Rob Cole wrote:
Lightroom's new basic control set is defined by the need for Revel compatibility, not to provide maximum control for professionals and picky users like me.
No, actually it isn't...not sure where you got that faulty info...the ORDER that the controls are is has been influenced by Mark Hamburg (the founding engineer on Lightroom and Revel) for use in Revel minus the Whites and Blacks...but LR4's controls and Revel's controls really aren't the same (for example, Revel only processes JPEGs, not raws).
Look Rob, you simply don't know anything about the internal development of Lightroom. Saying these things shows your ignorance...
Jeff Schewe wrote:
No, actually it isn't...not sure where you got that faulty info...the ORDER that the controls are is has been influenced by Mark Hamburg (the founding engineer on Lightroom and Revel) for use in Revel minus the Whites and Blacks...but LR4's controls and Revel's controls really aren't the same (for example, Revel only processes JPEGs, not raws).
Maybe Rob got his "faulty" info from Eric who wrote:
"Another aspect of the design is that only the first 4 sliders -- Exposure thru Shadows -- are available in one of our other products, and so maximizing the usefulness of just these 4 without relying on Whites/Blacks was important. And of course, consistent rendering and interoperability of the products was also important."
He also wrote "...Lr itself is a standalone product. But there are other Adobe products and consistent rendering across products is important to us and to our customers. So, our design for the new controls in PV 2012 had to keep this in mind."
So your information seems to be the "faulty" one to me.
BTW, what professional body has deemed Mark Hamburg's education as being compliant to an "engineering" standard? I have a feeling that many "engineers" in the LR team (past and present) are just programmers as opposed to professionally accredited engineers. Not being a proper engineer need not imply lack of talent, but I have a feeling that the term "engineer" is used inflationary in the context of LR development.
tgutgu wrote:
In order to be able to discuss with you, could you PLEASE, backup your claims with some examples, shown here publicly.
Have you seen the example in this thread:
http://forums.adobe.com/message/4179562
?
Seems to show that a small and quick set of adjustments with PV2010 take a very long time to reproduce with PV2012.
Disclaimer: I haven't looked at this example in detail, but it may be what you are looking for.
TK2142 wrote:
Maybe Rob got his "faulty" info from Eric who wrote:
"Another aspect of the design is that only the first 4 sliders -- Exposure thru Shadows -- are available in one of our other products, and so maximizing the usefulness of just these 4 without relying on Whites/Blacks was important. And of course, consistent rendering and interoperability of the products was also important."
And what part of Eric's quote says Lightroom has had it's processing dictated by Revel? It hasn't..."consistent rendering and interoperability of the products" also goes along with Camera Raw...Rob thinks he knows what goes on in LR dev, he doesn't.
TK2142 wrote:
BTW, what professional body has deemed Mark Hamburg's education as being compliant to an "engineering" standard?
Uh huh...you're joking right? Google him...
TK2142 wrote:
BTW, what professional body has deemed Mark Hamburg's education as being compliant to an "engineering" standard?
Let's see ... I Googled "Mark Hamburg" and received 78,100,000 results ... with the vast majority of refernces pointing to Adobe, Photoshop and Lightroom ...
I Googled TK2142 and received 13,200 results with the first page containing links to Turkish airlines, ebay and more or less gibberish.
Perhaps before you question or otherwise attempt to sully the credibility of someone very publicly linked to an endeavor ... you could legitimize your own identiy beyond a vauge collection of alpha-numeric symbols ...
Until then, Mr. Hamburg's contributions to the cause will carry considerably much more weight than your comments here ....
This whole thread is such a waste of server space ...
IF Mr. Cole had expressed his findings and opinions as succintly and concisely as he did in his most recent contribution to this thread ... without his prior offerings of conjecture, speculation and editorializing his own self-proclamied importance ... there would be little to no contraversy on the matter.
There is no doubt that Lr4b and pv 2012 is NOT perfect ... I don't think anyone involved expected the beta to be so ... however, the incredulous contempt shown by some over those who are offering the opportunity for us to test Lr4 is beyond comprehension.
BETAS ARE NOT INTENDED TO BE PERFECT! If they were, there would be no purpose whatsoever for public beta testing. Betas are a starting point so the engineers may receive feedback to enable them to fine-tune for the final relaese. Expecting more, or placing too much importance on the findings of individual testers goes far beyond the intended means ...
The issue here are not about the credibilty, motives or work environment of those Adobe employees tasked with presenting the software ... it is for us to test, evaluate and trial the capabilities of the software as presented, then report back our issues, problems and concerns after putting it through the paces ... anything more that we may want to offer beyond that is pointless and counter to the cause. Whether or not we as individuals question particular team members as "qualified" for their job titles is meaningless ... I'm sure Adobe doesn't make a habit of hiring "posers" to work on such important projects.
Beta testing feedback should be based upon results, or the lack thereof and not emotionaly induced narrative ... so if you want to question Adobe staff selction, their academic or professional accreditation, you actually may want to consider another venue for that cause. ... And ... if you want to proclaim your own sel-importance you may be better off joining the blogoshpere and take your chances there to make an impact ... because thumping on Mr. Hamburg and Mr Chan here ... ain't gonna getcha very far ...
Jeff Schewe wrote:
And what part of Eric's quote says Lightroom has had it's processing dictated by Revel?
No one claimed it was "dictated by Revel". Eric's statement, however, clearly states that the function of the "Highlights & Shadows" sliders was influenced by compatibility considerations and since ACR will have "Whites & Blacks" controls, that leaves Revel to be the influence.
Jeff Schewe wrote:
It hasn't..."consistent rendering and interoperability of the products" also goes along with Camera Raw...
No alignment between LR and ACR is necessary as the former's Develop module is essentially just a UI for the latter.
Jeff Schewe wrote:
Uh huh...you're joking right? Google him...
No, it wasn't a joke. It was a question. And none of the Google results -- I went through that exercise before I asked -- tell me what education Mark Hamburg has enjoyed and whether that education satisfies the requirements of the Washington Accord. Of course you can call someone an "engineer" but that doesn't make them one.
Again, someone not being an engineer doesn't imply that they are bad at what they are doing, not at all, but AFAIC, you should not refer to someone as an "engineer" unless said person has been certified to be an engineer by a professional engineering organisation.
PaoloAvezzano wrote:
I can't see the point in that, apart from being offensive.
Sorry, I didn't mean it to be offensive.
I repeatedly said that not being an engineer doesn't imply that one is bad at what one is doing, so I don't know why my question is being perceived as being offensive.
I just keep reading people using the term "engineer" and wonder whether they actually mean "programmer" or "developer".
If I had written that from what I have gleaned from the Adobe LR podcasts that Thomas Knoll comes across as someone who really knows what he is doing, while this is much less the case for Mark Hamburg, then I would have understood that someone said this could regard this as being "offensive". But I didn't. I merely asked a question about whether Jeff Schewe is correct in referring to Mark Hamburg as an engineer.
Just because someone does things an engineer could do, doesn't make them an engineer. FWIW, a pure computer scientist has a qualification, but isn't an engineer either.
From Wikipedia...
An engineer is a professional practitioner of engineering, concerned with applying scientific knowledge, mathematics and ingenuity to develop solutions for technical problems. Engineers design materials, structures, machines and systems while considering the limitations imposed by practicality, safety and cost.[1][2] The word engineer is derived from the Latin roots ingeniare ("to contrive, devise") and ingenium ("cleverness").[3][4]
Engineers are grounded in applied sciences, and their work in research and development is distinct from the basic research focus of scientists.[2] The work of engineers forms the link between scientific discoveries and their subsequent applications to human needs.[1]
The key item seems to be the work they do in applying solutions.
To my knowledge there is no licensing requirement or board accreditation required to be called a software engineer.
John
Hal P Anderson wrote:
I think PV 2010 has to be the exact same code as LR3--otherwise, the backward compatibility wouldn't work.
I think you're right BUT there is a bug in the PV2010 snapshots (in Lr4b1) - they don't save/restore the point curve.
So, if you toy with both process versions, and include a point curve manipulation in the PV2010 version and make a snapshot of it, then switch to PV2012 and back to PV2010 via the snapshot, then you'll lose your point curve. I haven't verified whether the problem is saving or restoring the snapshot, so it could be the same problem would occur if snapshot was saved in Lr3 - dunno.
Note: Sometimes people (including me) use the terms "point curve" and "tone curve" synonymously, but in this case, I really mean "point curve", not "tone curve".
TK2142 wrote:g offensive.
If I had written that from what I have gleaned from the Adobe LR podcasts that Thomas Knoll comes across as someone who really knows what he is doing, while this is much less the case for Mark Hamburg, then I would have understood that someone said this could regard this as being "offensive". But I didn't. I merely asked a question about whether Jeff Schewe is correct in referring to Mark Hamburg as an engineer.
Yes,,,Mark is a software engineer....but he's actually much more than that. He was the #2 engineer to work on Photoshop with Thomas. He has been named an Adobe Fellow. He was the founding engineer and designer of Lightroom. Many of the functions in Develop came directly from Mark such as; HSL/Grayscale, Parametric Curves, Split Tone. Thomas was impressed enough with Mark's work to incorporate those functions into Camera Raw. Mark left Adobe for a while to work on UI and usability at MSFT but came back to Adobe. Rather than go back to work on Lightroom, he decied to do a consumer app–Revel.
Mark has always thought that ACR/LR has had too many controls (ironic since he added 24 HSL controls) so when he worked on Revel, he decided to try to distill a lot of functionality into less controls. But trying to understand the team's motives from the outside is a waste of time.
It seems to me there are two different flavors of toning issues that could be talked about:
1. General toning flexibility given the basic control set (and shadow/highlight recovery processing - but I'll not be talking about that aspect in this post).
2. Toning characteristics of the "adaptive" algorithm.
Previously I didn't understand the difference between the two, and so often was talking about one or the other or both without knowing which.
Part of Lightroom 4's new magic is to separate tones in a locally adaptive fashion. Lr3 never did that.
It's the reason Lr4 photos can look so brilliant and clear and shiny and wonderful (without added clarity).
What that means is that you can have toning that's a little different in different parts of the photo, even for the same tones.
Personally, I often like the effects of the locally adaptive toning algorithm in some parts of a photo, but not others. The reason it's so exasperating to try and correct it using the basic tone controls is that it is essentially not correctable. The best one can ever hope for is a balanced compromise.
Type 1 toning issues are relatively easy to adjust using the tone curve.
Type 2 toning issues are impossible (literally) to adjust using the tone curve - locals required.
I feel relieved that I finally understand what's going on (or at least I understand a little better - I always like to leave plenty of room for new enlightment...), and thus armed will be better able to get the results I want with less fiddling.
PS - No way to "correct" type 2 toning issues without losing the Lr4 magic. Eric may fine tune some, but I doubt there will be any changes in this other than very subtle.
Type 1 issues are constrained by the number of sliders available - dunno the extent to which Eric could find ways to improve handling, my guess is that changes will be relatively minor.
Cheers,
Rob
To be honest: I love every aspect of the PV2012 algorithm. But, as you all stated, that's personally. For me it works.
It's also easy for me to reproduce the colors EXACTLY as they should be in real life.
- I calibrated my monitor with the X-Rite i1 Display Pro
- I took a photo of the X-Rite ColorChecker Passport, with the lighting exactly as my calibrated Sekonic 758-DR Lightmeter told me
- I profiled the photo of the ColorChecker Passport and applied the profile
- I verified the custom white balance (although I already custom made it in camera via the Passport)
- I downloaded and opened this PDF: http://www.babelcolor.com/download/RGB%20Coordinates%20of%20the%20Macb eth%20ColorChecker.pdf
- I adjusted all the controls 'till the numbers matched in LR4 with this settings:
Exposure: 0,00
Contrast: -33
Highlights: 0
Shadows: 0
Whites: 0
Blacks: +25
Clarity: 0
Vibrance: 0
Saturation: 0
Tone Curve: Linear
The values on my ColorChecker matched the values in the PDF. Of course, it looks more dull. I saved those settings as a new preset. I applied it on other images with the ColorChecker (with different backgrounds but every image with perfect light metering) and the values were consistent as stated in the PDF.
I calibrated my screen at D65 2.2 and have calibrated norm lights in my post processing rooms. Comparing the ColorChecker with my calibrated screen and the adjustments in LR4 gives me an exact match.
PV2012 gives me a good and flexible starting point, I'm happy with it.
sandy_mc wrote:
The problem I have is that the basic controls just aren't predictable, at least to me. I adjust for one thing, and three other issues spring up in the other corner, quite unexpectedly.
Now maybe after a few more weeks, this will go away...
It may get tweaked, but it will never go away, no matter how many weeks you use it.
(these are the "Type 2" toning issues mentioned above)
The whole "adaptive" thing hit me recently when I switched process versions and the exact same tone went brighter in one section of the photo, and dimmer in another.
The place where it went brighter, was near a darker place, and the place it went darker was near a lighter place. This is the local adaptive behavior that is responsible for at least part of Lr4's magic, and a part of it's seeming squirrelyness, and uneven-ness.
It's like a very nicely done but more subtle implementation of Topaz Adjust - built-in.
I'm thinking maybe a two layer system - The PV2010 rendition on one layer, and the PV2012 rendition on another - probably a bad idea, but at the moment, I don't see a way to have our cake and eat it too... (other than by painting locals)
PS - I almost always use Topaz Adjust in conjunction with a layer mask, when I use it, which I confess is not too often anymore.
It will probably be stacked with Noise Ninja (stacked when Lr3 was released) after Lr4 is released.
Cheers,
Rob
TK2142 wrote:
Have you seen the example in this thread:
http://forums.adobe.com/message/4179562
?
The issue with that photo is what I identified as Type 1 earlier in this thread
i.e. it had nothing to do with the local adaptive stuff, and everything to do with the basic control set not having enough flexibility.
Thanks for bringing it up TK.
Cheers,
Rob
From:"Rob Cole
So, if you toy with both process versions, and include a point curve
manipulation in the PV2010 version and make a snapshot of it, then switch
to PV2012 and back to PV2010 via the snapshot, then you'll lose your point
curve. I haven't verified whether the problem is saving or restoring the
snapshot, so it could be the same problem would occur if snapshot was
saved in Lr3 - dunno.
Isn't that why LR has virtual copies? So that you can play with a different
develop process without upsetting your existing one?
Bob Frost
Andrew Rodney wrote:
Why not look at the Lightroom AND Photoshop splash screens for a hint.
Andrew, that would be way too obvious ... I just wanted to contrast the search results for a well-known individual who has been a integral component in the development and production of Photoshop and Lightroom ... to ... an anonymous forum poster fixated on the credibility of a job title ... which really has little to do with this discussion or the advancements in the Lr product ...
Why some think it more important to focus on speculation and conjecture ... or the personal motivations of Adobe employess ... rather than the software itself escapes me ...
A follow up question for Eric,
Putting aside the technology, can you explain whether LR makes a different creative assumption than LR3 in terms of a starting point. For instance it's apparent, and many have commented on the highlight recovery change. While it can be argued that LR4 develop controls may ultimately provide the same or greater range of aesthetic control, I'm keen to know whether your default "treatment" has changed.
Let's say I like to shoot high key and don't mind some blown highlights. Will achieving that look be at odds with LR4's default assumptions, and require more effort to achieve?
I believe this is related to Rob's questions. As a workflow product initially conceived to address volume (in contrast to PS's pursuit of ultimate creative expression for individual images) it is completely legitimate to be concerned about the throughput implications of the develop changes and whether the starting creative assumptions of LR4 are harder to undo than those of LR3. To be clear I'm not asking about the "zeroed baseline" that was discussed in other threads, I accept the fact that assumptions have to made, I just want you to characterize the assumptions of the LR4 process relative to LR3, if in fact they are different.
thanks
Rob Cole wrote:
It seems to me there are two different flavors of toning issues that could be talked about:
1. General toning flexibility given the basic control set (and shadow/highlight recovery processing - but I'll not be talking about that aspect in this post).
2. Toning characteristics of the "adaptive" algorithm.
To elaborate on what I meant by Type 1 toning flexibility.
As it stands:
- the shadows slider is critical (in conjunction with the blacks slider) for defining intra-shadow contrast.
- the shadows slider is critical (in conjunction with the highlights slider) for defining midtone contrast.
- the highlights slider is critical (in conjunction with exposure & whites) for intra-highlight definition.
After deleting badly shot images, the rest fall into one of these categories:
Category 1: When the needs for the shadow slider in the darks are the same as the needs of the shadow slider for the midtones, and the needs for the highlights slider for the mids are the same as the needs for the highlights slider in the uppper region, then the photo falls right into place quickly with no Type-1 issues, and can be tweaked easily without running into any Type-1 issues.
Category 2: When the needs for the shadow slider in the darks are compatible with the needs of the shadow slider for the midtones, and the needs for the highlights slider for the mids are compatible with the needs for the highlights slider in the uppper region, then after a few iterations, and maybe a bit of tone-curve manipulation, the photo can be satisfactorily edited, but then you'd better not ever touch it again, or you'll have to start over... ("you" in this context means "me" and others who have or will have the same problem eventually, even if not yet).
Category 3: When the needs for the shadow slider in the darks are incompatible with the needs of the shadow slider for the midtones, or the needs for the highlights slider for the mids are incompatible with the needs for the highlights slider in the uppper region, then it takes a heroic effort to balance the basics with the tone curve in order for the effects to be satisfactory, if its even possible at all. Locals required for satisfactory results.
Category 4: Photo can absolutely not be edited to satisfaction with Lr4b1.
I'm talking about my experience here - your experience may be different.
And yes, the contrast slider factors in too - I've tried to simplify in the hopes of making a point.
As always, I recommend experimenting with your own photos, to see whether any of this applies to you and your photography, or not.
My present estimate @Lr4b1:
- 80% fall in category 1.
- 15% fall in category 2. This is how I spend the bulk of my time when using Lr4b1.
- 4% fall in category 3. I've learned to spend far less time on these photos. After beating my head against the wall for a while, I head for the locals.
- 1% fall in category 4 (and I really don't care).
Looking forward to beta #2.
To quote Scoop-Nisker from KFOG/SF - "if you don't like the news, go make some of your own...".
Rob
ROb...
WIth all respect....you are not helping yourself....or us, for that matter.
TYpe 1, type2....category 1,2,3,4......?????
MAybe it is just me, but I have no idea what you are talking about....without some illustrative examples.
ALso...what is the reference to a "Beta#2"? Do you have some knowledge the rest of us do not have? I have no idea, but if I were a betting man, I would suspect....sort of a last second "Oh, sh*t"....beta2 will be LR 4.0
I have no knowlege of a second beta - it was an assumption. Could be Lr4-final is next for all I know...
-blacks and +shadows creates intra-shadow contrast.
-shadows and +highlights creates midtone contrast.
Note: the sign on the shadows adjustment above.
Not unlike the tone curve, one can only have so much intra-shadow contrast before it starts cutting into the mid-tone contrast, and vice versa.
(contrast slider also increases midtone contrast and reduces shadow and highlight contrast, it just does it differently, and I shan't be discussing how it factors into all this...)
My point was that whether or not one can get sufficient toning in all three major regions depends on the tonal distribution of the photo, and how that jives with PV2012, and of course what you want from the photo.
As I understand it, some people have had no trouble whatsoever getting adequate toning for any of their images using PV2012 - I envy them. I've not been so fortunate.
Here is one instance of what I call a Type-1 toning issue:
http://forums.adobe.com/message/4179562
PS - I have shared plenty of images with people privately, I have reasons why I don't want to do it in the open forum. Sorry if that makes it hard on you.
Legend: contrast = separation of tones.
Cheers,
Rob
Burn after reading:
-----------------------
When PV2010 first came out, I noticed a very small percentage of my photos looked better in PV2003 than PV2010 (having to do mostly with dark image tones being mistaken for color noise, but also sharpening artefacts - "the zipper effect" in PV2010). I decided to just move forward and forget about PV2003, since it usually wasn't a big enough deal (decreasing color NR and/or increasing color NR detail when bitten, or local desharpening, respectively). I have already prepared myself that the same will not be the case with PV2012, assuming no significant changes are made before final release - I will be keeping my eagle-eye out for the tell-tale signs of when PV2010 will produce superior results, and use it instead.
I know this is really gonna ruffle some of y'all's feathers, and although that is not the intent, my loyalty is with the people reading this thread trying to get the real skinny, instead of a glass of Kool-Aid.
So, stick that in your pipe and smoke it! ![]()
R
Rob Cole wrote:
I have already prepared myself that the same will not be the case with PV2012, assuming no significant changes are made before final release...
Reasons that's a stupid decision:
- First, the obvious. You have a big assumption in there, that's already known to be wrong.
- Second, you've spent years and years getting used to and comfortable with PV2010 and PV2003. You've spent only weeks messing with PV2012. You can't expect to learn all the tricks that quickly.
- Third, any tiny amount of experience you've obtained from the last weeks has been marred by a little bit of beta-itis.
- Fourth, you haven't been trying very hard to following the workflow suggestions you've been given. Once you do (if you do), you may find your conclusions were wrong.
- Lastly, it's not necessary to duplicate the "look" of PV2010 if you can get an equivalently good or superior but different look in PV2012.
North America
Europe, Middle East and Africa
Asia Pacific