Thus I'm curious to know if LR uses floating point math for all of its internal operations on pixels before they are presented or integer. This is important because the use of integer means that all brightening operations should be done first because there's a potential loss of information from darkening.

I could be wrong but I don’t believe it uses floating point math for this (Schewe?) but I do know the order of edits is not in the order applied by the user but rather one optimized for image processing.

The ‘non destructive’ discussion based on the idea that the original isn’t touched doesn’t wash with me. If that’s the definition, we’ve had this kind of editing from the day “Save As” was an option in any application.
Title: Re: Darkening, brightening and information loss
Post by: dreed on July 11, 2011, 10:08:50 am
The ‘non destructive’ discussion based on the idea that the original isn’t touched doesn’t wash with me. If that’s the definition, we’ve had this kind of editing from the day “Save As” was an option in any application.

Indeed we have, but if you chose Save instead then that action was destructive. In LR, there is no concept of a Save, you have to use Export to get the image with its deltas applied to it in a file.

The closest LR has to Save is saving the metadata out to XMP files.
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 11, 2011, 10:37:35 am
In LR, there is no concept of a Save, you have to use Export to get the image with its deltas applied to it in a file.

Yup, and if the original data is a raw, the new data is built from scratch so to speak from the raw data, this process I’d consider non destructive (its pixel creation).

Take a rendered image, apply edits in LR and export out a new iteration. No damage?
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 11, 2011, 11:02:24 am

The closest LR has to Save is saving the metadata out to XMP files.

That's not a save at all. Lightroom stores all edits you do in the Lightroom catalog (or database) and you can choose have this data replicated in an xmp file but only of the file is the original RAW file. xmp save cannont be done for virtual copies.

I understand your question about precision, but think of it: The Adobe engineers would be rather stupid if the lost precision on the way, wouldn't they? As mentioned Lightroom reorders the operations to achieve the best result and you can be assured that the internal algorithms will work with the needed precision to not loose quality. And I have no doubt that the question about quality is a lot more than just the precision of the number representations along the way.
Title: Re: Darkening, brightening and information loss
Post by: NikoJorj on July 11, 2011, 11:37:15 am
Surely in 14 bit, or even 12, quantization errors are trivial?
Yes, and from what I've understood the math is done in 16-bits integer, ie more than the 14 or 12-bits data used in DSLRs.
Title: Re: Darkening, brightening and information loss
Post by: stamper on July 11, 2011, 11:50:04 am
Short answer is that LR always keeps the the original 10,10,10 pixel stored, regardless of what adjustments you make. That's what makes it "non-destructive". To display, print or whatever, it starts with that original pixel value, and does all the adjustments on that to get to an output value. So the kind of math doesn't matter, and you can go back and forth on adjustments as many times as you want.

Sandy

A lot of posters state that LR is non-destructive. That implies the raw data can be destroyed, so the question is what program does destroy it?
Title: Re: Darkening, brightening and information loss
Post by: Jeremy Roussak on July 11, 2011, 01:03:48 pm
Take a rendered image, apply edits in LR and export out a new iteration. No damage?
Correct - no damage. Even after the export, you can backtrack to any stage you like in the history of your LR editing changes, which LR will preserve for you permanently.

Of course, if you overwrite the initial rendered image with the newly exported one, you're opening a whole can of worms.

A lot of posters state that LR is non-destructive. That implies the raw data can be destroyed, so the question is what program does destroy it?
rm, especially rm -P  ;)

Jeremy
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 11, 2011, 01:28:39 pm
Correct - no damage. Even after the export, you can backtrack to any stage you like in the history of your LR editing changes, which LR will preserve for you permanently.

I’m not referring to the original, that’s not important (again, the Save As analogy). The question is about the new iterations FROM the rendered original. No damage?
Title: Re: Darkening, brightening and information loss
Post by: Jeremy Roussak on July 11, 2011, 03:24:58 pm
I’m not referring to the original, that’s not important (again, the Save As analogy). The question is about the new iterations FROM the rendered original. No damage?
Improvements, I'd rather hope!

It will depend on the saved format, won't it? If you use JPEG, there'll be loads of damage. TIFF, probably none. But we were talking about non-destructive editing.

Perhaps we're at cross-purposes.

Jeremy
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 11, 2011, 03:30:17 pm
Improvements, I'd rather hope!

Yes, that’s why we edit images. But that has nothing to do with the terms used (destructive or non destructive).

Quote
TIFF, probably none. But we were talking about non-destructive editing.

Exactly, so is it or isn’t it?
Title: Re: Darkening, brightening and information loss
Post by: eliedinur on July 11, 2011, 05:42:59 pm
Quote
TIFF, probably none. But we were talking about non-destructive editing.

Quote
Exactly, so is it or isn’t it?

Aside from any question of whether edits done in linear space are destructive to the "RAW" data (two thirds of which is new data generated by the demosaicing and not contained in the RAW file), the final, inevitable operation of gamma correction is certainly destructive, as shadows are stretched and highlights compressed. It is the price we have to (willingly) pay for a normal looking image, in a way similar to the fact that cellular attrition is the price of life.
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 11, 2011, 05:49:34 pm
Aside from any question of whether edits done in linear space are destructive to the "RAW" data (two thirds of which is new data generated by the demosaicing and not contained in the RAW file), the final, inevitable operation of gamma correction is certainly destructive, as shadows are stretched and highlights compressed. It is the price we have to (willingly) pay for a normal looking image, in a way similar to the fact that cellular attrition is the price of life.

Although what you say is right, this is not what is meant by nondestructive ;) Nondestructive simply means that we can generate a final image from the RAW data without making a pixel version at any stage and the editing operations are pure metadata and not operations done on pixels. So until we export from Lightroom all we have is the original RAW file and a bunch of editing instructions!
Title: Re: Darkening, brightening and information loss
Post by: Schewe on July 11, 2011, 06:13:58 pm
I could be wrong but I don’t believe it uses floating point math for this (Schewe?) but I do know the order of edits is not in the order applied by the user but rather one optimized for image processing.

I could tell you but then I would have to kill you :~)

Actually, it's pointless to speculate and irrelevant whether or not ACR/LR using integer or floating point cause there's nothing a user can do to change it. And while you might think you know something, compared to the Bright Boys™ at Adobe, we all know nothing (as Sargent Schultz used to say). So, my suggestion is quit worrying about it.

As for the rest of the discussion, the order in which ACR/LR actually processes is fixed in a certain order by the engineers. Users can't change that and the order in which a user sets their settings has zero impact on the actual order of the processing.

In terms of non-destructive editing, that refers to parametric editing vs pixel based editing. Inside of ACR/LR is all edits are parameter edits and no pixel editing is done until you render an image. Photoshop can also do non-desructing editing via Adjustment Layers which also parametric edits.

Once you render a file, in ACR/LR or PS, subsequent edits will be data destructive to a certain extent. Some formats are also more destructive such as JPEG.

So, is the OP still in the building? Does this answer your question?

:~)
Title: Re: Darkening, brightening and information loss
Post by: eliedinur on July 12, 2011, 07:07:50 am
Although what you say is right, this is not what is meant by nondestructive ;) Nondestructive simply means that we can generate a final image from the RAW data without making a pixel version at any stage and the editing operations are pure metadata and not operations done on pixels. So until we export from Lightroom all we have is the original RAW file and a bunch of editing instructions!
True, but that is not what D.D. and the OP are talking about:
Quote
@digitaldog: I’m not referring to the original, that’s not important (again, the Save As analogy). The question is about the new iterations FROM the rendered original. No damage?
I pointed out that even the first rendering - and even if it is written to tif/psd - will involve "damage" to the RAW capture data.

A distinction needs to be made here between "destruction", which is not what the OP asked about and the introduction of the term "non-destructive" already in the second post was OT and based on misunderstanding, and "damage" to data in the pipeline to an export.
Title: Re: Darkening, brightening and information loss
Post by: Alan Goldhammer on July 12, 2011, 08:08:24 am
The OP's question (as I read it) was "will there be any artifacts introduced into an image as the result of LR processing steps" and not whether the underlying RAW file was altered in any way (we all know that as long as one stays in LR the RAW file is untouched).  Whether such artifacts are visible or not is maybe the key point here.  As Jeff points out, the order of operations and how they are accomplished is within the LR code design and we are not likely to find out from Adobe how this is accomplished.  I don't think anyone has answered the OP's question.
Title: Re: Darkening, brightening and information loss
Post by: John R Smith on July 12, 2011, 08:24:22 am
The OP's question (as I read it) was "will there be any artifacts introduced into an image as the result of LR processing steps" and not whether the underlying RAW file was altered in any way (we all know that as long as one stays in LR the RAW file is untouched).  Whether such artifacts are visible or not is maybe the key point here.  As Jeff points out, the order of operations and how they are accomplished is within the LR code design and we are not likely to find out from Adobe how this is accomplished.  I don't think anyone has answered the OP's question.

That’s how I read it, too, Alan. And in general terms, yes of course LR can screw up your image in all sorts of ways if you push any of the processing steps to their extremes. The advantage is that in RAW at 16-bit you have a bit more headroom than if you were working with a pixel editor. I have accidentally produced all sorts of nasty artefacts in LR by ham-fisted editing, which if I had then rendered them to a TIFF would have been a permanent change. There is no especial voodoo charm to working in RAW, and this applies to any RAW editor, not just LR. My own 3FR files have an enormous amount of latitude, but I can still find the limits and they will then fall apart whatever the software.

But I think the OP’s post was very specific, and basically asked – if I darken the image using one slider (say EV or Brightness), and then lighten with another (say Shadow Fill) are the areas which started off at say 50, 50, 50 when lightened back to the same values compromised in some way? It seems to me that without documentation for the exact nature of the underlying process engine we have no way of knowing.

John
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 12, 2011, 09:20:15 am
I pointed out that even the first rendering - and even if it is written to tif/psd - will involve "damage" to the RAW capture data.

How so, the raw is read only.

The damage I’m asking about is the iteration of a TIFF (no reason to talk JPEG) from processing through LR or ACR FROM a rendered original, not raw.
Title: Re: Darkening, brightening and information loss
Post by: dreed on July 12, 2011, 03:10:56 pm
That’s how I read it, too, Alan. And in general terms, yes of course LR can screw up your image in all sorts of ways if you push any of the processing steps to their extremes. The advantage is that in RAW at 16-bit you have a bit more headroom than if you were working with a pixel editor. I have accidentally produced all sorts of nasty artefacts in LR by ham-fisted editing, which if I had then rendered them to a TIFF would have been a permanent change. There is no especial voodoo charm to working in RAW, and this applies to any RAW editor, not just LR. My own 3FR files have an enormous amount of latitude, but I can still find the limits and they will then fall apart whatever the software.

But I think the OP’s post was very specific, and basically asked – if I darken the image using one slider (say EV or Brightness), and then lighten with another (say Shadow Fill) are the areas which started off at say 50, 50, 50 when lightened back to the same values compromised in some way? It seems to me that without documentation for the exact nature of the underlying process engine we have no way of knowing.

John

Pretty much, yes.  And after reading Schewe's comments, I think I am (or we are) just going to have to settle for not getting a detailed answer on this and trust that somehow LR will do the right thing and do what I want it to :)
Title: Re: Darkening, brightening and information loss
Post by: eliedinur on July 12, 2011, 06:21:44 pm
Quote
Posted by: elied
I pointed out that even the first rendering - and even if it is written to tif/psd - will involve "damage" to the RAW capture data.
Posted by: digitaldog
How so, the raw is read only.

The file is not affected, but the data extracted from it is certainly changed. The application of a gamma correction TRC skewers and distorts the linear data and in this sense it can be seen as "damaging" it.
Title: Re: Darkening, brightening and information loss
Post by: digitaldog on July 12, 2011, 06:52:29 pm
The file is not affected, but the data extracted from it is certainly changed.

The last bit will involve "damage" to the RAW capture data. seemed to be directed at the raw, not the results. And I was again referring to a rendered image (not a raw). A rendered image processed from a rendered image. So it seems you are saying the rendered data from the rendered original is changed (damaged?).
Title: Re: Darkening, brightening and information loss
Post by: eliedinur on July 12, 2011, 07:53:06 pm
I remember five years ago when LR was first released, one of the advantages claimed for it was that the integrity of the capture data is preserved because all calculations are made in linear space. I accept that claim as valid and justified. But the time inevitably comes when the conversion to ProPhoto RGB (gamma 1.8), Adobe RGB (gamma 2.2) or sRGB (gamma sort of 2.2 but not exactly) must be made. And at that point the data gets stretched at one end of the DR and squashed at the other end and generally turned into something that no mother sensor would recognize as her progeny.
Title: Re: Darkening, brightening and information loss
Post by: hjulenissen on July 13, 2011, 02:56:38 am
When you have a raw sensor file in the Bayer format of 12-14 (linear) bits per channel and non-standard color filter response that you want to display on your typical display/printer of 8bits (2.2 gamma) sRGB/ARGB (standardized primaries) device, there is going to be some "loss" somewhere along the chain nomatter what. There is also going to be a radical change in numerical values, and this is a good thing as it makes the resulting image worth looking at.

The argument about Lightroom being "non-destructive" is that it separates image data (pixels) and editing instructions. Just like Microsoft Word is different from a typewriter - it does not save an image file containing the pixels of the rendered text. Rather, it works on a set of instructions that can be used to render the text at any time, but also do spell-checking and swapping the font. The original raw file is of course never touched, but the neatness of Lightroom is that it will read that raw file into memory, do your edits, and render the result to screen/export/printer. If you change one parameter, it will re-do this process (it may do internal optimizations, but analyzed as a user-oriented black-box I think this is correct). The important "philosophical" implication is that Lightroom isolates user 1. intent ("make the image brighter") and 2. image processing ("multiply all pixels by 2"). If the Adobe guys at some point find more clever ways to transform 1->2, they can do so (and atleast with the demosaicing in LR3, they did). They may theoretically do limited-precision integer processing in one version of Lightroom, and convert to single-precision floating point in another version as long as they deem that it gives only better end-result without compromising the intent of the original edit (this is mind-bogling because some users might actually want editing-induced banding and quantization and consider it a part of their edit. They may not be pleased with a radical new engine that does the processing "better").

Are you having problems with the end-results in Lightroom today? I guess that if you went to dcraw or something similar, you would have access to the ideal documentation of image processing (open-source code) so that you could put your mind at rest, or improve parts that you did not like.

-h
Title: Re: Darkening, brightening and information loss
Post by: Alan Goldhammer on July 13, 2011, 07:37:47 am
If you change one parameter, it will re-do this process (it may do internal optimizations, but analyzed as a user-oriented black-box I think this is correct). The important "philosophical" implication is that Lightroom isolates user 1. intent ("make the image brighter") and 2. image processing ("multiply all pixels by 2"). If the Adobe guys at some point find more clever ways to transform 1->2, they can do so (and atleast with the demosaicing in LR3, they did). They may theoretically do limited-precision integer processing in one version of Lightroom, and convert to single-precision floating point in another version as long as they deem that it gives only better end-result without compromising the intent of the original edit (this is mind-bogling because some users might actually want editing-induced banding and quantization and consider it a part of their edit. They may not be pleased with a radical new engine that does the processing "better").
They did change the processing with LR 3 and for those of us who had a lot of processed images using LR2, this is duly noted when you are in LR and the new sharpening and noise reduction capabilities argue for the redoing of some images.  Who knows what will happen with LR 4 when it appears (other than the longed for soft proofing feature that many are clamoring for).
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 13, 2011, 10:32:27 am
They did change the processing with LR 3 and for those of us who had a lot of processed images using LR2, this is duly noted when you are in LR and the new sharpening and noise reduction capabilities argue for the redoing of some images.  Who knows what will happen with LR 4 when it appears (other than the longed for soft proofing feature that many are clamoring for).

That's correct, but I think it is worth noting that the old algorithms known as process version 2003 were kept as default and only if you wanted the new algorithm 2010 you would choose to do so. Old pictures were 2003 by default, so I Adobe did a good thing not to change the look of the pictures by one upgrade.

I did change all my pictures to process version 2010 since I felt the improvement was worth the change.
Title: Re: Darkening, brightening and information loss
Post by: kenlip on July 16, 2011, 06:54:48 am
After reading this thread so far, I have had to rephrase the OP's question in my mind to be sure that I am fully understanding what he is asking.   I thought I might share with you how I see it.


Assume ...

I have a file - let's call it Test.RAW.  I make two copies of it, one called Test_E_F.RAW and the other called Test_F_E.RAW

I open Test_E_F.RAW in Lightroom, wang  the Exposure slider to the left a certain amount and then wang the Fill slider to the right a certain amount

I then open Test_F_E.RAW and reverse the order of the wanging - first the Fill slider to the right and then the Exposure slider to the left, both by the identical amounts I did for Test_E_F.

The Question:   Will there be any difference in the two images, even before exporting them into a usable format such a TIFF?

The OP seemed to believe there might be.  He explained why he thinks so, based on the algorithms/maths that Lightroom MAY be using, and suggested that, if there is a difference, it could be important to make adjustments in a particular sequence.

So far the answers seem to have been to the effect that Adobe has (probably!) considered the maths and will compensate for any anomalies that might arise from the sequence in which the changes are applied.  Therefore, Test_E_F and Test_F_E will be identical, and will produce identical images when exported from Lightroom into a different format e.g. TIFF or if they are printed from Lightroom.

Of course, even after all the slider wanging, the three files - Test.RAW, Test_E_F.RAW and Test_F_E.RAW are exactly as they were originally and will never be changed irrespective of what sliders are wanged in Lightroom or what format is used when exporting the images from Lightroom.


I hope this rephrasing of the OP's question helps.
Ken

Title: Re: Darkening, brightening and information loss
Post by: sandymc on July 16, 2011, 07:37:26 am
So far the answers seem to have been to the effect that Adobe has (probably!) considered the maths and will compensate for any anomalies that might arise from the sequence in which the changes are applied.  Therefore, Test_E_F and Test_F_E will be identical, and will produce identical images when exported from Lightroom into a different format e.g. TIFF or if they are printed from Lightroom.

The two TIFF output images will be identical, but it has nothing to do with the math or any compensation for anomalies. As I pointed out in reply 8 above, because LR aways starts from the original data and processes in a fixed sequence, if the various sliders are in the same position, then the output will be the same. How many time the sliders have moved in whatever direction doesn't matter. This would be the case even if you used 4-bit integer arithmetic - the images would look terrible, but they would be identical.

Sandy
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 16, 2011, 09:44:00 am
No, the images will not be identical which you can easily test. Fill light is not the inverse of exposure. As also brightness is not either.
Title: Re: Darkening, brightening and information loss
Post by: Richowens on July 16, 2011, 04:55:52 pm
Hans,

If the slider movements are the same amount then the two files will be the same. It matters not what order the movements are made, just the amount.

Rich
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 16, 2011, 05:30:41 pm
I didn't mention order... As said above fill light is not the inverse of exposure and the also btw. have different metrics, so an equal amount does not make sense. Exposure sets the white point, fill light lifts the shadows and does not move the white point.
Title: Re: Darkening, brightening and information loss
Post by: Eric Myrvaagnes on July 16, 2011, 07:28:55 pm
But there's no (good) reason to assume that any slider is the inverse of any other slider in LR. The only version of the OP's questionthat makes sense is if you end up with EACH slider in the SAME position at the end, but do the operations in a different order (NOT on different sliders). As numerous posters have mentioned, LR does its processing based only on the position of the sliders at rendering time and not on the order in which they were manipulated in getting to that final position. And the user has no control over the order in which LR does the processing steps.
Title: Re: Darkening, brightening and information loss
Post by: Richowens on July 16, 2011, 08:00:45 pm
Hans,

  Perhaps I did not make myself clear. On picture 1 he slid the slider for exposure to 4.34, recovery he slid to 2.76, in that order.
 On picture 2 he slid the recovery slider to 2.76, exposure he slid to 4.34, in that order.

  Both pictures will be the same no matter what the order. He did not put both exposure and recovery to say 3.39 on picture 1, then set both exposure and recovery to say 2.54 on picture 2. If he did that of course they will look different.

Rich
Title: Re: Darkening, brightening and information loss
Post by: Schewe on July 16, 2011, 09:00:38 pm
Yeah, ya know, let's be crystal clear on this...the order of the settings adjustment makes zero difference to the order that ACR/LR will end up doing the processing. They are two completely different animals.

The parametric nature of the settings means you can keep going back and changing the settings applied to the raw file (not yet processed) and keep changing them forever and have zero impact on the raw file itself–only the settings stored.

Once you process the rendered file, it doesn't matter how many times your changed the settings and in what order, the actual order of the processing pipeline is dictated by the ACR/LR engineering team. The pipeline will always process the image in it's own internal optimal order which is fixed and unchanged by the user...
Title: Re: Darkening, brightening and information loss
Post by: kenlip on July 16, 2011, 11:09:11 pm
I didn't mention order...
As said above fill light is not the inverse of exposure and the also btw. have different metrics, so an equal amount does not make sense. Exposure sets the white point, fill light lifts the shadows and does not move the white point.

But I did mention order.  

Please read my posting again, with specific attention on lines 4 and 5.

That was the whole point of the exercise.  

The OP was suggesting that, if LR used integer maths or even floating point to just one or two decimal points, then as one wanged sliders back and forth the rounding errors would induce artifacts.  Order of using the sliders might make a difference if LR wasn't using floating point maths for its algorithms and if it doesn't have other algorithms built in to optimise the image irrespective of the order in which the sliders are used.

Most of us are in agreement that LR uses floating point and has algorithms to make the sequence of using the sliders irrelevant.  


Ken

Ken

Title: Re: Darkening, brightening and information loss
Post by: Schewe on July 16, 2011, 11:31:35 pm
Order of using the sliders might make a difference if LR wasn't using floating point maths for its algorithms and if it doesn't have other algorithms built in to optimise the image irrespective of the order in which the sliders are used.

And this is where I say, forget about it...regardless of HOW ACR/LR is doing what it's doing, the order that the user makes adjustment is irrelevant and will have ZERO impact on the results. Ya got that yet? Do what you want to do with the adjustments and ACR/LR will do what it wants to do with the adjustments and in the order ACR/LR decides is the optimal order. Really, we need to move on from this...
Title: Re: Darkening, brightening and information loss
Post by: Richowens on July 17, 2011, 12:45:45 am
You know..........sometimes t's better to marvel at the magic show than to try to figure out how the magician does it.
Title: Re: Darkening, brightening and information loss
Post by: Alan Goldhammer on July 17, 2011, 09:05:22 am
And this is where I say, forget about it...regardless of HOW ACR/LR is doing what it's doing, the order that the user makes adjustment is irrelevant and will have ZERO impact on the results. Ya got that yet? Do what you want to do with the adjustments and ACR/LR will do what it wants to do with the adjustments and in the order ACR/LR decides is the optimal order. Really, we need to move on from this...
From a visual point of view this is probably true but from a computational point of view it may not be IF AND ONLY IF LR does its calculations in a different order depending on how the user has made the changes.  I suspect that this will not happen since LR probably only recognizes the final changes to the sliders and will optimize the order (otherwise the code starts to become really complicated).  The only ones who can answer this for sure are the Adobe LR engineers and they probably have much better things to do with their time.
Title: Re: Darkening, brightening and information loss
Post by: Schewe on July 17, 2011, 11:28:13 am
From a visual point of view this is probably true but from a computational point of view it may not be IF AND ONLY IF LR does its calculations in a different order depending on how the user has made the changes.

Again...the order that the user does something has ZERO (you understand the word zero) impact on the order that ACR/LR does it's thing.

I'm not sure why it's so hard to understand that the ACR/LR pipeline is pre-determined by the engineers to always do things in the order that will be optimal for processing, not the order that the user may have adjusted things. Also understand that the order of the pipeline is really, really complicated using a series of pretty impressive algorithms to deal with tone and color as well as sharpening and noise reduction and through in local adjustments and even lens corrections. Really, you don't want to know how to make the sausage...just eat it.
Title: Re: Darkening, brightening and information loss
Post by: Alan Goldhammer on July 17, 2011, 11:56:54 am
Again...the order that the user does something has ZERO (you understand the word zero) impact on the order that ACR/LR does it's thing.

I'm not sure why it's so hard to understand that the ACR/LR pipeline is pre-determined by the engineers to always do things in the order that will be optimal for processing, not the order that the user may have adjusted things. Also understand that the order of the pipeline is really, really complicated using a series of pretty impressive algorithms to deal with tone and color as well as sharpening and noise reduction and through in local adjustments and even lens corrections. Really, you don't want to know how to make the sausage...just eat it.
Jeff, you are not reading my statement correctly (and I am in your camp on this).  The only people who really can answer the question definitively are the software engineers who built the program.  I've done enough programming in my lifetime to realize that the order one does certain calculations with algorithms can lead to disparities in data.  Usually these disparities are meaningless since they are down to many zeros to the right of the decimal point but they are there none-the-less (and even if they were present in LR as the result of calculations they would be meaningless).  I agree that Adobe has designed an optimal product that will give excellent results and that there is NO IMPACT on the rendered image when it comes to printing.  On this we are in violent agreement.
Title: Re: Darkening, brightening and information loss
Post by: Hans Kruse on July 17, 2011, 12:18:57 pm
Hans,

  Perhaps I did not make myself clear. On picture 1 he slid the slider for exposure to 4.34, recovery he slid to 2.76, in that order.
 On picture 2 he slid the recovery slider to 2.76, exposure he slid to 4.34, in that order.

  Both pictures will be the same no matter what the order. He did not put both exposure and recovery to say 3.39 on picture 1, then set both exposure and recovery to say 2.54 on picture 2. If he did that of course they will look different.

Rich

You are right, I somehow misread the post I responded to with the examples. I guess I thought we had left the discussion about the order at that point ;)
Title: Re: Darkening, brightening and information loss
Post by: Schewe on July 18, 2011, 04:59:05 am
I've done enough programming in my lifetime to realize that the order one does certain calculations with algorithms can lead to disparities in data.

Uh huh...and if the order that the user might make an adjustments had ANY impact on the processing order, that would be interesting and useful.

Alas, that is not the case...regardless of what the user may do, the guys from Adobe have already chosen the optimal processing pipeline for ACR/LR. Regardless of what the user may or may not do, the actual order that ACR/LR has already been determined and fixed by the engineers...

While it's possible for a user to incorrectly adjust the settings for a given image, nothing the user might do will alter the order in which ACR/LR processes the image. Sure, users can screw up an image by selecting inappropriate settings but the order that the user may make the adjustments is simply not a factor.

OK? Can we move on? I really doesn't matter HOW the engineers do it (really, it doesn't) what matters is how you use the settings...

Are we done yet?
Title: Re: Darkening, brightening and information loss
Post by: Alan Goldhammer on July 18, 2011, 07:08:57 am
Are we done yet?
I'm done! ;)