The goal here was to make a version that was pretty close to yours without going nuts. Note my rendering is a bit warmer. I see the biggest differences in the water in the foreground.
This isn't the Whipping Post, and I don't think anyone wants a contest here. I just want to offer some observations, which I hope are helpful (to Pindy, at least). There are things I like about each of these versions, and things I don't (personal preference, perhaps, but I think in some cases not just that).
I agree with Andrew's take that the biggest difference is in the water in the foreground. This seems to me an artistic choice more than one being superior. There are also differences in the sail cover in the boat in the left rear (and other blues in the shot, which are move evident in Pindy's version), sharpening (Pindy apparently didn't sharpen), and the dock area (and generally highlights in the foreground, where Pindy's version seems to me overdone, saturation-wise).
As Andrew implies, it comes down to a cost-benefit trade-off. What do you want to get out of this particular shot, and how much are you willing to spend on it (not only time to do the post-processing, but time to learn the tools). This is a highly individual thing, and consequently I don't think there's a single answer.
This isn't the Whipping Post, and I don't think anyone wants a contest here. I just want to offer some observations, which I hope are helpful (to Pindy, at least). There are things I like about each of these versions, and things I don't (personal preference, perhaps, but I think in some cases not just that).
(Pindy apparently didn't sharpen)
Evidently I didn't sharpen enough! I sharpened a bit in LR and did some USM in PS, but I'm still trying to figure out how much is enough.
I appreciate your comments (all of you) especially about the over-saturated dock, which I was seeing more as a warming than an over-saturation, though it did go quite red. The sail cover, I noticed in Andrew's version, went way darker than it originally appeared, but all's fair when you're trying to conjure a vision I suppose. I realise there is more I can get from Lightroom, but sometimes for me it's the difference between being able to fix a tool's side effects (think blend modes in PS) and not being able to in LR. I felt on this image, and in particular on photos with a lot of foliage (I have hundreds), that my contrast/tone curve adjustments in LR send to over-saturate and I haven't learned to de-saturate them by other means in a way I find pleasing, including the vibrance and saturation sliders going negative. Looking at Andrew's DNG is enlightening in some ways, which I why I asked for it in the first place. Lots more experimentation is required.
Evidently I didn't sharpen enough! I sharpened a bit in LR and did some USM in PS, but I'm still trying to figure out how much is enough.
I didn't touch LR.'s sharpening controls, those are the defaults.
Andrew's DNG is enlightening in some ways, which I why I asked for it in the first place. Lots more experimentation is required.
And that's my point. You or others would have to ask Dan or Scott why they didn't render the image from the Raw converter closer to the goals instead of using a default rendering (you can see how hugely "off" it is in terms of your rendering goals) then dismiss the Raw converter, not collect $200 and go directly to Photoshop. Least we forget, one of those above actually recommended setting ACR/LR in a default mode for ALL conversions and doing all further work in Photoshop. Damn silly I say.
This is like a darkroom printer not exposing their papers correctly for the neg, then letting the paper sit in a developer for a few hours until it appeared better, then god knows what technique after to make a decent appearing print. Rendering to the print starts in the Raw converter:
In fairness to Dan, his recommendation to stick with the defaults was based on an earlier version of ACR. He specifically said that his conclusions could, and probably would, change as ACR and other developers improved. He also said that certain of the ACR tools may be worth using even as it was -- notably the white balance tool.
His real objection to ACR, then and now, as I understand it, is the curve function is a global function, and that it cannot be isolated for particular colors. In theory, he certainly has a point. In practice, I'm not sure how much difference it makes.
Also, he maintains that by getting the picture in as good a shape as possible, you may be putting yourself into a position where you can't get it into the best possible shape in Photoshop. On this point, I don't know if he's right or not. It makes sense that if you set the endpoints too hard in ACR, then you are depriving yourself of some room to work on contrast with channel specific curves in PS. But I haven't actually run into too many shots where that was really an issue. And I'm also sure that I'm not that good a judge of when those situations arise.
So, Dan may still be holding onto his recommendation with the current version of ACR. But I think its pretty clear that the tide is against this recommendation. Future versions of ACR will probably let you have more control over particular colors, and contrast within those colors, either by curves or by some new device. That should put an end to whatever controversy might exist now over this point.
In fairness to Dan, his recommendation to stick with the defaults was based on an earlier version of ACR.
And his current recommendations?
His real objection to ACR, then and now, as I understand it, is the curve function is a global function, and that it cannot be isolated for particular colors. In theory, he certainly has a point. In practice, I'm not sure how much difference it makes.
Well Adobe designed these behaviors for a specific reason. I'd suggest this article by Mark Segal:
I don't know if he's got any current recommendation. I've never heard him address the flexibility the HSL tab now gives. I know that at one point, after the current ACR release, he said that he still though it was lacking, primarily because of the global curve question.
As I recall, you (or another on the color theory list, but I think it was you) issued a correction challenge. I would have liked to see the results of such an event.
One of the problems, of course, with a comparison of these tools, is that so much depends on the sensitivity and skill level of the people using the tools. I have no doubt that Dan could do a better job, using default settings and PS, than I could do with the full compliment of tools. That says more about Dan and me than it does about the tools themselves.
I do know that Dan looks at these sorts of things as like a competitive sport. if someone attended one of his classes and consistently got better results using ACR than Dan could get just using PS, then I'm pretty sure he would come around. I don't know if that has happenned. I'm willing to give it a shot, if someone will front me the money for one of his classes
Another possibility would be to have something like a challenge for a few images here. I think it could be a really good learning experience. One of the complaints that I've had about ACR, even in its new form, is that there is very little showing how to do anything more than the most basic manipulations.
When I purchased PSCS3, I didn't really take the time to really learn and appreciate the vast improvement that ARC 4 offered over that in CS2, and just jumped in using my techniques from ARC 3 in CS2. Big, big mistake.
As you and Andrew, both, know very well, it takes a while to fully appreciate all the new controls ( bells and whistles ) in ARC 4.3. It is like getting a P-51 Mustang, after only having a Spad to fly about in.
Jeffe Schewe and Bruce Fraser's Real World "Camera RAW with Adobe Photoshop CS3" is well worth reading more than twice.
I examine each image after it is rendered into a jpg in Photoshop in each of the color channels. Set white and black points, then CTRL-1, CTRL-2, CTRL-3 Bling, bling, bling to find where the contrast lies, and to see if I can utilize that contrast to effect in the image. CTRL-~ and I am back to color viewing all in less than 20 seconds. If I can use the contrast in one of the RGB channels to create more contrast between subject and background, I will then include the Apply Image command in my workflow. If I do not need that separation, I will omit it, but it can be very helpful to create more depth in an image if used subtly and carefully, by using layers and blending modes.
In short, I am not an either/or kind of fellow, but I am willing to try whatever works quickly and easily to help me maximize my images.
But, Adobe Raw Converter 4.3 is my foundation these days.
I do know that Dan looks at these sorts of things as like a competitive sport. if someone attended one of his classes and consistently got better results using ACR than Dan could get just using PS, then I'm pretty sure he would come around.
That's a challenged that was made to him and ignored.
That's a challenged that was made to him and ignored.
Do you guys (Duffy and Andrew) believe that there is one and only one "right" way to make corrections and anyone using a different way must learn that they should switch to the "right" way?
Or do you believe that there are many different ways to achieve a desired result and any way that a particular photographer is familiar with and has the time for that gives them the result they are after is perfectly fine?
I ask because it looks like the way this thread is going is that it's simply wrong to do anything in CSx that could have been done in ACR/LR. I will agree that doing as much as possible in LR can offer some workflow efficiency advantages (particularly when all adjustments can be kept in LR and none are needed in CSx).
But, if you focus only on the final quality of one particular image, once you've got the basic tone and white balance in the proper ballpark in the RAW processor, does it really matter to the quality of that final image whether the rest of the fine tuning of the image is done in LR/ACR or CSx?
For example, if you want to emphasize the contrast of one particular element in the photo that's in a mid-tone range, does it matter whether you do that in ACR/LR or in CSx? If you want to darken a sky, does it matter whether you do that in ACR/LR or in CSx?
There's no right and wrong but there is a modern method and an older one taking place here. The first question is, where is metdata versus pixel editing the most appropriate technique for making a desired color appearance?
One is totally non destructive, the other isn't.
One is simply a set of metadata instructions and the other is baked and rebaking pixels.
One is fast the other isn't.
One provides a history of every edit ever done, the other doesn't.
One is done high bit, linear encoded data, the other happens (often) in lower bit and gamma corrected color space (on baked pixels).
One provides only global corrections (for the time being) the other doesn't have that restriction.
There's no right and wrong but there is a modern method and an older one taking place here. The first question is, where is metdata versus pixel editing the most appropriate technique for making a desired color appearance?
One is totally non destructive, the other isn't.
One is simply a set of metadata instructions and the other is baked and rebaking pixels.
One is fast the other isn't.
One provides a history of every edit ever done, the other doesn't.
One is done high bit, linear encoded data, the other happens (often) in lower bit and gamma corrected color space (on baked pixels).
One provides only global corrections (for the time being) the other doesn't have that restriction.
Yeah, I understand most of those points.
Fast depends upon what you know. There are still several types of corrections that I know how to do faster in CS3 or maybe don't know how to do at all in LR/ACR. Maybe you could do them faster in LR, but fast for me is what I personally know how to do best.
My main reason for going to CS3 these days is for non-global corrections where I use a mask or a blend-if. I'm also much more comfortable with sharpening in CS3, but that could just be due to a learning curve in LR.
The reason I asked this question is I look at all of these as just different tools and different artists prefer different tools. There's no right or wrong tool. Each tool has its advantages and its disadvantages and it's up to the artist to choose the tool that works best for them. There will be "old school" artists who get great results using tools that have been around for awhile and there will be "new school" artists who use the newest tools for the maximum that they can achieve. There will be in-between artisits who know how to do some things in the new tools, but still use the old tools for other things (some of which could be done in the new tools). Each is doing what they're doing because it works best for them. I see no right or wrong in any of the choices.
Is the "new school" artist getting better results? Who knows? It probably depends more on their skill as an artist than it does on which tools they pick.
Is the "new school" artist being more efficient by using the new tools? Maybe, but that also depends on a bunch of other things too. It's possible to spend a ridiculous amount of time in the new tools trying to achieve a result that's really easy in the old tools, so it can work both ways.
Is the "new school" artist being more efficient by using the new tools? Maybe, but that also depends on a bunch of other things too. It's possible to spend a ridiculous amount of time in the new tools trying to achieve a result that's really easy in the old tools, so it can work both ways.
That's what we're talking about, tools, not art.
What's the right tool for the right job?
There's no right or wrong tool
Ever tried to use a kitchen knife as a screw driver? Not recommended.
Would you ever consider it acceptable for someone to use LAB mode in CS3 to achieve an editing goal? I'm not asking if you would use it for your own image editing or even if you would ever recommend a LAB mode technique, but would you ever consider it acceptable for someone else to use it who is familiar with how to achieve a result they like in LAB mode?
Also, thanks for stating your opinions of my tone for all readers ("its very clear to all").
The quote works for me, I may not be all readers - but I am one reader that sadly knows more about this than I would like to know.
This is one of the issues Andrew, your past track record precedes you, in some respects in a negative fashion. Even if this is not what you 'state as your intent', many readers will take away what they will knowing where your preferences lie and the history of your interaction with some other authors, your feelings on various methods and how often you bring them up in other posts on other lists when they are not a current topic or the person is not there to comment or set the record straight with thier side of the story (which is often different from the side you post).
Andrew, i think this post was intended for your therapist and not this forum?
The OP rendered the best version that they knew how before using the Kelby Apply Image method. Dan did not encourage Pindy to render a default version from ACR.
Dan was not mentioned in the OP. Dan was mentioned in some follow up replies, as he is famous for being an authority on channel blending.
If I could read 100 consecutive posts of yours that did not mention Dan Margulis or associated subjects I would be very happy (and I am sure that many other viewers would be too).
This is one of the issues Andrew, your past track record precedes you, in some respects in a negative fashion. Even if this is not what you 'state as your intent', many readers will take away what they will knowing where your preferences lie and the history of your interaction with some other authors, your feelings on various methods and how often you bring them up in other posts on other lists when they are not a current topic or the person is not there to comment or set the record straight with thier side of the story (which is often different from the side you post).
I've asked questions and provided answers. What does the above have to do with the facts of the statements made?
Duffy, thankfully I am not a moderator here (one unpaid role is engough for me!), but I would mention from my previous interactions with Andew on and related subjects - that you take care in swallowing strawman bait. This thread was about the Kelby Apply Image method, but if you let Andrew get off his leash, the thread will go off topic and move onto other areas and lose all benefit to the topic at hand (why not split the thread up and a mod move the posts that are off topic)...Look how many pages and posts, if one to count them, how many would be on topic and how many off?
Dan is not the only one that does not like a curve that affects saturation for certain images, even photographers do not always like/wish/want this:
Those suggesting and or familar with such workflows (Lab, LCH, HSI, HSB etc) know and understand that if one decouples the hue/saturation edits from the tone curve that they will need to be addressed separately and that nobody is suggesting simply using pure tone edits without addressing colour in a separate move. Andrew Rodney from his past LinoColor days should know this as well.
Some think it is optimal to increase saturation, if required/desired.
Some think it is optimal to decrease saturation, if required/desired (ACR/ALR users that have no other option).
In fairness to Dan, his recommendation to stick with the defaults was based on an earlier version of ACR. He specifically said that his conclusions could, and probably would, change as ACR and other developers improved. He also said that certain of the ACR tools may be worth using even as it was -- notably the white balance tool.
His real objection to ACR, then and now, as I understand it, is the curve function is a global function, and that it cannot be isolated for particular colors. In theory, he certainly has a point. In practice, I'm not sure how much difference it makes.
Also, he maintains that by getting the picture in as good a shape as possible, you may be putting yourself into a position where you can't get it into the best possible shape in Photoshop. On this point, I don't know if he's right or not. It makes sense that if you set the endpoints too hard in ACR, then you are depriving yourself of some room to work on contrast with channel specific curves in PS. But I haven't actually run into too many shots where that was really an issue. And I'm also sure that I'm not that good a judge of when those situations arise.
So, Dan may still be holding onto his recommendation with the current version of ACR. But I think its pretty clear that the tide is against this recommendation. Future versions of ACR will probably let you have more control over particular colors, and contrast within those colors, either by curves or by some new device. That should put an end to whatever controversy might exist now over this point.
Another possibility would be to have something like a challenge for a few images here. I think it could be a really good learning experience. One of the complaints that I've had about ACR, even in its new form, is that there is very little showing how to do anything more than the most basic manipulations.
Duffy
I don't think a Texas Death Match between Dan and Andrew would be particularly useful.
I suggest something I think much more useful, especially in light of comments on this thread about a need for more workflow recommendations for raw processors.
Dan Margulis is currently advocating a specific workflow for getting pretty good results on a large range of images in a few minutes of work (his so called "Picture Postcard Workflow), which he's demonstrated at Photoshop World and is currently evangelizing though his videos on Kelby Training.
It would be great for someone to propose a different workflow using raw processor tools to get the same effects. This would allow anyone, of (almost) any skill level, to do the test themselves.
For what it's worth, the "Old School" vs "New School" argument doesn't work for me (think String Theory, very New School and very much in trouble these days). What works for me is practics - specific techniques that get much better results or sound theory that explains why better results will follow using techniques that follow this particular theory. Unfortately, Andrew, the theory arguments I've heard aren't that compelling. But first order, it's practics. The above suggestion was made from this perspective.
I don't know who recommended flipping over to PS from a RAW converter's default rendering, but it was neither myself nor Scott. I got fairly close in LR before wanting more color control than I felt I was getting from the HSL in LR. I didn't completely render that image in PS—I would say PS did about 30% of the work.
Kelby's book has you start in ACR or LR and take it as far as you can until (he recommends) that you would want local editing, Lab or anything that PS offers that ACR/LR doesn't handle, such as output sharpening.
Andrew, as you and I have discussed on this forum back when I knew less than a damn thing about RAW conversion, the default rendering in a RAW converter is useless and in no way anything other than a starting-off point. I don't believe anyone's saying it's anything else.
I don't think there is one right way to do things. In general, if a technique gets better results, or gets nearly the same results in less time, then I think its worth learning. And which techniques I will use will depend alot on how much time I have allocated. I used to use LAB techniques all the time for simple and quick fixes on pictures I could only devote a few minutes to. I don't use LAB for that type of image as much anymore since the release of the current ACR.
Stephen:
Your point is well taken. Once the main question in a thread has been answered (as I believe it has here), then I think the continuing discussion is just that. So hijacked threads tend not to bother me so much, but I can see how it can be a concern. Also, since the Kelby Technique that was the foundation of this thread had its genesis (as far as I know) in Dan's work, and was the subject of a chapter in the LAB book, I wasn't at all surprised at the way this thread veered.
JJBong:
I think you stated better what I was trying to get at. On Dgrin, I think it could be really useful to see how people are getting excellent results out of Camera Raw. There just isn't very much of that kind of analysis that I've seen. As for the "Texas Death Match," I would be very curious to see whether I prefered the Photoshop or ACR centered approach in the hands of very highly skilled people.
I was looking through Kelby's book about a week ago, and I noticed that he had people do curves in Photoshop, and not in ACR. Am I remembering that right? If so, does he give any explanation why? (I realize that its a Kelby book, so the "whys" might be few and far between, but I thought he might have said something about it.)
I don't think a Texas Death Match between Dan and Andrew would be particularly useful.
I suggest something I think much more useful, especially in light of comments on this thread about a need for more workflow recommendations for raw processors.
Dan Margulis is currently advocating a specific workflow for getting pretty good results on a large range of images in a few minutes of work (his so called "Picture Postcard Workflow), which he's demonstrated at Photoshop World and is currently evangelizing though his videos on Kelby Training.
It would be great for someone to propose a different workflow using raw processor tools to get the same effects. This would allow anyone, of (almost) any skill level, to do the test themselves.
For what it's worth, the "Old School" vs "New School" argument doesn't work for me (think String Theory, very New School and very much in trouble these days). What works for me is practics - specific techniques that get much better results or sound theory that explains why better results will follow using techniques that follow this particular theory. Unfortately, Andrew, the theory arguments I've heard aren't that compelling. But first order, it's practics. The above suggestion was made from this perspective.
I was getting worried about the direction this thread was drifting since earlier this afternoon, but it seems to be getting back on track. The field of image editing and processing workflow is filled with numerous individuals whom do not always agree completely in all areas - and I know that is a strong understatement. Hopefully, we can keep that disagreement civil here, even though I know that might be challenging.
I want to thank jjbong for this suggestion. I am a current subscriber to the Picture PostCard videos and hope to understand them better in a few weeks.
I, personally, find that knowing and understanding LAB is a useful skill, but one that I use relatively infrequently with the advent of CS3. LAB can certainly help in evaluating an image for color casts, as Dan describes in his writings and videos.
Andrew, as you and I have discussed on this forum back when I knew less than a damn thing about RAW conversion, the default rendering in a RAW converter is useless and in no way anything other than a starting-off point. I don't believe anyone's saying it's anything else.
One person is! But I'm told he is basing this on an old version of ACR (he doesn't appear to own or admit to owning Lightroom) and what is currently different in his advise is totally unclear.
As to Scott, don't have the book but, the question is not (aside from Dan), dismissing the silliness of setting a Raw converter in default mode and fixing in Photoshop, the real point is, when do we decide and know that we can no longer render the image with the controls provided and must now dive into Photoshop and what are the repercussions of using these two distinct tools as best as we can?
The OP showed an image of a scene using a Lab trick which didn't produce anything at all useful on the image I tried so I asked "what's it do?". Then we saw the final image (Lab) and a default LR rendering (not useful as I pointed out), then a rendering in LR using the controls and skill of the OP. I then used my understanding of LR to get a very close rendering in short order. So if anything, it proves, at least with this one example that we do have the control to render the image totally in a Raw processor to get very close to the other process and I think there are some advantages in both the tools and controls used and the results of the test that so far, haven't been discussed (Stephen would rather dismiss the real meat and potatoes here and talk about how unfortunate this conversation is and probably wishes he had moderator control to censor posts here has he does on the CT list, a list where a discussion, even a challenge like this would never see the light of day).
Different tools and techniques. What do each bring to the party? The bit about "I had to use Lab to produce this color appearance", at least in this case doesn't wash. Now onto the results of the two techniques, what some might feel is best practice, based on image quality, time, flexibility etc. Or maybe we should just censor ideas and just fix everything using older Photoshop techniques. When all you know is a hammer, everything looks like a nail!
I was looking through Kelby's book about a week ago, and I noticed that he had people do curves in Photoshop, and not in ACR. Am I remembering that right? If so, does he give any explanation why? (I realize that its a Kelby book, so the "whys" might be few and far between, but I thought he might have said something about it.)
Duffy
He tends to trade off. There are some lessons where you Tone Curve (Point always) and others where you Curves. I remember reading one explanation of why he had you bypass TC in favour of Curves, but I would have to look again to remember it.
In general, he employs Curves much more often in the book as an local editing/adjustment layer/mask/painting kind of thing. For the most part he will get more contrast boost out of black point/white point settings than automatically using Curves or Tone Curve to do a global contrast enhancement, though he will do this in some lessons. I don't disagree with this, as you can get a little too hung up on contrast sometimes.
There were a couple times I wondered why he didn't have you get more out of the TC.
Thanks. I'm going to have to go to the store again and see if the book is worth getting. The last Kelby book I got was the channels book, and I thought it was OK. I learned a couple of things from it, but overall I thought I would have done better getting a book with more thorough explanations. The thing that I find attractive about this book is that, even without explanations, the repetitive nature of the "steps" will probably give some more insight on when to do what. (I was a little disappointed that it doesn't go a bit deeper into ACR, at least covering the HSL tab. But it's his method...)
The thing that I find attractive about this book is that, even without explanations, the repetitive nature of the "steps" will probably give some more insight on when to do what. (I was a little disappointed that it doesn't go a bit deeper into ACR, at least covering the HSL tab. But it's his method...)
Duffy
Agreed. Clearly more could be had from HSL (he doesn't touch, so far, the Split Toning or most of the stuff in the latter end of the Develop module and it's ACR equivalent) but despite the sometimes light use of ACR, I've learned a ton. The repetitive nature of the lessons really beat a lot into my head. Between this book and JP Cap.'s Dodging/Burning tutorial at Kelby Training, I'm fearless about layers, blending and the brush tool.
I took another look at the seven point book today, only to see if there was some rhyme or reason to when he used ACR to do a luminosity curve, and when he did curving in PS. I couldn't find any. At one point, he mentions that he is doing the curve in PS because sometimes you don't decide you want extra contrast until you get to PS. That's surely accurate, and honest, but its not a very good recommendation.
I also noticed this time that he relies quite a bit on the preset curves, especially when using the ACR tone curves. This practice might be a useful shortcut, but in general I don't like this sort of one size fits all approach. And again, I couldn't find any explanations on why you would use the preset on one image, but not another.
I've decided my money is probably better spent elsewhere. This looks like it could be a very useful book. But given my past experiences with Kelby, I don't think its worth it to me to work through this book constantly thinking "Why? Why? WHY???"
Back to apply image. As we have seen, it is possible to mix colour modes, making use of both RGB in one file and Lab in another file, blending the two files together using various source and destination channels and blend modes, opacity etc. This is very powerful stuff, I was probably first introduced to this general topic many years ago when Kai Krausse was using ChOps for special effects in his Power Tips, back before layers existed in Photoshop.
The default rendering from the raw image that Pindy posted makes a good example for another apply image move, using the third commonly available colour mode in Photoshop - CMYK.
I have used a similar technique to great effect in a somewhat specialized circumstance, although I believe it has more general applicability. It is similar to what Stephen described, and I got the idea from Chapter 7 of Dan Margulis' "Professional Photoshop (5th Ed.)".
I work with a non-profit that does home repair for elderly indigent in the area, focused on several weekends twice a year. There's a group photo for each house, with all of the volunteers who worked on it and a banner with logos (the non-profit's and the sponsor's). The photos are taken by volunteers (non-professionals) under poor conditions (early and mid afternoon in the sun). They're turned into momentos, so we try to make them as good as possible under the circumstances. In particular, we want the logos to really pop.
The idea is to build a black/white version of the banner which is pretty good, sharpen it, and blend it back into the base version in multiply mode, clearly with a mask and some opacity less than 100%. I've found doing this in CMYK works pretty well for a variety of colored logos on the banners. I usually can't rely on the black channel alone, but have to blend in one or more others into the K channel to get a decent black/white version. Here's an example, with the original on the left.
I can supply more details on the blending, if anyone's interested.
The default rendering from the raw image that Pindy posted makes a good example for another apply image move, using the third commonly available colour mode in Photoshop - CMYK.
It serves to prove an important point, don't think the default renderings of a Raw file are anything close to what you want from the scene referred data and move directly into Photoshop. USE the Raw converter! All those sliders are not placed there for visual balance, they do things that are kind of useful!
Why on earth shoot Raw if you're going to go this route. Shoot 8-bit sRGB JPEG and munch up those poor pixels with a real tool.
The default rendering from the RAW converter will probably be much less satisfactory than a jpg straight from the camera.
Use the Force, Luke!! Why shoot in RAW, if you do not plan to use the RAW converter to your advantage. It would be like driving around in only first gear.
The power of ARC 4.3 is awesome, with shadow/highlights controls and sharpening with a mask and chromatic aberration correction ability. With Smart Objects you can actually use the RAW converter more than once on different parts of the image for even more control.
Comments
I agree with Andrew's take that the biggest difference is in the water in the foreground. This seems to me an artistic choice more than one being superior. There are also differences in the sail cover in the boat in the left rear (and other blues in the shot, which are move evident in Pindy's version), sharpening (Pindy apparently didn't sharpen), and the dock area (and generally highlights in the foreground, where Pindy's version seems to me overdone, saturation-wise).
As Andrew implies, it comes down to a cost-benefit trade-off. What do you want to get out of this particular shot, and how much are you willing to spend on it (not only time to do the post-processing, but time to learn the tools). This is a highly individual thing, and consequently I don't think there's a single answer.
I appreciate your comments (all of you) especially about the over-saturated dock, which I was seeing more as a warming than an over-saturation, though it did go quite red. The sail cover, I noticed in Andrew's version, went way darker than it originally appeared, but all's fair when you're trying to conjure a vision I suppose. I realise there is more I can get from Lightroom, but sometimes for me it's the difference between being able to fix a tool's side effects (think blend modes in PS) and not being able to in LR. I felt on this image, and in particular on photos with a lot of foliage (I have hundreds), that my contrast/tone curve adjustments in LR send to over-saturate and I haven't learned to de-saturate them by other means in a way I find pleasing, including the vibrance and saturation sliders going negative. Looking at Andrew's DNG is enlightening in some ways, which I why I asked for it in the first place. Lots more experimentation is required.
I didn't touch LR.'s sharpening controls, those are the defaults.
And that's my point. You or others would have to ask Dan or Scott why they didn't render the image from the Raw converter closer to the goals instead of using a default rendering (you can see how hugely "off" it is in terms of your rendering goals) then dismiss the Raw converter, not collect $200 and go directly to Photoshop. Least we forget, one of those above actually recommended setting ACR/LR in a default mode for ALL conversions and doing all further work in Photoshop. Damn silly I say.
This is like a darkroom printer not exposing their papers correctly for the neg, then letting the paper sit in a developer for a few hours until it appeared better, then god knows what technique after to make a decent appearing print. Rendering to the print starts in the Raw converter:
http://tinyurl.com/33msxz
Author "Color Management for Photographers"
http://www.digitaldog.net/
His real objection to ACR, then and now, as I understand it, is the curve function is a global function, and that it cannot be isolated for particular colors. In theory, he certainly has a point. In practice, I'm not sure how much difference it makes.
Also, he maintains that by getting the picture in as good a shape as possible, you may be putting yourself into a position where you can't get it into the best possible shape in Photoshop. On this point, I don't know if he's right or not. It makes sense that if you set the endpoints too hard in ACR, then you are depriving yourself of some room to work on contrast with channel specific curves in PS. But I haven't actually run into too many shots where that was really an issue. And I'm also sure that I'm not that good a judge of when those situations arise.
So, Dan may still be holding onto his recommendation with the current version of ACR. But I think its pretty clear that the tide is against this recommendation. Future versions of ACR will probably let you have more control over particular colors, and contrast within those colors, either by curves or by some new device. That should put an end to whatever controversy might exist now over this point.
Duffy
And his current recommendations?
Well Adobe designed these behaviors for a specific reason. I'd suggest this article by Mark Segal:
http://www.luminous-landscape.com/essays/Curves.shtml
Author "Color Management for Photographers"
http://www.digitaldog.net/
I don't know if he's got any current recommendation. I've never heard him address the flexibility the HSL tab now gives. I know that at one point, after the current ACR release, he said that he still though it was lacking, primarily because of the global curve question.
As I recall, you (or another on the color theory list, but I think it was you) issued a correction challenge. I would have liked to see the results of such an event.
One of the problems, of course, with a comparison of these tools, is that so much depends on the sensitivity and skill level of the people using the tools. I have no doubt that Dan could do a better job, using default settings and PS, than I could do with the full compliment of tools. That says more about Dan and me than it does about the tools themselves.
I do know that Dan looks at these sorts of things as like a competitive sport. if someone attended one of his classes and consistently got better results using ACR than Dan could get just using PS, then I'm pretty sure he would come around. I don't know if that has happenned. I'm willing to give it a shot, if someone will front me the money for one of his classes
Another possibility would be to have something like a challenge for a few images here. I think it could be a really good learning experience. One of the complaints that I've had about ACR, even in its new form, is that there is very little showing how to do anything more than the most basic manipulations.
Duffy
When I purchased PSCS3, I didn't really take the time to really learn and appreciate the vast improvement that ARC 4 offered over that in CS2, and just jumped in using my techniques from ARC 3 in CS2. Big, big mistake.
As you and Andrew, both, know very well, it takes a while to fully appreciate all the new controls ( bells and whistles ) in ARC 4.3. It is like getting a P-51 Mustang, after only having a Spad to fly about in.
Jeffe Schewe and Bruce Fraser's Real World "Camera RAW with Adobe Photoshop CS3" is well worth reading more than twice.
I examine each image after it is rendered into a jpg in Photoshop in each of the color channels. Set white and black points, then CTRL-1, CTRL-2, CTRL-3 Bling, bling, bling to find where the contrast lies, and to see if I can utilize that contrast to effect in the image. CTRL-~ and I am back to color viewing all in less than 20 seconds. If I can use the contrast in one of the RGB channels to create more contrast between subject and background, I will then include the Apply Image command in my workflow. If I do not need that separation, I will omit it, but it can be very helpful to create more depth in an image if used subtly and carefully, by using layers and blending modes.
In short, I am not an either/or kind of fellow, but I am willing to try whatever works quickly and easily to help me maximize my images.
But, Adobe Raw Converter 4.3 is my foundation these days.
Moderator of the Technique Forum and Finishing School on Dgrin
That's a challenged that was made to him and ignored.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Do you guys (Duffy and Andrew) believe that there is one and only one "right" way to make corrections and anyone using a different way must learn that they should switch to the "right" way?
Or do you believe that there are many different ways to achieve a desired result and any way that a particular photographer is familiar with and has the time for that gives them the result they are after is perfectly fine?
I ask because it looks like the way this thread is going is that it's simply wrong to do anything in CSx that could have been done in ACR/LR. I will agree that doing as much as possible in LR can offer some workflow efficiency advantages (particularly when all adjustments can be kept in LR and none are needed in CSx).
But, if you focus only on the final quality of one particular image, once you've got the basic tone and white balance in the proper ballpark in the RAW processor, does it really matter to the quality of that final image whether the rest of the fine tuning of the image is done in LR/ACR or CSx?
For example, if you want to emphasize the contrast of one particular element in the photo that's in a mid-tone range, does it matter whether you do that in ACR/LR or in CSx? If you want to darken a sky, does it matter whether you do that in ACR/LR or in CSx?
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
One is totally non destructive, the other isn't.
One is simply a set of metadata instructions and the other is baked and rebaking pixels.
One is fast the other isn't.
One provides a history of every edit ever done, the other doesn't.
One is done high bit, linear encoded data, the other happens (often) in lower bit and gamma corrected color space (on baked pixels).
One provides only global corrections (for the time being) the other doesn't have that restriction.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Yeah, I understand most of those points.
Fast depends upon what you know. There are still several types of corrections that I know how to do faster in CS3 or maybe don't know how to do at all in LR/ACR. Maybe you could do them faster in LR, but fast for me is what I personally know how to do best.
My main reason for going to CS3 these days is for non-global corrections where I use a mask or a blend-if. I'm also much more comfortable with sharpening in CS3, but that could just be due to a learning curve in LR.
The reason I asked this question is I look at all of these as just different tools and different artists prefer different tools. There's no right or wrong tool. Each tool has its advantages and its disadvantages and it's up to the artist to choose the tool that works best for them. There will be "old school" artists who get great results using tools that have been around for awhile and there will be "new school" artists who use the newest tools for the maximum that they can achieve. There will be in-between artisits who know how to do some things in the new tools, but still use the old tools for other things (some of which could be done in the new tools). Each is doing what they're doing because it works best for them. I see no right or wrong in any of the choices.
Is the "new school" artist getting better results? Who knows? It probably depends more on their skill as an artist than it does on which tools they pick.
Is the "new school" artist being more efficient by using the new tools? Maybe, but that also depends on a bunch of other things too. It's possible to spend a ridiculous amount of time in the new tools trying to achieve a result that's really easy in the old tools, so it can work both ways.
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
That's what we're talking about, tools, not art.
What's the right tool for the right job?
Ever tried to use a kitchen knife as a screw driver? Not recommended.
Author "Color Management for Photographers"
http://www.digitaldog.net/
Would you ever consider it acceptable for someone to use LAB mode in CS3 to achieve an editing goal? I'm not asking if you would use it for your own image editing or even if you would ever recommend a LAB mode technique, but would you ever consider it acceptable for someone else to use it who is familiar with how to achieve a result they like in LAB mode?
Homepage • Popular
JFriend's javascript customizations • Secrets for getting fast answers on Dgrin
Always include a link to your site when posting a question
The quote works for me, I may not be all readers - but I am one reader that sadly knows more about this than I would like to know.
This is one of the issues Andrew, your past track record precedes you, in some respects in a negative fashion. Even if this is not what you 'state as your intent', many readers will take away what they will knowing where your preferences lie and the history of your interaction with some other authors, your feelings on various methods and how often you bring them up in other posts on other lists when they are not a current topic or the person is not there to comment or set the record straight with thier side of the story (which is often different from the side you post).
Stephen Marsh
http://members.ozemail.com.au/~binaryfx/
http://members.ozemail.com.au/~binaryfx/
http://prepression.blogspot.com/
Andrew, i think this post was intended for your therapist and not this forum?
The OP rendered the best version that they knew how before using the Kelby Apply Image method. Dan did not encourage Pindy to render a default version from ACR.
Dan was not mentioned in the OP. Dan was mentioned in some follow up replies, as he is famous for being an authority on channel blending.
If I could read 100 consecutive posts of yours that did not mention Dan Margulis or associated subjects I would be very happy (and I am sure that many other viewers would be too).
Stephen Marsh
http://members.ozemail.com.au/~binaryfx/
http://members.ozemail.com.au/~binaryfx/
http://prepression.blogspot.com/
I've asked questions and provided answers. What does the above have to do with the facts of the statements made?
Author "Color Management for Photographers"
http://www.digitaldog.net/
Dan is not the only one that does not like a curve that affects saturation for certain images, even photographers do not always like/wish/want this:
http://21stcenturyshoebox.com/essays/scenereferredworkflow.html
http://21stcenturyshoebox.com/essays/color_reproduction.html
http://21stcenturyshoebox.com/tools/curvetools.html
http://21stcenturyshoebox.com/tools/tonability.html
Those suggesting and or familar with such workflows (Lab, LCH, HSI, HSB etc) know and understand that if one decouples the hue/saturation edits from the tone curve that they will need to be addressed separately and that nobody is suggesting simply using pure tone edits without addressing colour in a separate move. Andrew Rodney from his past LinoColor days should know this as well.
Some think it is optimal to increase saturation, if required/desired.
Some think it is optimal to decrease saturation, if required/desired (ACR/ALR users that have no other option).
Stephen Marsh
http://members.ozemail.com.au/~binaryfx/
http://members.ozemail.com.au/~binaryfx/
http://prepression.blogspot.com/
I suggest something I think much more useful, especially in light of comments on this thread about a need for more workflow recommendations for raw processors.
Dan Margulis is currently advocating a specific workflow for getting pretty good results on a large range of images in a few minutes of work (his so called "Picture Postcard Workflow), which he's demonstrated at Photoshop World and is currently evangelizing though his videos on Kelby Training.
It would be great for someone to propose a different workflow using raw processor tools to get the same effects. This would allow anyone, of (almost) any skill level, to do the test themselves.
For what it's worth, the "Old School" vs "New School" argument doesn't work for me (think String Theory, very New School and very much in trouble these days). What works for me is practics - specific techniques that get much better results or sound theory that explains why better results will follow using techniques that follow this particular theory. Unfortately, Andrew, the theory arguments I've heard aren't that compelling. But first order, it's practics. The above suggestion was made from this perspective.
Kelby's book has you start in ACR or LR and take it as far as you can until (he recommends) that you would want local editing, Lab or anything that PS offers that ACR/LR doesn't handle, such as output sharpening.
Andrew, as you and I have discussed on this forum back when I knew less than a damn thing about RAW conversion, the default rendering in a RAW converter is useless and in no way anything other than a starting-off point. I don't believe anyone's saying it's anything else.
I don't think there is one right way to do things. In general, if a technique gets better results, or gets nearly the same results in less time, then I think its worth learning. And which techniques I will use will depend alot on how much time I have allocated. I used to use LAB techniques all the time for simple and quick fixes on pictures I could only devote a few minutes to. I don't use LAB for that type of image as much anymore since the release of the current ACR.
Stephen:
Your point is well taken. Once the main question in a thread has been answered (as I believe it has here), then I think the continuing discussion is just that. So hijacked threads tend not to bother me so much, but I can see how it can be a concern. Also, since the Kelby Technique that was the foundation of this thread had its genesis (as far as I know) in Dan's work, and was the subject of a chapter in the LAB book, I wasn't at all surprised at the way this thread veered.
JJBong:
I think you stated better what I was trying to get at. On Dgrin, I think it could be really useful to see how people are getting excellent results out of Camera Raw. There just isn't very much of that kind of analysis that I've seen. As for the "Texas Death Match," I would be very curious to see whether I prefered the Photoshop or ACR centered approach in the hands of very highly skilled people.
Duffy
I was looking through Kelby's book about a week ago, and I noticed that he had people do curves in Photoshop, and not in ACR. Am I remembering that right? If so, does he give any explanation why? (I realize that its a Kelby book, so the "whys" might be few and far between, but I thought he might have said something about it.)
Duffy
I was getting worried about the direction this thread was drifting since earlier this afternoon, but it seems to be getting back on track. The field of image editing and processing workflow is filled with numerous individuals whom do not always agree completely in all areas - and I know that is a strong understatement. Hopefully, we can keep that disagreement civil here, even though I know that might be challenging.
I want to thank jjbong for this suggestion. I am a current subscriber to the Picture PostCard videos and hope to understand them better in a few weeks.
I, personally, find that knowing and understanding LAB is a useful skill, but one that I use relatively infrequently with the advent of CS3. LAB can certainly help in evaluating an image for color casts, as Dan describes in his writings and videos.
Moderator of the Technique Forum and Finishing School on Dgrin
One person is! But I'm told he is basing this on an old version of ACR (he doesn't appear to own or admit to owning Lightroom) and what is currently different in his advise is totally unclear.
As to Scott, don't have the book but, the question is not (aside from Dan), dismissing the silliness of setting a Raw converter in default mode and fixing in Photoshop, the real point is, when do we decide and know that we can no longer render the image with the controls provided and must now dive into Photoshop and what are the repercussions of using these two distinct tools as best as we can?
The OP showed an image of a scene using a Lab trick which didn't produce anything at all useful on the image I tried so I asked "what's it do?". Then we saw the final image (Lab) and a default LR rendering (not useful as I pointed out), then a rendering in LR using the controls and skill of the OP. I then used my understanding of LR to get a very close rendering in short order. So if anything, it proves, at least with this one example that we do have the control to render the image totally in a Raw processor to get very close to the other process and I think there are some advantages in both the tools and controls used and the results of the test that so far, haven't been discussed (Stephen would rather dismiss the real meat and potatoes here and talk about how unfortunate this conversation is and probably wishes he had moderator control to censor posts here has he does on the CT list, a list where a discussion, even a challenge like this would never see the light of day).
Different tools and techniques. What do each bring to the party? The bit about "I had to use Lab to produce this color appearance", at least in this case doesn't wash. Now onto the results of the two techniques, what some might feel is best practice, based on image quality, time, flexibility etc. Or maybe we should just censor ideas and just fix everything using older Photoshop techniques. When all you know is a hammer, everything looks like a nail!
Author "Color Management for Photographers"
http://www.digitaldog.net/
He tends to trade off. There are some lessons where you Tone Curve (Point always) and others where you Curves. I remember reading one explanation of why he had you bypass TC in favour of Curves, but I would have to look again to remember it.
In general, he employs Curves much more often in the book as an local editing/adjustment layer/mask/painting kind of thing. For the most part he will get more contrast boost out of black point/white point settings than automatically using Curves or Tone Curve to do a global contrast enhancement, though he will do this in some lessons. I don't disagree with this, as you can get a little too hung up on contrast sometimes.
There were a couple times I wondered why he didn't have you get more out of the TC.
Duffy
Agreed. Clearly more could be had from HSL (he doesn't touch, so far, the Split Toning or most of the stuff in the latter end of the Develop module and it's ACR equivalent) but despite the sometimes light use of ACR, I've learned a ton. The repetitive nature of the lessons really beat a lot into my head. Between this book and JP Cap.'s Dodging/Burning tutorial at Kelby Training, I'm fearless about layers, blending and the brush tool.
I also noticed this time that he relies quite a bit on the preset curves, especially when using the ACR tone curves. This practice might be a useful shortcut, but in general I don't like this sort of one size fits all approach. And again, I couldn't find any explanations on why you would use the preset on one image, but not another.
I've decided my money is probably better spent elsewhere. This looks like it could be a very useful book. But given my past experiences with Kelby, I don't think its worth it to me to work through this book constantly thinking "Why? Why? WHY???"
Duffy
I have used a similar technique to great effect in a somewhat specialized circumstance, although I believe it has more general applicability. It is similar to what Stephen described, and I got the idea from Chapter 7 of Dan Margulis' "Professional Photoshop (5th Ed.)".
I work with a non-profit that does home repair for elderly indigent in the area, focused on several weekends twice a year. There's a group photo for each house, with all of the volunteers who worked on it and a banner with logos (the non-profit's and the sponsor's). The photos are taken by volunteers (non-professionals) under poor conditions (early and mid afternoon in the sun). They're turned into momentos, so we try to make them as good as possible under the circumstances. In particular, we want the logos to really pop.
The idea is to build a black/white version of the banner which is pretty good, sharpen it, and blend it back into the base version in multiply mode, clearly with a mask and some opacity less than 100%. I've found doing this in CMYK works pretty well for a variety of colored logos on the banners. I usually can't rely on the black channel alone, but have to blend in one or more others into the K channel to get a decent black/white version. Here's an example, with the original on the left.
I can supply more details on the blending, if anyone's interested.
It serves to prove an important point, don't think the default renderings of a Raw file are anything close to what you want from the scene referred data and move directly into Photoshop. USE the Raw converter! All those sliders are not placed there for visual balance, they do things that are kind of useful!
Why on earth shoot Raw if you're going to go this route. Shoot 8-bit sRGB JPEG and munch up those poor pixels with a real tool.
Anyone for a hammer or a nail about now? <g>
Author "Color Management for Photographers"
http://www.digitaldog.net/
Use the Force, Luke!! Why shoot in RAW, if you do not plan to use the RAW converter to your advantage. It would be like driving around in only first gear.
The power of ARC 4.3 is awesome, with shadow/highlights controls and sharpening with a mask and chromatic aberration correction ability. With Smart Objects you can actually use the RAW converter more than once on different parts of the image for even more control.
Moderator of the Technique Forum and Finishing School on Dgrin