Luminous Landscape Forum

Raw & Post Processing, Printing => Adobe Lightroom Q&A => Topic started by: pcrov on July 07, 2007, 06:02:57 pm

Title: Curves
Post by: pcrov on July 07, 2007, 06:02:57 pm
I'm not liking lightroom's tone curve tool. Is there any way to use an old-fashioned photoshop-type curve instead?
Title: Curves
Post by: Schewe on July 07, 2007, 06:48:35 pm
No...
Title: Curves
Post by: michael on July 07, 2007, 07:23:09 pm
Beyond Jeff's accurate but somewhat direct response, I would add the following...

Parametric curves in Lightroom take a little while to learn and become comfortable with. But, when you have, especially in conjunction with the Targeted Adjustment Control, they provide unparalleled control and flexibility.

But, in other words – no.

Michael
Title: Curves
Post by: X-Re on July 07, 2007, 07:41:46 pm
I admit freely to not being a Curves Wizard. Frankly, I just don't have the patience!

The Tone Curve in Lightroom makes a lot more intuitive sense to me - so while you can't do direct adjustments to it, I find I get a lot more out of it than I've ever been able to really do in the Photoshop Curves tool...
Title: Curves
Post by: Schewe on July 07, 2007, 10:02:04 pm
Quote
But, in other words – no.
[a href=\"index.php?act=findpost&pid=127048\"][{POST_SNAPBACK}][/a]

That's what I said...

:~)
Title: Curves
Post by: pcrov on July 11, 2007, 03:35:56 am
I guess it's time to learn curves all over again.
Title: Curves
Post by: digitaldog on July 11, 2007, 11:54:05 am
The 'problem' is often the user, not LR.'s curves. In the past (and based on who's books you read), you are given the impression that the first stop for any image correction is to go directly to curves and not pass go. But LR isn't Photoshop and shouldn't be looked at in such a light. LR.'s tone and color controls are arranged in a fixed order. Now you don't have to use them this way but going outside this route can lead to issues that cause you to spend more time chasing your tail then getting work done. Bottom line is, the curves are useful for very fine tune adjustments and I find that I often don't even need to use them. They are useful for tiny tone tweaks you may not even need. Do all the heavy lifting using the Basic controls.
Title: Curves
Post by: pcrov on July 11, 2007, 12:29:14 pm
Quote
The 'problem' is often the user, not LR.'s curves.
[a href=\"index.php?act=findpost&pid=127598\"][{POST_SNAPBACK}][/a]

Well what I wasn't liking was the method of altering the curves - the interface itself. I don't know how that makes me the problem.

In any case it's pretty lame to blame the user when they're having trouble with a piece of software.
Title: Curves
Post by: digitaldog on July 11, 2007, 12:45:14 pm
Quote
Well what I wasn't liking was the method of altering the curves - the interface itself. I don't know how that makes me the problem.

In any case it's pretty lame to blame the user when they're having trouble with a piece of software.
[a href=\"index.php?act=findpost&pid=127603\"][{POST_SNAPBACK}][/a]

The Parametric curves don't allow you to alter them to hose the image which is good, especially when you understand that you're doing fine tweaks.

There's a pencil option in Photoshop's curves that allow you to severely Posterize the image and you can of course do this using the standard point curves if you pull them too much. That's not possible in LR and as yet, I don't see why anyone would want to do this.

You can't pull individual color channels in LR.'s curves because again, this isn't the place to be doing such work.

LR's Parametric curves are for fine tweaks, end of story.

Now the question that I'd find reasonable would be, if there is point curves in ACR, why not LR? Considering that it appears the good people at Adobe are trying for parity between the two, its a reasonable question. Lose them in ACR or add them in LR, I'm OK with either.
Title: Curves
Post by: Jack Flesher on July 11, 2007, 12:48:36 pm
Quote
Well what I wasn't liking was the method of altering the curves - the interface itself. I don't know how that makes me the problem.

In any case it's pretty lame to blame the user when they're having trouble with a piece of software.
[a href=\"index.php?act=findpost&pid=127603\"][{POST_SNAPBACK}][/a]

I'm with you pcrov... And it's one of the (many) reasons I don't like or use LR.

Adobe spends what, the last 10 years having us do curves one way in Photoshop? We learn it and get accustomed and proficient with one way, then all of a sudden they expect us all to fall in love with parametric sliders?   Sorry, I like the old way, the ability to ctrl/cmd-click a point onto the curve and adjust it. IMO, many improvements aren't...

Cheers,
Title: Curves
Post by: digitaldog on July 11, 2007, 12:53:44 pm
Quote
Are you seriously telling me I'm wrong for preferring an interface I've used for years? That it's my problem for not understanding the LR workflow and the role of curves in it?

For fuck's sake, this conversation was already over before you decided to jump in with this bullshit.
[a href=\"index.php?act=findpost&pid=127614\"][{POST_SNAPBACK}][/a]

You're incorrect in assuming that editing gamma corrected images in Photoshop is anything like rendering linear gamma raw data. This is a problem I've seen with some so called Photoshop guru's who look at raw modules as if they were Photoshop correction tools. If all you know is a hammer, everything appears like a nail. IF you're rendering raw data (not using LR or ACR to edit existing rendered gamma corrected images which I think is silly), then you have to look at this tool in a different light.

Yes, you're wrong and you don't really know how to communicate and behave in a public forum, at least based on your poor choice of language.
Title: Curves
Post by: pcrov on July 11, 2007, 12:53:53 pm
Quote
Adobe spends what, the last 10 years having us do curves one way in Photoshop? We learn it and get accustomed and proficient with one way, then all of a sudden they expect us all to fall in love with parametric sliders?   Sorry, I like the old way, the ability to ctrl/cmd-click a point onto the curve and adjust it. IMO, many improvements aren't...

Cheers,
[a href=\"index.php?act=findpost&pid=127613\"][{POST_SNAPBACK}][/a]

Exactly. I'm sure the parametric sliders are just lovely, but the photoshop-style curves tool is second-nature to me. It's going to be hard to adjust.
Title: Curves
Post by: DarkPenguin on July 11, 2007, 12:54:57 pm
what an odd reaction.
Title: Curves
Post by: digitaldog on July 11, 2007, 12:56:56 pm
Quote
Go fuck yourself.

Is that better?
[a href=\"index.php?act=findpost&pid=127617\"][{POST_SNAPBACK}][/a]

Yes, much better at illustrating your lack of decorum and intelligence.
Title: Curves
Post by: pcrov on July 11, 2007, 01:00:14 pm
Why are you even here? Did I insult your favorite new toy or something?
Title: Curves
Post by: digitaldog on July 11, 2007, 01:05:13 pm
Quote
Why are you even here? Did I insult your favorite new toy or something?
[a href=\"index.php?act=findpost&pid=127621\"][{POST_SNAPBACK}][/a]

No.
Title: Curves
Post by: stewarthemley on July 11, 2007, 01:42:44 pm
PCROV Please be aware that many people will be offended by your choice of language on this public forum. My friend's children (quite young) are learning about photography and often visit. I don't see why you felt you had been offended and also why you had to respond as you did. It would be a sign of maturity if you first edited (ie removed) your most offensive posts and then issued an apology.
Title: Curves
Post by: pcrov on July 11, 2007, 01:46:06 pm
I don't take kindly to self-important assholes telling me my *opinion* is wrong and implying I'm somehow stupid or ignorant for having it. I'll not be editing my posts or apologizing to anyone.
Title: Curves
Post by: Schewe on July 11, 2007, 03:13:55 pm
Quote
I don't take kindly to self-important assholes telling me my *opinion* is wrong and implying I'm somehow stupid or ignorant for having it. I'll not be editing my posts or apologizing to anyone.
[a href=\"index.php?act=findpost&pid=127629\"][{POST_SNAPBACK}][/a]


Well, I won't imply ANYTHING...you ARE stupid and ignorant for both your opinion and your behavior and your attitude. You may THINK you have the right to do or say anything you wish, but you don't. Not if you wish to maintain the honor of posting in this group of people.

You need not reply to this message (or you can if you wish) I have chosen to ignore you as a user which means nothing you post will be visible to me-regardless of what you may write. It's a form of banishment open to us in these forums-although truth be told I suspect when Michael sees this thread and your behavior and language, I figure he'll do a bit more.

See ya, have a nice life (and maybe think about changing your meds or getting off caffiene).
Title: Curves
Post by: Mark D Segal on July 11, 2007, 04:16:42 pm
pcrov: I have no role moderating, but from past experience I would simply observe there are rules regarding behaviour on this website, and there are consequences for not respecting them.
Title: Curves
Post by: X-Re on July 11, 2007, 05:11:51 pm
Quote
Is that better?
[a href=\"index.php?act=findpost&pid=127617\"][{POST_SNAPBACK}][/a]

     Take a hike, bub.
Title: Curves
Post by: michael on July 11, 2007, 06:06:43 pm
The individual has now been deleted and banned.

Michael
Title: Curves
Post by: Jack Flesher on July 11, 2007, 06:16:11 pm
Quote
IF you're rendering raw data (not using LR or ACR to edit existing rendered gamma corrected images which I think is silly), then you have to look at this tool in a different light.
[a href=\"index.php?act=findpost&pid=127615\"][{POST_SNAPBACK}][/a]

Andrew:

I hear you on rendering raw, BUT...  ACR has a curves point mode in addition to the sliders -- and thus I have a good old fashioned curves type interface in RAW that I can place points on. This option is missing in LR...  My only point was I prefer that type of adjustment freedom (and I submit higher precision) to the parametric sliders.  Moreover,  it's what I'm used to doing.  (Not sure why the other guy went postal, but I thought that was his point too...)

Cheers,
Title: Curves
Post by: JeffCharles on July 11, 2007, 06:42:42 pm
Thank you for the ban.
Title: Curves
Post by: Mark D Segal on July 11, 2007, 06:58:23 pm
Quote
The individual has now been deleted and banned.

Michael
[a href=\"index.php?act=findpost&pid=127675\"][{POST_SNAPBACK}][/a]

Good. And thanks for telling us - it will serve as a public future reference point in case other such people join the site and behave likewise.
Title: Curves
Post by: Mark D Segal on July 11, 2007, 07:08:35 pm
Quote
Andrew:

I hear you on rendering raw, BUT...  ACR has a curves point mode in addition to the sliders -- and thus I have a good old fashioned curves type interface in RAW that I can place points on. This option is missing in LR...  My only point was I prefer that type of adjustment freedom (and I submit higher precision) to the parametric sliders.  Moreover,  it's what I'm used to doing.  (Not sure why the other guy went postal, but I thought that was his point too...)

Cheers,
[a href=\"index.php?act=findpost&pid=127676\"][{POST_SNAPBACK}][/a]

Hi Jack,

OK, now that this thread is back to sanity we can discuss technicalities in peace and harmony. I agree with your observation that the point curve is extremely useful, it's what we know best, and I applaud Thomas Knoll that it was retained in Adobe Camera Raw. I also applaud the Adobe team on the development of the Parametric Curve which appears in both Lightroom and in Camera Raw. I also agree it would be helpful if the point curve were also included in Lightroom as it is in Camera Raw. When I attended a Lightroom seminar at PhotoshopWorld in Boston this past April I met a number of the key players on the Lightroom team and I made this recommendation to them. Of course many people recommend all kinds of things to them, and they must decide whether a point curve is a good philosophical fit for their concept of what Lightroom is supposed to be about.

At first I was a bit skeptical about the usefulness of this parametric curve, because I was forever frustrated that I could not detach the end points from their anchors - that is really the main hiccup with it. But I must report, now that I have played with this parametric curve ALOT, I have really become a convert - don't get me wrong - I still want my point curve, but this tool is GOOD. It's a very efficient way of doing zonal-type adjustments on images - as a complement if needed after you finish with Exposure, Recovery, Fill Light and Blacks in the Basic Tab. Once you combine the adjustability of the zonal demarcators with the adjustability of the four portions of the Curve itself, you can get a surprising amount of very refined and powerful control out of this system. I'm developing techniques for setting it which in some ways surpass the flexibility of the point curve in terms of acceptable impacts on images.

Cheers,

Mark
Title: Curves
Post by: Schewe on July 11, 2007, 07:18:05 pm
Quote
I applaud Thomas Knoll that it was retained in Adobe Camera Raw. [a href=\"index.php?act=findpost&pid=127685\"][{POST_SNAPBACK}][/a]

It had to be retained for backwards compatibility reasons. It should also be noted that Lightroom 1/1.1 CAN render a point curve, you just can't actually edit the point curve.

However, if you'all want the point curve editor in Lightroom, you would do better posting in the Feature Request forum for Lightroom and be prepared to make a use case why certain edits HAVE to be done with points and why the lack of points in Lightroom is causing less than optimal processing results. That is the only real factor with weight.

Merely saying you LIKE points better than parametric or you WANT point editing won't get any traction, what so ever. You must make a case and prove it. Fact is, most curves CAN be done quicker and easier with parametric than points–if you know what you're doing. There are only a very few things that can't be done-like editing extreme highlights to tease textural detail out. That's an example...and the engineers already know that...but it would be useful to find more use cases to strengthen the case.

P.S. Mike, thanks for making butthead go bye bye...of course now that's he listed as "registered" his darn vulgar posts reappeared. :~(
Title: Curves
Post by: Mark D Segal on July 11, 2007, 07:59:18 pm
Quote
It had to be retained for backwards compatibility reasons. It should also be noted that Lightroom 1/1.1 CAN render a point curve, you just can't actually edit the point curve.

However, if you'all want the point curve editor in Lightroom, you would do better posting in the Feature Request forum for Lightroom and be prepared to make a use case why certain edits HAVE to be done with points and why the lack of points in Lightroom is causing less than optimal processing results. That is the only real factor with weight.

Merely saying you LIKE points better than parametric or you WANT point editing won't get any traction, what so ever. You must make a case and prove it. Fact is, most curves CAN be done quicker and easier with parametric than points–if you know what you're doing. There are only a very few things that can't be done-like editing extreme highlights to tease textural detail out. That's an example...and the engineers already know that...but it would be useful to find more use cases to strengthen the case.

P.S. Mike, thanks for making butthead go bye bye...of course now that's he listed as "registered" his darn vulgar posts reappeared. :~(
[a href=\"index.php?act=findpost&pid=127689\"][{POST_SNAPBACK}][/a]

Long live backward compatibility!

It's not a matter of "liking it better". It happens to be a very efficient and effective way of achieving higher contrast and brightness at the same time with a distinctly smooth gradiant of luminosity change that you can get from a linear more easily than from a curvilinear transformation. I had hoped at least one or more of the huge number of beta testers and perhaps an alpha or two would have recommended that to them, and perhaps did. (I had a long delay getting into Lightroom Beta because I'm on Windows, then soon after the CS3 Beta came out, so I opted for that.) The feedback I got to my suggestion for the point curve at PSW indicated quite clearly that not including it in LR was a deliberate choice to "keep the program simple" - so it's not that they don't know the merits of it, they consciously decided not to include it. Well, they are the developers - their prerogative. Maybe if they get enough recommendations from enough customers all saying the same thing they'll rethink that decision. Clearly it's no big deal to just drop it in - the whole thing is there already in Camera Raw.

(edited)

Cheers,

Mark
Title: Curves
Post by: Jack Flesher on July 11, 2007, 08:11:38 pm
Quote
There are only a very few things that can't be done-like editing extreme highlights to tease textural detail out. That's an example...and the engineers already know that...but it would be useful to find more use cases to strengthen the case.
[a href=\"index.php?act=findpost&pid=127689\"][{POST_SNAPBACK}][/a]

I hear you Jeff, but maintaining detail in highlights is an area I personally pay lots of attention to, especially during the raw conversion;  I expend enough energy to make sure they're not blown during capture, that I certainly want to be able to tweak them without blowing them during conversion. The upper end of the highlights is probably not as important for folks posting mainly to the web -- a LR strength -- but IMO it makes a difference in obtaining the optimal tonal balance in a print.   I am often asked how I get the look I get in my images (web and print) by folks who know how to process, and I suspect the question is at least partially answered by what I do with my highlights.

Mark: If I could add another slider or two to the parametric curve so I had more control in the highlights without giving it up in the shadows, I might be more enthused with the tool, but even then I have doubts they would give me the discrimination I want in that adjustment...

Cheers,
Title: Curves
Post by: Mark D Segal on July 11, 2007, 08:35:40 pm
Quote
IMark: If I could add another slider or two to the parametric curve so I had more control in the highlights without giving it up in the shadows, I might be more enthused with the tool, but even then I have doubts they would give me the discrimination I want in that adjustment...

Cheers,
[a href=\"index.php?act=findpost&pid=127698\"][{POST_SNAPBACK}][/a]

Jack,

I find that by very careful placement of the first zone placer to the right on the x-axis of the dialogue box, and then iterating between adjusting it as well as the Highlights and Lights sliders themselves, one can achieve quite subtle control of what happens to real highlights, versus "Lights". As I'm sure you've discovered already, one really should play with those zonal markers to make the most of that tool. By the way, you don't give up a thing in the shadows by playing with the Lights and the Highlights, especially if watching the placement of the zone marker between Darks and Shadows.
Title: Curves
Post by: Jack Flesher on July 11, 2007, 10:18:39 pm
Quote
Jack,

I find that by very careful placement of the first zone placer to the right on the x-axis of the dialogue box, and then iterating between adjusting it as well as the Highlights and Lights sliders themselves, one can achieve quite subtle control of what happens to real highlights, versus "Lights". As I'm sure you've discovered already, one really should play with those zonal markers to make the most of that tool. By the way, you don't give up a thing in the shadows by playing with the Lights and the Highlights, especially if watching the placement of the zone marker between Darks and Shadows.
[a href=\"index.php?act=findpost&pid=127699\"][{POST_SNAPBACK}][/a]

I've played with it a bit, but admit I am not even close to proficient...  Explain to me how I would increase contrast in the highlights while keeping the maximum brightness at 97% (roughly 248) and at the same time not affecting any pixels below say 87% (roughly 222) using the parametric controls.  

Assume at the same time I want to add an inverse adjustment (lower contrast but maintain certain border values) in the shadows.  

Finally, suppose once the above are done, I might want to boost contrast slightly throughout the middle range from say 25% through 75%, and increase brightness slightly, but of course without significantly altering either of the earlier adjustments.

I'm sure it can be done and moreover understand a "similar" more general effect could be achieved, but I cannot see how to do it with precision using the slider interface.  Yet in curves it is accomplished pretty easily with good precision.  

Cheers,
Title: Curves
Post by: Schewe on July 11, 2007, 10:20:37 pm
Quote
The feedback I got to my suggestion for the point curve at PSW indicated quite clearly that not including it in LR was a deliberate choice to "keep the program simple" - so it's not that they don't know the merits of it, they consciously decided not to include it.
[a href=\"index.php?act=findpost&pid=127697\"][{POST_SNAPBACK}][/a]

You THINK they said "keep the program simple", what they were REALLY saying is that it didn't sound mission critical so they chose not to spend the engineering time required to engineer a UI into Lightroom to be able to edit points because they had so many other things they couldn't afford the time. You need to be able to read the context (engineer speak).

Will Lightroom get the UI to edit points?

I'll bet it will at some "point" :~) Not that hard since the code is already there, just no UI. Will it happen sooner vs later? The more people that post on the Feature Request forum asking for it and giving strong use cases the more likely it will to be soon than later.
Title: Curves
Post by: Jack Flesher on July 11, 2007, 10:29:55 pm
Quote
You THINK they said "keep the program simple", what they were REALLY saying is that it didn't sound mission critical [a href=\"index.php?act=findpost&pid=127712\"][{POST_SNAPBACK}][/a]

At least the sliders are more or less idiot-proof    I see it akin to having a stereo with bass, treble and volume adjustments, versus having a component system with a graphic equalizer.  I suspect from a business POV, it gets the job done "well enough" for 80% of the users, and for management, that's "good enough".

I liked Adobe better when Lamkin was there -- at least I had his ear on some of this stuff  

Cheers,
Title: Curves
Post by: Mark D Segal on July 11, 2007, 10:33:27 pm
Quote
At least the sliders are more or less idiot-proof    I see it akin to having a stereo with bass, treble and volume adjustments, versus having a component system with a graphic equalizer.  I suspect from a business POV, it gets the job done "good enough" for 80% of the users, and for management, that's "good enough".

Cheers,
[a href=\"index.php?act=findpost&pid=127714\"][{POST_SNAPBACK}][/a]

Actually, the sliders are NOT idiot proof. Use them badly and you can mess-up an image just as badly as you would with a traditional Curve  
Title: Curves
Post by: Mark D Segal on July 11, 2007, 10:44:02 pm
Quote
I've played with it a bit, but admit I am not even close to proficient...  Explain to me how I would increase contrast in the highlights while keeping the maximum brightness at 97% (roughly 248) and at the same time not affecting any pixels below say 87% (roughly 222) using the parametric controls. 

Assume at the same time I want to add an inverse adjustment (lower contrast but maintain certain border values) in the shadows. 

Finally, suppose once the above are done, I might want to boost contrast slightly throughout the middle range from say 25% through 75%, and increase brightness slightly, but of course without significantly altering either of the earlier adjustments.

I'm sure it can be done and moreover understand a "similar" more general effect could be achieved, but I cannot see how to do it with precision using the slider interface.  Yet in curves it is accomplished pretty easily with good precision. 

Cheers,
[a href=\"index.php?act=findpost&pid=127711\"][{POST_SNAPBACK}][/a]

For the highlight portion, you would set the rightmost zone marker far enough to the left to isolate only the highlights whose contrast you want to increase. Then you would steepen that portion of the curve with the Highlight slider.

For the shadows, likewise you would set the left most point far enough to the left to isolate just the shadow areas, and adjust the shadows slider to make it flatter.

For the middle range, you would then steepen the "Darks" and "Lights" portions, experimenting how much for each and where to place the middle zone marker to delineate the Lights from the Darks. You can have some quite refined control over the mid-tones - for example, suppose you Lighten the Lights and Darken the darks, or you Lighten both but the Darks by less than the Lights. Then by sliding that middle zone marker to the left the mid-tones darken, and by sliding it to the right they lighten. Etc., etc.
Title: Curves
Post by: Jack Flesher on July 11, 2007, 10:51:56 pm
Quote
For the highlight portion, you would set the rightmost zone marker far enough to the left to isolate only the highlights whose contrast you want to increase. Then you would steepen that portion of the curve with the Highlight slider.

For the shadows, likewise you would set the left most point far enough to the left to isolate just the shadow areas, and adjust the shadows slider to make it flatter.

For the middle range, you would then steepen the "Darks" and "Lights" portions, experimenting how much for each and where to place the middle zone marker to delineate the Lights from the Darks. You can have some quite refined control over the mid-tones - for example, suppose you Lighten the Lights and Darken the darks, or you Lighten both but the Darks by less than the Lights. Then by sliding that middle zone marker to the left the mid-tones darken, and by sliding it to the right they lighten. Etc., etc.
[a href=\"index.php?act=findpost&pid=127718\"][{POST_SNAPBACK}][/a]

Yeah, I get that -- as I said I could generate a similar effect... But can I lock down the specific end values as I listed them for the highlights?  I think not.
Title: Curves
Post by: Mark D Segal on July 11, 2007, 11:40:25 pm
Quote
Yeah, I get that -- as I said I could generate a similar effect... But can I lock down the specific end values as I listed them for the highlights?  I think not.
[a href=\"index.php?act=findpost&pid=127721\"][{POST_SNAPBACK}][/a]

Jack, it's a different kind of thinking. The language of "locking down values" is more applicable to traditional point curves - which I too still like very much.
My version of the LR interface doesn't even give me those composite Luminosity values to reference (or at least I haven't seen where). Where does one see when you are at luminosity 87%? (I only see individual R G B values on LR-1 for Windows.) I think luminosity in the LR interface is meant to be used more visually and less by specific numerical luminosity values as in the traditional way.
Title: Curves
Post by: Jack Flesher on July 11, 2007, 11:52:38 pm
Quote
I think luminosity in the LR interface is meant to be used more visually and less by specific numerical luminosity values as in the traditional way.
[a href=\"index.php?act=findpost&pid=127728\"][{POST_SNAPBACK}][/a]

Yep, I got that... Reminds of an old ad... "With the Bose radio you don't need any complicated dials or settings because our professional audiophiles have pre-balanced the radio to perfection for you!  Simply adjust the volume to your desired level and enjoy the music!"   ~~~  Want some more Kool-Aid?

All kidding aside, if LR works as-is for you, great.  Me, I want more flexibility and repeatable precision.

Cheers,
Title: Curves
Post by: BernardLanguillier on July 12, 2007, 12:53:04 am
Quote
However, if you'all want the point curve editor in Lightroom, you would do better posting in the Feature Request forum for Lightroom and be prepared to make a use case why certain edits HAVE to be done with points and why the lack of points in Lightroom is causing less than optimal processing results. That is the only real factor with weight.

Merely saying you LIKE points better than parametric or you WANT point editing won't get any traction, what so ever. You must make a case and prove it. Fact is, most curves CAN be done quicker and easier with parametric than points–if you know what you're doing. There are only a very few things that can't be done-like editing extreme highlights to tease textural detail out. That's an example...and the engineers already know that...but it would be useful to find more use cases to strengthen the case.
[a href=\"index.php?act=findpost&pid=127689\"][{POST_SNAPBACK}][/a]

Jeff,

I understand the constraints of software developement.

But... overlooking the "want" to focus on the "need" will not capture the fact that if I don't get what I "want" I'll probably decide at some point of time to use a piece of software that does give me what I "want".

Regards,
Bernard
Title: Curves
Post by: macgyver on July 12, 2007, 01:07:02 am
Quote
Jeff,

I understand the constraints of software developement.

But... overlooking the "want" to focus on the "need" will not capture the fact that if I don't get what I "want" I'll probably decide at some point of time to use a piece of software that does give me what I "want".

Regards,
Bernard
[a href=\"index.php?act=findpost&pid=127737\"][{POST_SNAPBACK}][/a]


That is the crux of the matter; I keep hearing from software companies what amounts to "we know you don't like it but its better this way, get used to it" with little or no thought to what the consumer actually desires.  It seems that many of these companies would benifit in taking a basic business course from their local community college.  But, then again, I doubt adobe will be losing money anytime soon because of something like this.
Title: Curves
Post by: Schewe on July 12, 2007, 02:22:23 am
Quote
But... overlooking the "want" to focus on the "need" will not capture the fact that if I don't get what I "want" I'll probably decide at some point of time to use a piece of software that does give me what I "want".
[a href=\"index.php?act=findpost&pid=127737\"][{POST_SNAPBACK}][/a]

And that's your right. Look, the Lightroom engineers can only do what they can do. LR 1.1 was required for a variety of bug fixes and functionality enhancements-which meant that that other work went undone. What current Lightroom feature would you give up in exchange for point editing? Prolly not mergable catalogs, prolly not sharpening, prolly not a lot that is in 1.1 that ain't in 1.0.

Want will always take a backseat to need. So, unless you can elevate a want to a need, it will remain optional as apposed to mission critical.

Can you get really great output now from Lightroom? Would a point editor make a big difference? Prove it. Try to find use cases where Camera Raw can do a substantially better job than Lightroom.
Title: Curves
Post by: BernardLanguillier on July 12, 2007, 03:07:32 am
Quote
And that's your right. Look, the Lightroom engineers can only do what they can do. LR 1.1 was required for a variety of bug fixes and functionality enhancements-which meant that that other work went undone. What current Lightroom feature would you give up in exchange for point editing? Prolly not mergable catalogs, prolly not sharpening, prolly not a lot that is in 1.1 that ain't in 1.0.

Want will always take a backseat to need. So, unless you can elevate a want to a need, it will remain optional as apposed to mission critical.

Can you get really great output now from Lightroom? Would a point editor make a big difference? Prove it. Try to find use cases where Camera Raw can do a substantially better job than Lightroom.
[a href=\"index.php?act=findpost&pid=127750\"][{POST_SNAPBACK}][/a]

Jeff,

As you might remember, I was not too enthousiastic about LR in the past for various reasons having to do with the need to import data and the lack of opening to other raw converters (I see that DxO is now proposing an intregration which is great).

This being said, I have to admit that the quality of the conversions in 1.1 is now excellent and have to acknowledge the fact that it is possible to get great results with 1.1. It works especially well with my ZD files as long as the exposures are reasonnably short.

Now, the main reasons why I think I need point curves in Lightroom nonetheless are:

- point control enables me to increase the local contrast of an image exactly where it has to be done based on the actual histogram. The current sliders work on pre-defined areas of the curve that do not always meet the actual needs of an image.

How about those high key images that do not have a lot of information in the lower 25% of the histogram but where more contrast is needed elsewhere for instance?

Since this is currently not easy to do - or at least I haven't found an easy way to do it - you often end up having to apply more global curves inside PS, which defeats the very purpose of LR as a comprehensive global image modification engine. Multiple curve application can only increase the entropy of the image and should be avoided.

- highlight control - we all know that those brightest bits can often make or break an image, especially those shot with natural light. I believe that this has been mentioned by others also,

- consistency with ACR. I am not yet sold on the idea that I need to batch import all my images in Lightroom, especially all the legacy ones. Using ACR on those is much faster and I would typically use point control for those images. I'd like to be able to stick the same workflow for those new images I'll handle with Lightroom,

- consistency with other RAW converters like Capture NX,

- I am used to working with point curve.

Regards,
Bernard
Title: Curves
Post by: Hermie on July 12, 2007, 03:32:25 am
Quote
Jeff,

As you might remember, I was not too enthousiastic about LR in the past for various reasons having to do with the need to import data and the lack of opening to other raw converters (I see that DxO is now proposing an intregration which is great).

This being said, I have to admit that the quality of the conversions in 1.1 is now excellent and have to acknowledge the fact that it is possible to get great results with 1.1. It works especially well with my ZD files as long as the exposures are reasonnably short.

Now, the main reasons why I think I need point curves in Lightroom nonetheless are:

- point control enables me to increase the local contrast of an image exactly where it has to be done based on the actual histogram. The current sliders work on pre-defined areas of the curve that do not always meet the actual needs of an image.

How about those high key images that do not have a lot of information in the lower 25% of the histogram but where more contrast is needed elsewhere for instance?

Since this is currently not easy to do - or at least I haven't found an easy way to do it - you often end up having to apply more global curves inside PS, which defeats the very purpose of LR as a comprehensive global image modification engine. Multiple curve application can only increase the entropy of the image and should be avoided.

- highlight control - we all know that those brightest bits can often make or break an image, especially those shot with natural light. I believe that this has been mentioned by others also,

- consistency with ACR. I am not yet sold on the idea that I need to batch import all my images in Lightroom, especially all the legacy ones. Using ACR on those is much faster and I would typically use point control for those images. I'd like to be able to stick the same workflow for those new images I'll handle with Lightroom,

- consistency with other RAW converters like Capture NX,

- I am used to working with point curve.

Regards,
Bernard
[a href=\"index.php?act=findpost&pid=127754\"][{POST_SNAPBACK}][/a]

Wouldn't you just love the fine control of LightZone's ZoneMapper in Lightroom :-) ?

Adobe should acquire Lightcrafts and use its nifty tooling like ZoneMapper, Re-light and especially its selection tools.

Herman
Title: Curves
Post by: Mark D Segal on July 12, 2007, 08:23:35 am
Quote
Jeff,

As you might remember, I was not too enthousiastic about LR in the past for various reasons having to do with the need to import data and the lack of opening to other raw converters (I see that DxO is now proposing an intregration which is great).

This being said, I have to admit that the quality of the conversions in 1.1 is now excellent and have to acknowledge the fact that it is possible to get great results with 1.1. It works especially well with my ZD files as long as the exposures are reasonnably short.

Now, the main reasons why I think I need point curves in Lightroom nonetheless are:

- point control enables me to increase the local contrast of an image exactly where it has to be done based on the actual histogram. The current sliders work on pre-defined areas of the curve that do not always meet the actual needs of an image.

How about those high key images that do not have a lot of information in the lower 25% of the histogram but where more contrast is needed elsewhere for instance?

Since this is currently not easy to do - or at least I haven't found an easy way to do it - you often end up having to apply more global curves inside PS, which defeats the very purpose of LR as a comprehensive global image modification engine. Multiple curve application can only increase the entropy of the image and should be avoided.

- highlight control - we all know that those brightest bits can often make or break an image, especially those shot with natural light. I believe that this has been mentioned by others also,

- consistency with ACR. I am not yet sold on the idea that I need to batch import all my images in Lightroom, especially all the legacy ones. Using ACR on those is much faster and I would typically use point control for those images. I'd like to be able to stick the same workflow for those new images I'll handle with Lightroom,

- consistency with other RAW converters like Capture NX,

- I am used to working with point curve.

Regards,
Bernard
[a href=\"index.php?act=findpost&pid=127754\"][{POST_SNAPBACK}][/a]

Bernard,

While I hear Jeff loud and clear (no way you can't)    and he has a point (or two) I still find myself firmly within the group who "wants" point curves. Do I really "need" them? - yes and no - I'm moving from being less convinced of yes to not convinced of no - if you see what I mean. So while I still "want" the Point Curve, at the same time I recognize the value and potential of the Parametric approach.

You say above the current sliders work on pre-defined areas of the Curve. This is not necessarily quite correct, depending on what you mean by "pre-defined" - you can re-define those areas by shifting the zone demarcation sliders on the x-axis of the Curve U.I. box. In fact you have seven controls in that U.I. - the four sliders below the box and three markers immediately under the box. You can work back and forth between them to shape that Curve with remarkable effectiveness. I've processed over a thousand images over the past several months using the new Camera Raw (same thing as LR Develop except ACR does have the point curve in a separate tab), and I find myself needing the Point Curve less and less - but when I need it, I do need it.

This is not because I absolutely can't do with Parametric what I can do with Point, but because the distinction between "want" and "need" is less clear-cut than Jeff's posts would imply. There is also a consideration of "time-effectiveness" in all this. I can probably get 90% of what I "need" from a Parametric Curve quite easily. The other 10% I could probably get as well - with much more time spent fiddling with sliders rather than plunking down several points. I would also like to see composite luminosity, Lab and HSB data read-outs in an expanded info palette. Maybe this needs to be a "feature request".

I agree with Jeff that there are trade-offs between priorities, especially when a company - which is after all a for profit enterprise (absent which we wouln't have all this stuff) feels it's under commercial pressure to put a product on the market by a certain date. That said, the Point Curve add-in is probably one of the easiest and fastest things they could do, because all the technology for it is sitting right under their own roof in Camera Raw. OK, says me who isn't a software engineer - and I know everything takes time even if it's been done before - but common sense would suggest this is not a monumental development problem for them. And I'm still not convinced that leaving this out of Lightroom was a matter of this priority versus that. I do believe that I heard and interpreted correctly what I was told: they wanted to keep LR "simple" - well - it ain't all that simple, but..............that's what they "wanted" - and felt they "needed".

Bernard, your last point - "I am used to working with Point curve" - yah, so am I, so is Jack, so is everyone. I don't know how old you guys are, but I'm getting on up there, and I still like to believe that old dogs can learn new tricks. But they'd best be better ones!  

Cheers,

Mark
Title: Curves
Post by: digitaldog on July 12, 2007, 08:35:07 am
Given the fact that ACR has point curve and given the fact (well it may not be a fact but it appears to be true) that Adobe is trying to produce feature parity between LR and ACR with minor exceptions, and given the fact that users appear to want point curves, I think it should be easily accomplished should enough people make their voices heard. It seems obvious that Thomas Knoll feels point curves are necessary and useful. Now you can import the point curve settings from ACR into LR but man, that's a kludge.

Unlike Jeff, I don't know the rational or have heard discussions of the two teams lead engineers but it appears from the outside as if the two team leaders don't necessarily agree in the need for point curve. More the reason why it would be useful to provide examples of images using BOTH products to illustrate your points to the LR group.
Title: Curves
Post by: Jack Flesher on July 12, 2007, 10:06:10 am
Quote
More the reason why it would be useful to provide examples of images using BOTH products to illustrate your points to the LR group.
[a href=\"index.php?act=findpost&pid=127782\"][{POST_SNAPBACK}][/a]

OR we can choose to use -- and pay for -- a tool that gives us what we want in lieu of a tool that doesn't, even though some third party tells us it has everything we need... (AKA, "drink their Kool-Aid!")  Here I use the word "want" specifically as respects my purchasing criteria.  

Cheers,
Title: Curves
Post by: Fred Ragland on July 12, 2007, 10:15:40 am
Quote
...it would be useful to provide examples of images using BOTH products to illustrate your points to the LR group...
[a href=\"index.php?act=findpost&pid=127782\"][{POST_SNAPBACK}][/a]
This has been a very interesting discussion.  Jack, could you show us how we can benefit from point curves in post processing?  And could others show us how that can be done in LR?

Thank you all.

Fred
Title: Curves
Post by: Mark D Segal on July 12, 2007, 10:25:54 am
Quote
OR we can choose to use -- and pay for -- a tool that gives us what we want in lieu of a tool that doesn't, even though some third party tells us it has everything we need... (AKA, "drink their Kool-Aid!")  Here I use the word "want" specifically as respects my purchasing criteria. 

Cheers,
[a href=\"index.php?act=findpost&pid=127799\"][{POST_SNAPBACK}][/a]

Jack - we all know nothing is perfect out there and I think you'd have a hard time - if not an impossible one - replicating in one package all which LR offers. If what you lack is a point Curve, use Camera Raw instead - does the same stuff with this added benefit.

Cheers,

Mark
Title: Curves
Post by: Mark D Segal on July 12, 2007, 10:27:38 am
Quote
This has been a very interesting discussion.  Jack, could you show us how we can benefit from point curves in post processing?  And could others show us how that can be done in LR?

Thank you all.

Fred
[a href=\"index.php?act=findpost&pid=127803\"][{POST_SNAPBACK}][/a]

OK, I'm not Jack so I won't get into the specifics of answering your question, but I must confess to being puzzled by te question itself. A point Curve is the same animal we've had in Photoshop from the get-go. Presumably you've been using it?
Title: Curves
Post by: Fred Ragland on July 12, 2007, 10:50:18 am
Quote
OK, I'm not Jack so I won't get into the specifics of answering your question, but I must confess to being puzzled by te question itself. A point Curve is the same animal we've had in Photoshop from the get-go. Presumably you've been using it?
[a href=\"index.php?act=findpost&pid=127806\"][{POST_SNAPBACK}][/a]
Yes, but it would be helpful to see an example of how an image enhanced with point curves could be enhanced with LR.  The baseline is the point curve image.  The question is whether LR can do as well or better with equal or less effort.

Fred
Title: Curves
Post by: Jack Flesher on July 12, 2007, 10:52:11 am
Quote
Jack, could you show us how we can benefit from point curves in post processing?  [a href=\"index.php?act=findpost&pid=127803\"][{POST_SNAPBACK}][/a]

Fred: I gave a verbal example in detail a few posts back.  If, as Mark asked, you use point curves already, it should be very easy to understand the benefit; more precise adjustments to small tonal ranges of the image.  If you were a user of point curves, you would know how *REALLY* sweet the workflow is.  To simply ctrl/cmd-click a tone in your image, then have the point for that tone appear on the curve ready to be adjusted with the cursor keys or directly, is about as convenient as it gets.

The debate here as I see it, is really over whether we *need* them or just *want* them.  And apparently Adobe feels we don't need them, so they are not a priority.  Personally, I want them so the argument ends there for me.  However, at this point in time, I also need them since *I* cannot get the parametric sliders to give me the adjustment precision I want in a format I'm used to using -- yes, I can get close with the LR sliders, but not as easily nor as precisely.  

Mark, you are of course correct, and it's what I am doing; using ACR because it works well for me.

Cheers,
Title: Curves
Post by: Mark D Segal on July 12, 2007, 11:17:29 am
Quote
Yes, but it would be helpful to see an example of how an image enhanced with point curves could be enhanced with LR.  The baseline is the point curve image.  The question is whether LR can do as well or better with equal or less effort.

Fred
[a href=\"index.php?act=findpost&pid=127810\"][{POST_SNAPBACK}][/a]

Frankly Fred, it wouldn't be very helpful at all. There are images and there are images. Many images will come out just as well either way, some won't. Then there is the efficiency dimension - with some you'll get to home base faster and more easily using LR sliders, with others you'll still get there but not as easily.

I share Jack's view that if we have a tool that millions of people are using successfully and efficiently it should be retained unless something truly better (in terms of both efficiency and effectiveness) is developed to replace it. That hasn't happened with Curves yet, as nifty as the Parametric curve really is.
Title: Curves
Post by: madmanchan on July 12, 2007, 11:37:08 am
What about the Target Adjustment Tool in LR? Does this give you "point curve"-like functionality? (haven't used it much myself, but sounded relevant so I brought it up)
Title: Curves
Post by: Schewe on July 12, 2007, 01:42:11 pm
Quote
- consistency with other RAW converters like Capture NX,
[a href=\"index.php?act=findpost&pid=127754\"][{POST_SNAPBACK}][/a]

That seriously isn't the way to try to move Adobe...pointing to OTHER products is a sure way to watch them shut down and quit listening. They seriously couldn't care less what other products do and will go out of their way to innovate around or beyond the competition.

Bringing up the competition is like the kiss of death in an argument.

Quote
- I am used to working with point curve.

And while true, would make little impact on their thinking...you have to understand that Hamburg in particular is really and truley trying to do new things. So, they will bust their butts to do something in a manner and approach that is new–even if it's a lot more work–than to do the expected and comfortable.

No, you really need to make the arguments only on the basis of optimal image output...that's the one area where proving better results will make them blink because regardless of ALL the other factors, it's really all about image quality. That's the one driving force behind their commitment where you can get traction.
Title: Curves
Post by: Mark D Segal on July 12, 2007, 01:56:53 pm
Quote
What about the Target Adjustment Tool in LR? Does this give you "point curve"-like functionality? (haven't used it much myself, but sounded relevant so I brought it up)
[a href=\"index.php?act=findpost&pid=127822\"][{POST_SNAPBACK}][/a]

Eric, the Targeted Adjustment Tool is fine-tuning with the HSL sliders according to colour family in the HSL/Color section of the Develop Module. It isn't really akin to point-curve functionality in the sense that Jack means it (making fine adjustments to Luminosity in specific tonal ranges of the image). That said, depending on the relationship between the colours and their luminosity distribution in the image, those sliders can be used to impact on localized luminosity directly associated with particular colour families. It's still not the kind of functionality you get via the direct route of a point on a luminosity or composite curve.
Title: Curves
Post by: Schewe on July 12, 2007, 02:08:02 pm
Quote
Eric, the Targeted Adjustment Tool is fine-tuning with the HSL sliders according to colour family in the HSL/Color section of the Develop Module.
[a href=\"index.php?act=findpost&pid=127848\"][{POST_SNAPBACK}][/a]


Yeah, well, the parametric curves ALSO has a TAT...it ain't just for HSL.
Title: Curves
Post by: Jack Flesher on July 12, 2007, 02:13:14 pm
Quote
And while true, would make little impact on their thinking...you have to understand that Hamburg in particular is really and truley trying to do new things. [a href=\"index.php?act=findpost&pid=127844\"][{POST_SNAPBACK}][/a]

And he doesn't care about the interface Adobe has had us using for the last 10 years, an interface that works intuitively and brilliantly?  Come on Jeff, that would be corporate suicide and any software developer will tell you that.  This must be the Macromedia influence speaking.
Title: Curves
Post by: Mark D Segal on July 12, 2007, 02:28:37 pm
Quote
Yeah, well, the parametric curves ALSO has a TAT...it ain't just for HSL.
[a href=\"index.php?act=findpost&pid=127852\"][{POST_SNAPBACK}][/a]

Jeff, if you mean by that the use of the cursor over the specific luminosity areas in the image you want to affect with the Parametric Curve - yes - but that is more of an interface convenience rather than an additional layer of fine-tuning control over the image in the sense that Jack means. It is in this latter context I was speaking. In Lightroom Help - I just checked - I did not find reference to the Parametric Curve under TAT.
Title: Curves
Post by: digitaldog on July 12, 2007, 02:30:21 pm
Quote
And he doesn't care about the interface Adobe has had us using for the last 10 years, an interface that works intuitively and brilliantly?  Come on Jeff, that would be corporate suicide and any software developer will tell you that.  This must be the Macromedia influence speaking.
[a href=\"index.php?act=findpost&pid=127856\"][{POST_SNAPBACK}][/a]

To play devils advocate and based on years of teaching Photoshop, there's really nothing intuitive about using Curves initially. Once you 'get it' it seems simple. But intuitive? I don't think so.
Title: Curves
Post by: Schewe on July 12, 2007, 02:30:42 pm
Quote
And he doesn't care about the interface Adobe has had us using for the last 10 years, an interface that works intuitively and brilliantly?[a href=\"index.php?act=findpost&pid=127856\"][{POST_SNAPBACK}][/a]

He's worked on Photoshop. . .he was the second engineer to work on Photoshop, he was the architect of Photoshop. He knows the code (prolly wrote most of it) and I'm telling you that the LAST thing he would want to do is do it the Photoshop Way™. Lightroom is The UnPhotoshop™.

And completely forget the Macromedia connection. . .nobody that I know working on Lightroom worked at MM...and I'm pretty sure nothing about LR is influenced by MM. Everything about Lightroom is Adobe bred (as far as I know). There is FAR more ImageReady influence than anything else since other than Hamburg, most of the top LR engineers worked on IR. And ironically, the UI is coming from a Kai Krause influence via Phil Clevenger, who worked for Kai.

Almost all of the assumptions people are making about Lightroom's genesis are incorrect as are the assumptions regarding motivation. Photoshop was Thomas' Photoshop. Lightroom is Hamburg's Photoshop and the less it looks and behaves like Photoshop the happier Mark will be. Given a choice between a feature looking and working like Photoshop and a feature looking and working like something you've never seen before, he will go towards the looking and working like something you've never seen before, 9 out 10.

And he has had a MAJOR impact...look at Camera Raw 3 (not 3.7 which started the migration to 4) and Camera Raw 4. Look at ALL the things Thomas adopted from Mark.

Then you start to get the idea...

There's a funny story (which I _THINK_ is true). When Thomas was a senior in HS he went out for the All Michigan Math Conference and placed in the top 10.

Mark being younger (but also from Michigan-Thomas in Ann Arbor, Mark in Midland) followed in Thomas' footsteps later. He went out for the Conference as a HS freshman...placed #1. As a sophomore placed #1. As a junior was invited not to compete...

Not many people can impress Thomas Knoll. Hamburg does...
Title: Curves
Post by: BernardLanguillier on July 12, 2007, 02:35:22 pm
Quote
That seriously isn't the way to try to move Adobe...pointing to OTHER products is a sure way to watch them shut down and quit listening. They seriously couldn't care less what other products do and will go out of their way to innovate around or beyond the competition.

Bringing up the competition is like the kiss of death in an argument.
And while true, would make little impact on their thinking...you have to understand that Hamburg in particular is really and truley trying to do new things. So, they will bust their butts to do something in a manner and approach that is new–even if it's a lot more work–than to do the expected and comfortable.
[a href=\"index.php?act=findpost&pid=127844\"][{POST_SNAPBACK}][/a]

Jeff,

Since you have only reacted nagatively to the last 2 of my 5 arguments in fabour or point curves, I assume that you agree with the first 3?

Cheers,
Bernard
Title: Curves
Post by: BernardLanguillier on July 12, 2007, 02:41:42 pm
Quote
You can work back and forth between them to shape that Curve with remarkable effectiveness. I've processed over a thousand images over the past several months using the new Camera Raw (same thing as LR Develop except ACR does have the point curve in a separate tab), and I find myself needing the Point Curve less and less - but when I need it, I do need it.

This is not because I absolutely can't do with Parametric what I can do with Point, but because the distinction between "want" and "need" is less clear-cut than Jeff's posts would imply. There is also a consideration of "time-effectiveness" in all this. I can probably get 90% of what I "need" from a Parametric Curve quite easily. The other 10% I could probably get as well - with much more time spent fiddling with sliders rather than plunking down several points. I would also like to see composite luminosity, Lab and HSB data read-outs in an expanded info palette. Maybe this needs to be a "feature request".
[a href=\"index.php?act=findpost&pid=127780\"][{POST_SNAPBACK}][/a]

Anyhow I read this comment, it looks like you see value in point curve too.

Quote
I agree with Jeff that there are trade-offs between priorities, especially when a company - which is after all a for profit enterprise (absent which we wouln't have all this stuff) feels it's under commercial pressure to put a product on the market by a certain date. That said, the Point Curve add-in is probably one of the easiest and fastest things they could do, because all the technology for it is sitting right under their own roof in Camera Raw.
[a href=\"index.php?act=findpost&pid=127780\"][{POST_SNAPBACK}][/a]

So why would we not want to request it and push for it?

Quote
Bernard, your last point - "I am used to working with Point curve" - yah, so am I, so is Jack, so is everyone. I don't know how old you guys are, but I'm getting on up there, and I still like to believe that old dogs can learn new tricks. But they'd best be better ones!   
[a href=\"index.php?act=findpost&pid=127780\"][{POST_SNAPBACK}][/a]

But why do we need to adapt to a new tool that is overall less effective than the old one?

Why would I want to try to seduce a girl that I don't find attractive? Just to show friends that I can get her? Frankly speaking, what a waste of time.

Cheers,
Bernard
Title: Curves
Post by: wolfnowl on July 12, 2007, 03:08:31 pm
Hi Folks:

This has been an interesting discussion, and I certainly do not consider myself to be a Curves guru, but for all of the bantering back and forth with each other here, it seems to me that it explains both sides of the situation and solves nothing.  I have no inroads into Adobe, but wouldn't it be smarter for those of us (users) who consider a point curve of great value to move this discussion over to an Adobe forum where, as Jeff suggested, the people at Adobe might be more persuaded to actually do something about it?

Just a thought...

Mike.
Title: Curves
Post by: Schewe on July 12, 2007, 03:33:53 pm
Quote
Since you have only reacted nagatively to the last 2 of my 5 arguments in fabour or point curves, I assume that you agree with the first 3?
[a href=\"index.php?act=findpost&pid=127865\"][{POST_SNAPBACK}][/a]


Correct.

And to be accurate I _DO_ want point curves in Lightroom...I want the ease and smoothness of the usability of parametrics while being able to access the accuracy of point when I need it.

I'm really just trying to explain to people what will or won't move the engineers. The single best argument to make is about image quality...that's something they respond to. All the "I want" or "I need" may make an impact on product and marketing managers but falls on deaf ears in engineering...

And yes, the BEST place to talk about the specifics of the point curves is in the Feature Requests in the Lightroom forums...and the _BEST_ thing to dwell on is image quality...none of the other "stuff" that the engineers don't respond to or care about.
Title: Curves
Post by: Jack Flesher on July 12, 2007, 04:00:06 pm
Okay Jeff... and thanks for the additional background.  I get it, but that doesn't mean I have to like it!  However, consider me done banging my drum on this thread

PS: Andrew, I've taught PS too, and while I agree curves may not be initially intuitive for some, they seem to become so pretty quickly once the relationship between input and output values is explained.

So...  /rant

Best,
Title: Curves
Post by: Schewe on July 12, 2007, 05:06:45 pm
Quote
However, consider me done banging my drum on this thread
[a href=\"index.php?act=findpost&pid=127884\"][{POST_SNAPBACK}][/a]


Ok. . .but keep banging the drum at the Feature Request forum...that's where it'll count for something.
Title: Curves
Post by: Mark D Segal on July 12, 2007, 05:59:08 pm
Quote
Anyhow I read this comment, it looks like you see value in point curve too.
So why would we not want to request it and push for it?

But why do we need to adapt to a new tool that is overall less effective than the old one?

Why would I want to try to seduce a girl that I don't find attractive? Just to show friends that I can get her? Frankly speaking, what a waste of time.

Cheers,
Bernard
[a href=\"index.php?act=findpost&pid=127868\"][{POST_SNAPBACK}][/a]

Bernard,

Yes of course I see a lot of value in retaining the fully, infintely adjustable point curve. We should push for NOT losing it in Camera Raw. People who only want to use LR can decide whether they wish to push for it in LR too.

The new tool is NOT NECESSARILY less effective than the older one. Leave the word "overall" out of your question and we can discuss the merit of having both. It's a different approach. There are many images for which it's all you need. I find, having played with it quite a bit, that I like it more and more. I've reached a point where I can isolate what I can do "real fast" in the Parametric Curve, and what I would prefer to leave for drilling down with a Point Curve. Now that I'm more "at home" with the strengths of BOTH, I think my image processing skills are more efficient.

So I guess I'm saying the girl is attractive, she's not a waste of time, but I'd be seeing the other girl on the side at the same time!  
Title: Curves
Post by: Ralph Eisenberg on July 13, 2007, 08:28:09 am
If there is power in numbers, I should also like to reiterate my feature request for the point-Curves tool, as in ACR 4.1, although not necessarily on the back of the bus. Because this feature obtains in ACR 4.1, I have been using Lightroom less and less, despite its congenial user environment. If I recall correctly, Bruce Fraser commented on what an achievement it was to have access to a (point -) Curves tool in ACR, that a tremendous amount was going on under the hood to make this possible. It would be a shame to forego this feature, which is not to diminish the accomplishment of parametric Curves.
Title: Curves
Post by: rainiershooter on July 15, 2007, 11:28:12 pm
As someone who is just getting started in digital after years with film I find the parametric curve control in Lightroom very easy to use.  The curves feature in PS looked intimitating (to me) and I wasn't particularly looking forward to having to try and figure it out.  

Pcrov:  no one is forcing you to use Lightroom.  Stay with PS and be happy.
Title: Curves
Post by: IanSeward on July 16, 2007, 12:51:54 am
Quote
As someone who is just getting started in digital after years with film I find the parametric curve control in Lightroom very easy to use.  The curves feature in PS looked intimitating (to me) and I wasn't particularly looking forward to having to try and figure it out. 

Pcrov:  no one is forcing you to use Lightroom.  Stay with PS and be happy.
[a href=\"index.php?act=findpost&pid=128367\"][{POST_SNAPBACK}][/a]


I was critical of the change to curves when Lightroom was introduced, as the main reason seemed to be to be different from Photoshop, but reserved judgment until I could evaluate the shipping version.  

Having now used Lightroom one aspect I have become aware off, is that for many images there is no need to use curves at all.  The extra controls available within Lightroom, such as, Recovery, Fill light etc., yield the required result.  Perhaps, because in Photoshop the only tools you had readily to hand were levels and curves you simply had to use curves to do simple adjustments like raising the shadows to bring out shadow detail.  With Lightroom, I have already clicked the fill button to accomplish the required adjustment.  There are some images that do require finessing with curves, and a small proportion of those could do with point curves.  However, if I am dealing with an image in this detail I will almost certainly require selective, rather than global adjustments, so a trip to Photoshop is necessary, and point curves are then available.

I am not an Adobe fan, and although Lightroom is far from being robust in its database functionality, it is an excellent Raw converter.  The apparent dedication of the development team also gives me confidence that the database functionality will improve, so much so that I have finally given in and purchased CS3  

To recap, I certainly found that in Photoshop I always had to use curves.  it would be interesting to know how many images people feel they have to use curves for, after having used the basic controls such as fill light, etc., in Lightroom?  

Having said all that, for good business reasons they might as well put point curves into Lightroom, simply for compatibility with CS3.  Compatibility with CS3 may be a significant driver for Photoshop; it was for me.

Ian
Title: Curves
Post by: digitaldog on July 16, 2007, 09:36:44 am
Quote
Having now used Lightroom one aspect I have become aware off, is that for many images there is no need to use curves at all.

That's been my experience as well. Rarely used. Which bodes for the idea of working top down, left to right where curves are way past the other controls, and the fact that it takes place after basic tone adjustments. This doesn't mean point curves are not useful or we shouldn't ask for them. But going back a week when that one banned and nssty poster went ballistic because I suggested *some* users jump directly to curves based on years of work in Photoshop, and this isn't necessary or recommended. There's at least one Photoshop guru who suggests that we should all set CR and LR to default and conduct the edits afterwards in Photoshop because he fails to understand what's really going on under the hood, and I suspect has spent so many years fixing poorly rendered images in Photoshop instead of doing the heavy lifting at acquisition that the prospect of actually creating good quality images from raw data is way off his radar.
Title: Curves
Post by: BernardLanguillier on July 16, 2007, 10:00:53 am
Quote
Having now used Lightroom one aspect I have become aware off, is that for many images there is no need to use curves at all.  The extra controls available within Lightroom, such as, Recovery, Fill light etc., yield the required result.  [a href=\"index.php?act=findpost&pid=128372\"][{POST_SNAPBACK}][/a]

While I agree that recovery and fill light and useful tools, IMHO they only do a sub-set of what I typically do with curves:

- recovery is a reconstruction tool that computes back some color from partially blown data. This is something that curves never did,
- fill light does selectively brigthen shadows and is something that can indeed be done with curves also.

What these 2 tools can not do is to control the level of contrast selectively for different brightness levels.

Regards,
Bernard
Title: Curves
Post by: Mark D Segal on July 16, 2007, 10:09:50 am
Quote
While I agree that recovery and fill light and useful tools, IMHO they only do a sub-set of what I typically do with curves:

What these 2 tools can not do is to control the level of contrast selectively for different brightness levels.

Regards,
Bernard
[a href=\"index.php?act=findpost&pid=128420\"][{POST_SNAPBACK}][/a]

That is true. For that you go to the Parametric Curve, where you can. That said, you can play between Fill Lights and Blacks to achieve some very interesting contrast.
Title: Curves
Post by: larsrc on July 18, 2007, 08:39:19 am
Quote
I was critical of the change to curves when Lightroom was introduced, as the main reason seemed to be to be different from Photoshop, but reserved judgment until I could evaluate the shipping version. 
[...]

To recap, I certainly found that in Photoshop I always had to use curves.  it would be interesting to know how many images people feel they have to use curves for, after having used the basic controls such as fill light, etc., in Lightroom? 

Having said all that, for good business reasons they might as well put point curves into Lightroom, simply for compatibility with CS3.  Compatibility with CS3 may be a significant driver for Photoshop; it was for me.

Ian
[a href=\"index.php?act=findpost&pid=128372\"][{POST_SNAPBACK}][/a]

I always found Curves to be frustrating -- not that I couldn't use them, but it was all too easy to make the image be a grey morass or have poor toning or both.  I continue to be amazed at the quality LR gives me just by adjusting Fill Light, Black and parametric curves.  Sometimes it's like it reads my mind and knows exactly which parts I want adjusted.  I am so much more productive with this interface than I ever was before.

If I were the LR developers, I'd put point curves on the list, but not very high -- it's one of those "emergency tools" that is the reason they have round-trip to other photo editors built-in.  If you'd only need it for 5% of your images, is it really worth cluttering up the interface?  And here I mean *need*, not just "want" because you're used to it.

-Lars
Title: Curves
Post by: seamus finn on July 18, 2007, 11:29:07 am
Would anybody like to see the various adjustment tools spread across the UI as in the latest Camera Raw rather than down as in LR? Personally, I find the Raw UI less work - it cuts out a lot of scrolling. Any thoughts?
Seamus
Title: Curves
Post by: john beardsworth on July 18, 2007, 12:31:39 pm
Quote
Would anybody like to see the various adjustment tools spread across the UI as in the latest Camera Raw rather than down as in LR? Personally, I find the Raw UI less work - it cuts out a lot of scrolling. Any thoughts?
Seamus
[a href=\"index.php?act=findpost&pid=128851\"][{POST_SNAPBACK}][/a]
Use Solo mode.
Title: Curves
Post by: seamus finn on July 19, 2007, 10:14:44 am
John, Re Solo Mode - a great tip.... Just tried it, like it. Thanks
Title: Curves
Post by: laughfta on July 20, 2007, 09:01:08 pm
Quote
There's at least one Photoshop guru who suggests that we should all set CR and LR to default and conduct the edits afterwards in Photoshop because he fails to understand what's really going on under the hood, and I suspect has spent so many years fixing poorly rendered images in Photoshop instead of doing the heavy lifting at acquisition that the prospect of actually creating good quality images from raw data is way off his radar.


These kind of comments are just as offensive to me as the language that was banned earlier in the thread—and actually feel less honest.  When experts disagree it would be helpful if it were not in the form of a potshot.

I imagine the guru to whom you are referring is one who has carefully and precisely shown the benefits of adjusting curves by the channel in PS instead of using the composite curve that is available in CR and LR. Some gurus aren't satisfied with "good" quality images from raw data. Some may need raw data that can provide an excellent final image. Those two sets of data are not necessarily the same.
Title: Curves
Post by: Schewe on July 20, 2007, 11:49:37 pm
Quote
Some may need raw data that can provide an excellent final image.[a href=\"index.php?act=findpost&pid=129244\"][{POST_SNAPBACK}][/a]

Well, I suppose I've processed more images through Camera Raw and Lightroom than most "experts" or "gurus" and I'm here to tell ya the ONLY reasons one wouldn'd be getting really excelent output from Camera Raw or Lightroom is if you don't know what you are doing or the images you run through are crap to start with.

There is NOTHING about Camera Raw/Lightroom's design or internal engine that precludes photographer from using it to get excellent raw processing results. The "guru" in question ain't even a photographer...

And CR/LR's curve is NOT a composite RGB curve ala Photoshop...it's a hue locked, saturation tuned luminance based curve. Which I proved to the guru in question COULD come real close to his "luminance curve" out of Photoshop. Yeah, it took a few extra slider adjustments..but Thomas Knoll explained exactly why he chose NOT to use a simple luminance curve–because he didn't think it looked right. So, he tweaked it to also increase saturation–not as much as an RGB composite curve–and made sure the hue didn't wander.

Does it match a straight luminance curve? Nope...all ya have to do to do that is wait to edit in Photoshop...if ya got your heart set on a pure luminance curve (or learn how to use LR's controls).
Title: Curves
Post by: Mark D Segal on July 21, 2007, 07:25:04 am
I agree with Jeff's remarks completely. Furthermore, There's nothing wrong with adjusting curves by the channel in Photoshop, but that doesn't invalidate other workflow ideas that can serve just as well - and in some cases better.

The raw data remains unchanged regardless of what we do with it thereafter - short of deleting it, in the sense that a raw file can always be returned to its "native", "as shot" position regardless of the instructions we generate for rendering that data. We can't really damage an image beyond repair or forever jeopardize the prospects of obtaining an excellent result by how we render a raw file. More often than not we would strive to render it properly so that it can be used optimally in Photoshop thereafter - if necessary, but if we're "too smart by half" and end-up with something that really does cause issues in Photoshop, nothing prevents us from going back to the raw file and re-rendering it, perhaps a bit sadder, but wiser.
Title: Curves
Post by: laughfta on July 21, 2007, 08:03:34 am
Quote
Well, I suppose I've processed more images through Camera Raw and Lightroom than most "experts" or "gurus" and I'm here to tell ya the ONLY reasons one wouldn'd be getting really excelent output from Camera Raw or Lightroom is if you don't know what you are doing or the images you run through are crap to start with.

I am absolutely convinced that one can get excellent output from CR and LR. I'm not  convinced that the output is identical—and workflow will certainly be different. I think the workflow is easier in CR and LR, but if some prefer PS's curve due to long standing experience and success with it, well, that makes sense, too.  And sometimes images are crap to start with...


Quote
The "guru" in question ain't even a photographer...


He don't even take pictures?  

Quote
... CR/LR's curve is NOT a composite RGB curve ala Photoshop...it's a hue locked, saturation tuned luminance based curve.


Understood. It's a different curve.

This CR/LR curve cannot be tweaked per channel as PS's curve can, and it can't be tweaked per point, which may be important to those who have successfully built that into their workflow. The two curves work differently. CR/LR curve can give excellent raw processing results. So can the advice to go lightly on raw processing and complete the work in PS—and there is little point in disparaging those who choose this well-worn path. Which is what prompted my reply.
Title: Curves
Post by: digitaldog on July 21, 2007, 10:08:26 am
Quote
I imagine the guru to whom you are referring is one who has carefully and precisely shown the benefits of adjusting curves by the channel in PS instead of using the composite curve that is available in CR and LR. [a href=\"index.php?act=findpost&pid=129244\"][{POST_SNAPBACK}][/a]

Really? Would you care to share how this was proven? I think you'll find, depending on image content and more importantly, the design of the curves in Photoshop by the authors, this isn't either a benefit nor necessary.
Title: Curves
Post by: digitaldog on July 21, 2007, 10:29:04 am
Quote
I am absolutely convinced that one can get excellent output from CR and LR.

The guru in question doens't believe this and calls CR a "non professional tool". He also said he can prove this after he created a spreadsheet that revels the exact math involved:

Quote
I also scripted a routine to batch open multiple variants and spreadsheet the results, to reveal the exact mathematics behind the command.

Multiple requests by multiple readers for proof by submission of said spreadsheet went unanswered for weeks and continues to be ignored. Do you really think said guru really could produce the exact mathematics by examining a few processed files?

Quote
I'm not  convinced that the output is identical—and workflow will certainly be different. I think the workflow is easier in CR and LR, but if some prefer PS's curve due to long standing experience and success with it, well, that makes sense, too.

Agreed entirely.

Quote
This CR/LR curve cannot be tweaked per channel as PS's curve can, and it can't be tweaked per point, which may be important to those who have successfully built that into their workflow. The two curves work differently. CR/LR curve can give excellent raw processing results. So can the advice to go lightly on raw processing and complete the work in PS—and there is little point in disparaging those who choose this well-worn path. Which is what prompted my reply.

No one (here) is arguing they are different. The questions are: Is it necessary or correct to not imply but outright state that the curves in Photoshop are inferior despite the intended design? Are the experience of tens of thousands of users working with the PS curves as designed, without any issues (reather than using them as described by the guru) problematic for anything but a small group of images (or that using a luminosity mode would easily adjust for these differences)? Or that the curves in CR are non professional and that users should leave the default rendering as set and then do the work in a gamma corrected, 8-bit, rendered image within Photoshop because of this behavior? Or that the guru even has a camera capable of producing a Raw file? (also a question gone unanswered).

It goes back to the saying about everything looking like a nail when all you know is a hammer. The guru only has a hammer in his toolbox, he's fine saying a screw driver or hack saw is a non professional tool because he tried to use one to hammer a nail, thus he has proof of this but will not share the proof, something that honest and professional peer review requires. Or course, some readers will take whatever is said as proof without actually testing the waters.
Title: Curves
Post by: laughfta on July 21, 2007, 12:53:26 pm
Quote
Multiple requests by multiple readers for proof by submission of said spreadsheet went unanswered for weeks and continues to be ignored. Do you really think said guru really could produce the exact mathematics by examining a few processed files?


No, I think he exaggerated. Let's hang him. Or, as generous and knowledge-seeking human beings, let's try to understand the point he was making.

Quote
Is it necessary or correct to not imply but outright state that the curves in Photoshop are inferior despite the intended design?


No, but it is a legitimate point to question. Was this something he said on his forum, to his readers?

Quote
Are the experience of tens of thousands of users working with the PS curves as designed, without any issues (reather than using them as described by the guru) problematic for anything but a small group of images


No, though not being problematic does not mean better.

Quote
(or that using a luminosity mode would easily adjust for these differences)?


I don't know. There doesn't seem to be a consensus there. So far the LR curve has produced great results for me,though I would never go so far as to say that it is unequivocally the best way to get the best image every time. So why not try to understand other methods used by other experts?

Quote
Or that the guru even has a camera capable of producing a Raw file? (also a question gone unanswered).


What is the point here? One can only work on files from one's own camera?

Quote
The guru only has a hammer in his toolbox...


Clearly an exaggeration, and therefore impossible to respond to except with sarcasm.

Quote
Or course, some readers will take whatever is said as proof without actually testing the waters.


I suspect you are talking about this reader, but regardless: I think most readers are trying to wade through a lot of information, and are having varying degrees of difficulty doing so. Testing is time consuming and difficult to do thoroughly, though to many of us it is an ongoing and fascinating process. We will all eventually find a workflow that makes sense personally.

Quote
Really? Would you care to share how this was proven? I think you'll find, depending on image content and more importantly, the design of the curves in Photoshop by the authors, this isn't either a benefit nor necessary.

 
Maybe that's the problem—it's a little hard to prove something that is both subjective in nature and different from image to image. But many examples of these comparisons are available for readers to compare and think about.
Title: Curves
Post by: digitaldog on July 21, 2007, 01:11:36 pm
Quote
No, I think he exaggerated. Let's hang him. Or, as generous and knowledge-seeking human beings, let's try to understand the point he was making.

The point he clearly made was he had the exact math used. We don't need to hang him but he could have said on any number of occasions when asked "no that's not what I meant". Instead, in typical fashion when asked to clarify his findings or explain his science, he ignores the issue. That's not a good sign!

Quote
No, but it is a legitimate point to question. Was this something he said on his forum, to his readers?

No, he's wrong about the curve behavior in CR and LR despite his (apparently bogus) math analysis. So first he says the curves behave a certain way (which the author of the software says isn't so), then he tries to back up his claims with a spreadsheet that disappears from the evidence room, then when asked about the proof (spreadsheet) he ignores the request. Sounds suspiciously kooky to me.

Quote
No, though not being problematic does not mean better.

He's saying its problematic to such a point the tool isn't fit for pro use. His words. And what does 'better' mean? How do you define it?

Quote
I don't know. There doesn't seem to be a consensus there. So far the LR curve has produced great results for me,though I would never go so far as to say that it is unequivocally the best way to get the best image every time. So why not try to understand other methods used by other experts?

Some what like the guru, you're trying to tilt the argument. NO ONE said its better. Said guru said it's a non professional tool that can be proven using the exact math based on some technique that cannot be varied. You're trying to turn the burden of proof on the guru by suggesting someone is implying that using the CR curves are better. That's not the topic or argument nor has anyone said that. We could open that up to discussion but at this time, it would only deter the point made by the guru so lets settle that first, then get onto which is 'better' OK?

Quote
What is the point here? One can only work on files from one's own camera?

He could clearly say he used this or that person's camera or even provide information about the file type. He didn't. He again ignored the questions from posters trying to get clarity on what he did. Again, that's highly suspicious. It is up to the person making the claim to back up the claim, not the other way around.

Quote
Testing is time consuming and difficult to do thoroughly, though to many of us it is an ongoing and fascinating process. We will all eventually find a workflow that makes sense personally.

Fine, then don't make blanket claims and expect everyone to accept them as proof without the proof. I can say the moon is made of cheese and I have to backup that claim unless everyone wants to automatically believe this based on my personal background (which I would suggest is a bad idea).

Quote
Maybe that's the problem—it's a little hard to prove something that is both subjective in nature and different from image to image. But many examples of these comparisons are available for readers to compare and think about.

Its not hard to explain the math. Especially when you tell people you have the exact math to prove your point but then don't want to provide the math. It smells fishy to say the least!
Title: Curves
Post by: laughfta on July 21, 2007, 02:02:09 pm
I see this has already been explored elsewhere, and one of the final comments made by said guru was this:

"In Photoshop CS2, I felt that the answer was clearly to be conservative in view of the range-handling issues. With the improvements in the current version this may no longer be the case, and if it is not, I would welcome the development."

Doesn't sound quite as extreme as you have portrayed it.

I think you have a point about proving blanket statements, but on the other hand, you make many of them in your own responses. I am going back on the sidelines to listen and learn.
Title: Curves
Post by: Mark D Segal on July 21, 2007, 02:23:28 pm
Quote
Its not hard to explain the math. Especially when you tell people you have the exact math to prove your point but then don't want to provide the math. It smells fishy to say the least!
[a href=\"index.php?act=findpost&pid=129326\"][{POST_SNAPBACK}][/a]

Well, I think it is hard to explain the math. The math we're talking about here is complex stuff, and there is obviously much happening inter-actively that would make it very difficult to explain easily.

One can talk about relationships between inputs and outcomes, but that isn't the math. Setting that aside,  these tools were designed for professional use, and they are being used by countless numbers of professionals, and the market is accepting what they do with them, otherwise we would have heard about it loud and clear by now - from more people than said guru.

Laughta's 1:02 PM post isolates one of said guru's quotes, but not others which leave quite a different impression about his views of the "merits" of Camera Raw 4. But that's fine - at a technical level this is a discussion that can be very revealing and rewarding. At some point one would hope for a generally accepted view of the relative merits of these tools.
Title: Curves
Post by: Schewe on July 21, 2007, 02:24:47 pm
Quote
I am absolutely convinced that one can get excellent output from CR and LR. I'm not  convinced that the output is identical—and workflow will certainly be different.[a href=\"index.php?act=findpost&pid=129282\"][{POST_SNAPBACK}][/a]

Well, both Thomas Knoll and Mark Hamburg say the results from Camera Raw 4.1 and Lightroom 1,1 are the same, so take it up with them. I'm inclined to believe them over somebody else-since they wrote the friggin' code...yes, Lightroom doesn't give you access to a point curve editor like Camera Raw does (but it CAN process point curve data from Camera Raw) and Camera Raw doesn't have the TAT tools that LR has...but equal settings in either will produce exactly the same output from either...

Quote
Maybe that's the problem—it's a little hard to prove something that is both subjective in nature and different from image to image. But many examples of these comparisons are available for readers to compare and think about.

Well, the "guru" said he had the EXACT MATH behind the Camera Raw curves function and said it was the same as Photoshop's, and thus not a pro level tool. The author of the math said he was wrong, That's not very subjective...either one is right or one is wrong. And Thomas also went on toe say that the EXACT MATH behind Camera Raw's code is actually available in the DNG SDK available to anybody-which kinda made the claims made by the "guru" that he deduced the code by running tests and plotting them in a spreadsheet pretty pointless–considering he was wrong.

This particular "guru" (Dan Margulis) gets his jollies off by making outrageous claims on his own private list where by his ground rules, he controls the nature of the discussion-it ain't a free and open environment like here. In fact Andrew is now banned from his list because of his "negative" approach (the actual reasons go a bit further but that's a generally accurate reason, right Andrew?).

So, to take Dan's position, anybody using Camera Raw for processing can't be a "professional" since Camera Raw ain't a pro level tool. Cause it doesn't have a pure luminance curve...see Dan is kinda attached to doing stuff in Lab since he wrote a book on it. And Camera Raw doesn't work in Lab (Pro Photo RGB chromaticities and a linear gamma) so obviously, to somebody like Dan (who ain't a photographer) Camera Raw just isn't a tool a pro would use...and by extension, same with Lightroom.

So, on one had you have Dan and the other hand you have Thomas Knoll who coauthored Photoshop and Camera Raw (and now controls the raw processing pipeline in Lightroom). Pick your side...I choose to back the horse with a winning record, (your milage may vary).
Title: Curves
Post by: digitaldog on July 21, 2007, 02:55:35 pm
Quote
Doesn't sound quite as extreme as you have portrayed it.

I'm not trying to portray anything, I'm providing quotes from the guru. A few more (this isn't a paraphrase, its exact word by word quotes):

Quote
Master-type corrections are undesirable because they increase and/or decrease saturation in ways that the user can't control. [AR despite this is how the tool was designed and most people prefer to see the results]
When a picture is being lightened by a master correction, saturation has to increase because the lightest channels are being hurt proportionally more than the dominating darker ones. Often that makes the image look better, but it's clearly the wrong policy to implement as a general rule--a serious user needs to make the decision of whether to saturate or not on a case-by-case basis.

Yes it makes it look better because that's how it was designed AND what nearly all users want. And its wrong because guru says its wrong? And this can be controlled if so desired, the saturation can be lowered to produce roughly the results of the individual curve method using a luminosity blend adjustment in PS or using additional tools in CR.

Quote
2) Those people who still use master RGB curves or, worse, levels, might take this occasion to discard these primitive tools (or limit them to very small moves, where the harm they do is unlikely to be noticeable). This image is a spectacular example of why channel-by-channel is better, but even images that superficially look good can benefit from proper color handling, which master adjustments don't offer.

Primitive? One guru's opinion but just that, an opinion without backing.
Better? Why, if it doesn't look better?

Quote
I studied whether there were any circumstances under which image-by-image manipulations in Camera Raw followed (if necessary) by manipulations in Photoshop would ever yield superior quality to a conservative, damage-free Camera Raw acquisition followed by Photoshop manipulation. After testing around 100 images, I concluded that in some cases opening the range in Camera Raw actually damaged the image to the point that it was no longer possible to get a good result without excessive effort, if it was possible at all. In most cases the impact was small. There were no cases I could identify where one could get better image-by-image results by using any of Camera Raw's functionality.

Damaged how? Never explained. Was the damage the "if necessary" manipulation in Photoshop after (on an 8-bit file)? Don't know.

My favorite quote:

Quote
It does, however, beg the question: if saving time is so important that quality compromises need to be made, why is the raw format being used at all? With rare image-specific exceptions, essentially anybody who is not a beginner will get better final results by shooting JPEG and correcting in Photoshop than an expert can who shoots raw but is not allowed to do any manipulation outside of the acquisition module. And in less time, too. The idea of a raw module is to *empower* the image-manipulation program, not replace it.

And

Quote
We are interested in good but not great quality. We will correct images individually in the raw module to make them look as best we can, but that's going to be it. It doesn't matter if we could do more in Photoshop
afterwards--we're just not going to do it, because the extra quality isn't worth the time.

Lastly:

Quote
Every other professional-level program gives us several options of how to establish range. But with ACR, you are getting more saturated colors, whether you like it or not--no way to bypass it. That's not acceptable for a professional user, and that's why the routine is inferior.

Not true as will soon be demonstrated. Of you can try on your own in CR or LR, the tools are there.

So working on an 8-bit JPEG is the way to go, not using CR? Must be due to the math of which we have yet to see proof (or for that matter, images). Anyone here who's worked on edited JPEGs and a good Raw converter buy this?

OK, my portrayal is we have a guru that seriously doesn't understand how Raw modules work or why, sees Photoshop as the only hammer in his tool chest and when asked to prove the short comings of a initial Raw converter rendering workflow (its not correction, its rendering, something else he fails to recognize), guru ignores requests for proof.

He's used similar tactics to dismiss high bit workflows or wide gamut working spaces. Since most users, other than those who believe everything he says, has moved past those two points, and adopted these techniques the next whipping boy is CR and LR specifically (Adobe products, which he loves to whale on).

The tactics he uses, are clearly demonstrated and have been for years with respect to his High Bit challenge:

http://www.brucelindbloom.com/index.html?DanMargulis.html (http://www.brucelindbloom.com/index.html?DanMargulis.html)

Its interesting to see how a true scientist views these tactics. Again, he says the moon is made of cheese, and we are not the ones that should disprove the notion, he has to prove the point. He didn't do it here with respect to high bit capture and editing, he didn't do it with Raw processing. My portrayal is he doesn't use good science, he uses religion. Some of us don't care to drink that koolaid.

On the other hand, IF he really did have math to prove his point, he could provide this as a  benefit to the imaging industry for people like Thomas Knoll. I suggested his role here could be useful to the industry but that's not on his radar. Or maybe the so called math is a figment of his imagination?
Title: Curves
Post by: digitaldog on July 21, 2007, 03:24:53 pm
The first ban on Dan's site I got was arguing with him about the U.S. Web Coated (SWOP) v2 profile in Photoshop. Since its a Photoshop supplied profile and of course an ICC profile, its bad, bad bad.

His take was that on EVERY device, the U.S. Web Coated (SWOP) v2 profile shifts blues to magenta. He used the Photoshop info palette to provide numbers that show this despite the fact that, when you send the U.S. Web Coated (SWOP) v2 profile to a device that expects these numbers, you get no shift. You get blue. This was demonstrated at a Seybold Seminar by Chris Murphy (with myself and Bruce Fraser) years ago in front of an audience. Using a SWOP proofing that did conform to TR001 (of which the U.S. Web Coated (SWOP) v2 profile describes), this profile provided the best output even when compared to custom build ICC profiles (built by Chris to the same proofing device).

Its convenient to say that the U.S. Web Coated (SWOP) v2 profile always does this or that. I questioned Dan about how he could say this when A, he couldn’t possibly test this profile on every CMYK device known to man and B, he didn't output tests to an actual device that DOES conform to TR001. In Dan's world, he's right based on whatever metric he wants to post. Ask him to send the CMYK numbers to a TR001 defined deivce and tell you what the blues look like, or better actually measure it (that would require he have a dreaded Spectrophotometer making him as he calls many of us, "calibrationists" an insulting term), he again ignores the request for scientific proofing methods. Being that I had seen the results and know lots of users who work with the U.S. Web Coated (SWOP) v2 profile and don't automatically, always get shifted blues, I pushed him to the point he could do nothing more than say I was being nasty (that's the pot calling the kettle black, I have many posts backing up his nastyness to Chris, Jeff and others), his only recourse was to silence the dissenter. That was for 6 months.

With the CR and LR nonsense he posted, he got tired of me requesting actual proof and the math.

Funny, the heat was just too great for him as he wrote me this about the reason for the lifetime ban:

Quote
We do this not because of any specific thread or because we believe you
cannot make a constructive contribution to the list. Indeed, some of your posts
have contributed valuable information. However, for every useful post we find an
unacceptably large mutliple of posts that we consider to be counterproductive
because they fall into one of the following categories:

*Personal attacks on others, containing no useful technical content.
*Gratuitous plugs for Adobe products, without useful discussion.
*Aggressive, rote repetition of arguments that you have made many times
previously, without new content.

The 'We' by the way isn't Dan, he would never allow censorship. Its his 5 moderators.

Am I upset by this? No, not really. Jeff and one or two other regular posters know of this history but this is the first (and should be last time) I've discussed it. Since Jeff brings it up and since some have asked "where you been on the ACT list" the cat's out of the bag.

Again, when good science and rational thinking don't apply and someone asks for clarity of rational thought, and you don't want this idea expressed, the best tactic if possible is to ban the discussion. That's what Dan has done.

Note too that until the 2nd cat was out of the bag, I went out of my way to refrain from using anyone's name here, simply using the term guru. Its not personal. I've known Dan for many years, I've dined with him, I run into him from time to time. I think in person he's a charming fellow. I think he has a LOT to contribute and I've learned a lot from him. I own all his books (paid with my own money<g>). It makes all this the more frustrating that he has such an enormous BS factor. He really doesn't need to slam Adobe or be controversial to direct attention to himself and yet, that's been his MO for years. Its a shame. Instead of truly contributing to the imaging industry, providing useful, not mean spirited suggestions to Adobe or making up imaginary spreadsheets, he could be working for the good of imaging-kind.
Title: Curves
Post by: Ray on July 23, 2007, 05:50:54 am
Quote
The first ban on Dan's site I got was arguing with him about the U.S. Web Coated (SWOP) v2 profile in Photoshop. Since its a Photoshop supplied profile and of course an ICC profile, its bad, bad bad.

His take was that on EVERY device, the U.S. Web Coated (SWOP) v2 profile shifts blues to magenta. He used the Photoshop info palette to provide numbers that show this despite the fact that, when you send the U.S. Web Coated (SWOP) v2 profile to a device that expects these numbers, you get no shift. You get blue. This was demonstrated at a Seybold Seminar by Chris Murphy (with myself and Bruce Fraser) years ago in front of an audience. Using a SWOP proofing that did conform to TR001 (of which the U.S. Web Coated (SWOP) v2 profile describes), this profile provided the best output even when compared to custom build ICC profiles (built by Chris to the same proofing device).

Its convenient to say that the U.S. Web Coated (SWOP) v2 profile always does this or that. I questioned Dan about how he could say this when A, he couldn’t possibly test this profile on every CMYK device known to man and B, he didn't output tests to an actual device that DOES conform to TR001. In Dan's world, he's right based on whatever metric he wants to post. Ask him to send the CMYK numbers to a TR001 defined deivce and tell you what the blues look like, or better actually measure it (that would require he have a dreaded Spectrophotometer making him as he calls many of us, "calibrationists" an insulting term), he again ignores the request for scientific proofing methods. Being that I had seen the results and know lots of users who work with the U.S. Web Coated (SWOP) v2 profile and don't automatically, always get shifted blues, I pushed him to the point he could do nothing more than say I was being nasty (that's the pot calling the kettle black, I have many posts backing up his nastyness to Chris, Jeff and others), his only recourse was to silence the dissenter. That was for 6 months.

With the CR and LR nonsense he posted, he got tired of me requesting actual proof and the math.

Funny, the heat was just too great for him as he wrote me this about the reason for the lifetime ban:
The 'We' by the way isn't Dan, he would never allow censorship. Its his 5 moderators.

Am I upset by this? No, not really. Jeff and one or two other regular posters know of this history but this is the first (and should be last time) I've discussed it. Since Jeff brings it up and since some have asked "where you been on the ACT list" the cat's out of the bag.

Again, when good science and rational thinking don't apply and someone asks for clarity of rational thought, and you don't want this idea expressed, the best tactic if possible is to ban the discussion. That's what Dan has done.

Note too that until the 2nd cat was out of the bag, I went out of my way to refrain from using anyone's name here, simply using the term guru. Its not personal. I've known Dan for many years, I've dined with him, I run into him from time to time. I think in person he's a charming fellow. I think he has a LOT to contribute and I've learned a lot from him. I own all his books (paid with my own money<g>). It makes all this the more frustrating that he has such an enormous BS factor. He really doesn't need to slam Adobe or be controversial to direct attention to himself and yet, that's been his MO for years. Its a shame. Instead of truly contributing to the imaging industry, providing useful, not mean spirited suggestions to Adobe or making up imaginary spreadsheets, he could be working for the good of imaging-kind.
[a href=\"index.php?act=findpost&pid=129351\"][{POST_SNAPBACK}][/a]

Andrew,
I find it difficult to believe that such a knowledgeable chap such as yourself should be banned from any site. I think you have a lot to contribute and I find it disturbing in general that opposing points of view should be censored on any site.. It happens of course. It happens in America, the land of the free with an impressive constitution that is often only paid lip service to.

I'm all for letting it all hang out. Let the facts prevail.
Title: Curves
Post by: laughfta on July 23, 2007, 08:16:41 am
Quote
I find it disturbing in general that opposing points of view should be censored on any site..



This is something to consider...when opinions are presented in anger and with a strong personal bias the message is diluted and the veracity unclear. This thread started with someone expressing his version of the truth, and he was banned from this site.

There is a lot of information out here, and I appreciate Andrew's contribution. I also appreciate Dan's. I would like unbiased, or at least less biased help in sorting it out.

Gloria
Title: Curves
Post by: john beardsworth on July 23, 2007, 09:41:36 am
Quote
This is something to consider...when opinions are presented in anger and with a strong personal bias the message is diluted and the veracity unclear. This thread started with someone expressing his version of the truth, and he was banned from this site.
[a href=\"index.php?act=findpost&pid=129499\"][{POST_SNAPBACK}][/a]

He wasn't banned for expressing his version of the truth, but for saying "Go fuck yourself." Not the most effective way to discuss the issue.

John
Title: Curves
Post by: Mark D Segal on July 23, 2007, 09:51:01 am
Quote
He wasn't banned for expressing his version of the truth, but for saying "Go fuck yourself." Not the most effective way to discuss the issue.

John
[a href=\"index.php?act=findpost&pid=129506\"][{POST_SNAPBACK}][/a]

Exactly
Title: Curves
Post by: laughfta on July 23, 2007, 12:07:00 pm
Quote
He wasn't banned for expressing his version of the truth, but for saying "Go fuck yourself." Not the most effective way to discuss the issue.

John
[a href=\"index.php?act=findpost&pid=129506\"][{POST_SNAPBACK}][/a]

  Good point.  

I had a point about effective communication, as well. Which brings me full circle:

Quote
These kind of comments are just as offensive to me as the language that was banned earlier in the thread—and actually feel less honest. When experts disagree it would be helpful if it were not in the form of a potshot.


Maybe I was over-reacting, myself. I'll just let it go.

Gloria
Title: Curves
Post by: sinc on July 25, 2007, 08:07:41 pm
I'm using Photoshop CS2. Does CS3 have the tools you're discussing? Is Camera Raw 4.0 part of CS3?
Title: Curves
Post by: Mark D Segal on July 25, 2007, 08:37:43 pm
Quote
I'm using Photoshop CS2. Does CS3 have the tools you're discussing? Is Camera Raw 4.0 part of CS3?
[a href=\"index.php?act=findpost&pid=129916\"][{POST_SNAPBACK}][/a]

Yes and yes
Title: Curves
Post by: macgyver on July 26, 2007, 01:28:35 am
Holy crap, stop the maddness.
Title: Curves
Post by: DaHen on July 27, 2007, 08:55:37 am
Quote
Holy crap, stop the maddness.
[a href=\"index.php?act=findpost&pid=129942\"][{POST_SNAPBACK}][/a]


Can't. We live on maddness. lol.

 
Title: Curves
Post by: sgs8r on July 27, 2007, 09:04:47 pm
Quote
That seriously isn't the way to try to move Adobe...pointing to OTHER products is a sure way to watch them shut down and quit listening. They seriously couldn't care less what other products do and will go out of their way to innovate around or beyond the competition.

Bringing up the competition is like the kiss of death in an argument.
And while true, would make little impact on their thinking...you have to understand that Hamburg in particular is really and truley trying to do new things. So, they will bust their butts to do something in a manner and approach that is new–even if it's a lot more work–than to do the expected and comfortable.

No, you really need to make the arguments only on the basis of optimal image output...that's the one area where proving better results will make them blink because regardless of ALL the other factors, it's really all about image quality. That's the one driving force behind their commitment where you can get traction.
[a href=\"index.php?act=findpost&pid=127844\"][{POST_SNAPBACK}][/a]

This (and similar) comments are a bit dismaying. You've heard of the "not invented here syndrome"? Dogmatically refusing to consider even good ideas because they're not yours? It's a difficult line to walk, I suppose. Pay too much attention to what others think and do and it becomes hard to think enough outside the box to do something really innovative (and hopefully good). But ignore everything completely and you become limited, no matter how great a genius you may be. Like Microsoft, Adobe's de facto monopoly on photo processing gives it substantial license to ignore users and competitors and any good ideas they may have.  But sooner or later there will come a tipping point that lets someone else into the game (maybe if/when photography become a subset of video as some are claiming is already starting to happen). Then Adobe/Photoshop/Lightroom might go down like the Titanic (c.f. Wang, DEC, Polaroid, Kodak(?)...). Meantime, their software won't be as good as it might be otherwise.
Title: Curves
Post by: Schewe on July 27, 2007, 09:17:23 pm
Quote
This (and similar) comments are a bit dismaying.[a href=\"index.php?act=findpost&pid=130208\"][{POST_SNAPBACK}][/a]

You're new here, huh? Well, you might want to try to blend in a little before you start condeming Lightroom and the other applications the industry needs. It would also be useful to know a little something about just how Adobe goes about getting feedback and input from users. They bend over backwards trying to get the right sort of feedback and input. But just pointing to some other product and saying "do this" ain't gonna fly bud. It's lazy if nothing else and doesn't tell Adobe ANYTHING it doesn't already know.

The primary reason they really DON'T want to look to steal other stuff is they can't afford to. It's not so much a "not invented here" sort of thing...more like they have to be VERY CAREFUL about how Adobe engineers go about doing research. Otherwise they leave Adobe wide open for lawsuits...

And truth be told, these guys are creatives, just like a lot of us here. They really don't want to just copy somebody else's work. That ain't their job...and as for video pushing photo. . .yeah, well, we'll see. It's two entirely different mindsets. And Adobe has a pretty good foot in the door with After Effects and Premier Pro-just returned to the Mac after a few years of being Win only.

Nope, I think you're just gonna have to be stuck with primarily Adobe for the long haul, so you may as well learn how to influence them–unless all you want to do is snipe from the sidelines.
Title: Curves
Post by: rick.mn on July 28, 2007, 11:43:35 am
I just wanted to add to Jeff's comments about law suits.  While not a software engineer, I have worked 23 years in the hi tech industry.  It is hard to imagine how difficult it is to be aware of all of the patent and other intellectual property art out there.  

I, for one, am immensely grateful to Adobe for paying attention to these issues.  The last thing I want as a user is to have my software taken back from me, because of an infringment injunction.

Rick
Title: Curves
Post by: Mike Ornellas on July 30, 2007, 04:50:44 pm
Dan is only interested in one thing and that ismaking money.,
Title: Curves
Post by: Mark D Segal on July 30, 2007, 05:13:44 pm
Quote
Dan is only interested in one thing and that ismaking money.,
[a href=\"index.php?act=findpost&pid=130724\"][{POST_SNAPBACK}][/a]

You know, I have to say once again - I have no role moderating this website - but I am involved in this discussion and I don't like being involved with discussions that turn into personal attacks against people - especially ones with serious professional reputations. You and I may not necessaarily agree with every position Dan takes on every issue and it is fair ball to say so and say why - at a technical level - but there is NO ROOM here for personal attacks with innuendo about peoples' motives when you most likely don't have the kind of relationship with them that would even give you a glimmer of insight into such things - and even if you did - not for reproduction on a website like this. This Forum is provided to us for exchanging ideas of educational value and comments like this DO NOT fit that description. I strongly recommend that you delete the post.

Mark