Intelligent Masking

Even I have seen most of these graphics software since I was responsible during some years in the early nineteenth for all Windows software product and in particular all Microsoft software as a technical product manager for Microsoft-products at first Sweden´s biggest Software distributor Esselte Datasoft that later became a Nordic one under the name Scribona and I worked there as a software developer until 2009 (product management info and pricing business systems)

When I started to be interested in digital photography, I was forced around 2006 or 2007 to use Lightroom 1.0 since Adobe made a hostile tale over of the Danish rising star on the RAW-converter software market called Pixmantec. Adobe feared that RAWShooter was going to threaten their market dominance so they just killed it and forced us to migrate without offering any migration path at all. We all lost all of our work that we had done in RAWShooter. I have never forgiven them for that. Both Lightroom 1.0 and 2.X was terrible and got usable first in version 3.X. - especially their useless preview rendering

I started to use Photoshop around 1992-93 when digital cameras wasn´t even thought of and pictures for postprocessing had to be scanned in TIFF. Today is 2023 and Photoshop has in these 30+ years undergone a constant structural growth. Demands on demands have resulted in a software so overloaded with functions that it´s just rediculous to use for normal modern postprocessing for digital photographers - if you are just to polish a RAW image. Photoshop is a tool optimized for creators and graphics artists that do something else with the images than just a few tweaks.

Digital postprocessing is all about RAW but Photoshop was designed to be used with TIFF that is a format originally designed by Aldus Corp. in the mid-eighties to be used as a standard image format for scanners. When Adobe acquired Aldus and the desktop publishing software Pagemaker they also acquired TIFF. The rest is history.

You wrote:
“Why would anyone try to use any of the making tools in any of the programs designed primerally for processing raw files not precision masking and essentially “retouching” type work? Why not just use dedicated program for that, like Photoshop?”

The simple answer to that question is that Photoshop is not a tool optimized to be used by digital photographers - it is so much more and also so much less and most of that more is irrelevant in this discussion and for the digital photographers. Digital photographers capture images in RAW that is a far more suitable format than TIFF is in digital postprocessing. Especially in DXO Photolab of all softwares, which is so centered around the world’s best “noise reduction”-system Deep Prime. Deep Prime and Deep Prime XD can´t be used on TIFF or even serialized (RGB) DNG. So of that reason using TIFF as a format in Photolab is just counter productive.

I think most photographers today tries to avoid TIFF as an intermediate file format and I really hate the so called “integration” with DXO´s own NIC Collection and Photolab because that always push us into using that terrible old scanner format. TIFF is forced on even Photolab users because NIK and even PureRaw is adapted to the old TIFF limitation of Photolab. Of this reason I think it is perfectly logical for the users of RAW-converters to ask for effective and intelligent masking tools in their converters, because that increases their efficiency.

Adding an unnecessary intermediate file format like old, bulky and inefficient TIFF in our workflows because of some ancient limitations in Photoshop, doesn´t sound like an all that smart idea these days (a tool photographers today mostly really don´t need when using a competent RAW-converter like Capture One or Lightroom). TIFF is nothing you like to have to day in modern Digital Asset Management workflows either. A modern RAW-converter today is optimized to process RAW and that´s why it´s perfectly logical to add all the modern smart intelligent masking tools exactly there so the users don´t need to take an inefficient unnecessary round trip to old Photoshop just of that reason.

If DNG had been better implemented in both CO and Photolab it could have been a better alternative but I can´t use it in my flow since Photolab doesn´t accept hardly any DNG-files than the ones it self exports. Unserialized DNG from CO can´t be read into Photolab. So in that case we are still forced to use TIFF and then we have to apply Deep Prime on the RAW before we export a TIFF since it doesn´t work on TIFF directly. It´s far from ideal.

I think you made a typo. That sentence does not make sense. What year did you mean I do not know, but it was not 2023.

Well, let me put it this way. If you are not a decently skilled user of Photoshop in 2023 and you are professional photographer. I am curious how do you make a living from it. Also if you have been around for so long, why aren’t you using Photoshop in your daily work?

And while Photoshop has indeed been expanded for a wide user base, it had tools for working with photographs for many decades now. And its not just an industry standard, but “Photoshopped” is an a Adjective and cultural staple.

My point is that I’m not talking about some obscure niche software. Photoshop is pretty much a standard even for beginners and hobbyists and amateurs as well as professionals.

No, it is not. Digital post processing is whatever processing needs to be done for desired final result. RAW files are just the start for many, a format that needs to be developed. If that is all you need, than that is fine, but since you are complaining about more advance masking and claiming RAW is all people do in digital workflow, I must wonder… what kind of place did you work at. Because it does not sound like any place I worked at… or worked with.

What? That makes zero sense. RAW file needs to be developed into sharable formats like TIFF, JPEG etc. But the idea that everything must be done in raw and TIFF or JPEG is just delivery format, is absurd. Maybe for some people, but everyone for wedding photographers, landscape, architectural, fine art, fashion and beauty, commercial work, product photography and just about everything else, retouch their images, and virtually all use Photoshop and TIFF or JPEG to do it.

There are people who shoot JPEG and deliver it, mostly for fast delivery in photojournalism or sports photography, everyone else usually shoots RAW, usually finishes in JPEG or TIFF or PSD and delivers for web in formats like JPEG, WebP or PNG. And sometimes TIFF is used for archival purposes or for high end prints or client delivery if they ask for it.

That also makes no sense. TIFF is not useless format, its a very important archival format, compatible format for wide variety of applications and supports layers and compression that RAW files cannot. Also if you are doing raster based adjustments, you cannot use RAW format, which is neither designed to be good for that, nor it needs to be. Clearly you are missing out on lot of thing happening in photography business if you have such low opinion of TIFF.

Is that so? I have a feeling your workflow than is very limiting indeed. I also don’t have a clue what you have been doing all these years to develop this kind of hatred for TIFF format, but its just not true what you say. The reasons for it are about a mile long.

If you have minimal needs than maybe that is enough for you. But it simply not true for vast majority of workflows and users out there. Also as you said yourself… “RAW-converter”… means just that. RAW format was always meant to give user the control over how data will be developed and processed at the expense of larger file size, more taxing processing and limitations of raw format itself. Often its just among first steps before further processing is done in Photoshop or similar application.

If you are a minimalist you can probably get away with just minimal work in RAW applications itself, but I think it would be hard to find working photographers who only rely on RAW processing and are still competitive. Maybe there are out there, but I very much doubt they are the norm. In fact I know they are not. Simply because of the demands of the market. And with AI generative imagery etc. For the foreseeable future even traditional work will be challenged. I’m sure you won’t be faking a bride throwing a bouquet, but you might retouch the image. Actually I’m sure you will. And you will be using tools that fit the purpose which for the time being means almost certainly Photoshop. Try working in commercial photography filed or product photography and get it all in camera. Try fashion or beauty work and see how that goes. Maybe you will get some crazy activists screaming “no Photoshop” etc, but that is a different story.

Bottom line is that all this is tools. What matters is final image, How you got there is up to you.

As for authenticity. I’ll use a quote: "The true authenticity of photographs for me is that they usually manipulate and lie about what is in front of the camera, but never lie about the intentions behind the camera.” ― Wolfgang Tillmans

1 Like

Sorry it should be 1992-93

I´m not a professional photographer in a common sense and have never been but I have used a lot of my images in my profession as a teacher educating teenagers during 9-10 years. I was a backpacker between 1971 and 1986 so I had a lot of useful images. I have around 70 000 images a lot of them are from my journeys.

1 Like

That would be more reasonable. Yes.

Well, that’s fine. I mean travel photography, family photos, even to some extend street photography, all that is fine. In most genres of photography where people make a living from it, demands are higher and so tools being used follow the need of the profession.

Even something simple as corporate head shots, high school photography. I forget the name, but taking photos of students graduating, even simple project like that may require some retouching. Usually Photoshop is the industry standard for that.

The more demanding photography requires more and so Photoshop becomes as indispensable as the camera or tripod or lighting kit or lenses, or backup drives etc. its just part of the process. .

Here is some basic retouching I did, that is quite standard in my experience. There is just no tools I can use in any of the RAW processing programs on the market to do this. While its a quick Photoshop job.

Lets say you have a client who wants some portrait shots of himself for a magazine, or social media to promote his singing career. They want to look their best and if its just a snapshot they can use their photo, so they need a photography service + retouching to the extent it is need.

On the other hand when I go for personal travel just to take some snapshots, I don’t need to worry about that. Unless I want to.

In this case I was talking a walk and took some casual shots, but why not enhance it in post to improve the image. I did it for my own desire.

Here I didn’t have fast lens for the pigeon shot, so I faked shallow dept of field in the post. But I think it helps the pigeon stand out. So why not.

I can do these things in Photoshop faster and easier than I could even in something like Luminar Neo which is heavily reliant on AI and it too I think also converts RAW files to JPEG or TIFF , but they might have changed that in latest version, I’m not sure. Anyway, pick the best tool for the job, be authentic in your intent and deliver the best image you can.

DXO is a great tool and I love it, but it can never replace Photoshop nor should it. Its apples and oranges.

Between 2009 and 2016 i was working at the City Museum of Stockholm. I was the technical project leader and also the main developer of “The Digital City Museum of Stockholm” I designed the integration with the museums old SQL-metadatadatabases and the Norwegian made DAM-system Fotoware and most of the automated workflows.

The museum has about 4 miljon historical images mostly in TIF of historical reasons (around 30 000 are digitized today) some examples are as big as 1 GB per image. Maybe 40 000 is digitally born and they are stored in the “newer” museum archive standard DNG. DNG is the new common archive standard for “RAW”-data used in most countries today.

I have seven years of working even with TIFF in DAM workflows and they are just terribly ineffective of two reasons. They are mostly very big 16 bit files and on top of that they lack a well defined metadata header like the ones present in JPEG or DNG. That´s why they are completely hopeless in metadata workflows. Believe me in DAM-systems no one use anything else than JPEG-files as metadata owners.

We had three type of files:
Base-files as is in TIFF or DNG (digitally born)
Delivery files (all pixels)
Sketch-files (1280 pixels on the long side)

When a TIFF Basefile was sent in to the automation hub it had som basic metadata on it. The file had a museum specific image number it had been registered in the old image SQL-databases. When this image was detected, it was matched with the database on image number and if it was registered it got all the matching metadata in the database automatically written to it.

Then there was a JPEG full size automatically created with all that metadata and finally a smaller Sketch-file was created and that file was then the owner of the metadata because the smaller files are are far more efficient in the DAM-workflows.

If the RAW was a DNG it had been developed by the photographers and saved with a full size developed embedded JPEG and when that file was processed the DNG was saved as it was as a Base-file. Then the automation hub copied the JPEG in the DNG and saved it as a full-size Delivery-file. Then there was a 1280 pixel file created with the metadata from the Base-file. So if you have read tis you might understand why most Cultural Heritage-institutions all over the world standardize on DNG these days.

If somebody wanted to download a Deliver-file they picked one of the Sketch-files displayed in the web-application and pushed the button for download. The system then picked the corresponding Delivery-file and then pushed the master metadata from the Sketch-file into it before it was downloaded. There were nothing else than small Sketchfiles (1280 pixels) handled in all of the metadata flows. That is best practise in all the DAM-industry.

I have been working very close to the museums photo archvarians of the museum when developing these systems. Some of the photographers have been using Photoshop but far fram all. They used Lightroom since it is the software to use working with DNG. I have also quite a lot of experiences from participating in the FotoWare metadata sphere since we for some years was their reference site because we then had the worlds most advanced integration setup using Fotowares DAM. We had both local visitors and people from the Tsingua University in Beijing and we even helped Ford Foundation with their studies of Photowares possibilities.

So I can say I have extensive professional experience not just with my own workflows but also some used in the City of Stockholm still today, because this system is still up and running after 10 years.

Digitala Stadsmuseet (stockholm.se)

This is a real DAM that handles not just images but also Documents and Publications as well as images take of Artefacts and Art, which makes it far more complicated since it has to handle not just one metadata schema but a handful in the same webinterface.

From the quick glance at the website you shown it seem the image uploaded to the web are JPEG, so I assume the archived ones were TIFF. TIFF is a good archival format because of the data it can store, and its long history means its compatible with virtually all programs. I don’t know about the specifics on the database and metadata you mentioned, but personally I’ve never had a problem with it.

I don’t know what specific requirements of workflow was allowed, so I can’t really say what the best format would be for that particular situation, since most goverment administrations are pretty inefficient as it is.

“Bureaucracy defends the status quo long past the time when the quo has lost its status.” – Laurence J Peter

That being said, TIFF doubles down both as archival format, working format, because it supports so many operations and is compatible with so many programs, and can be use as delivery format when one needs lossless compression.

DNG was developed by adobe as digital negative of sorts, if I’m not mistaken its also open source. The main draw back is that its not as compatible as TIFF since it has not been around for as long and does not support layers etc as TIFF does. If DNG is new standard for archives, it makes sense. Still not a good working format as TIFF , so its probably mostly for archive. Serving as a container.

The problem with TIFF in this case was that Stockholm had such a bad deal with Volvo IT at that time that it costed about 10 U$ to store just one single 1 GB image. When the museum had used the system it found that the cost for storage threatened the budget. One of our photographers said after a meeting when discussing this that maybe we shall skip loading up a lot of images because doing so will just resulting in us losing our jobs.

We suggested far better options but the city was not allowed by the agreement with Volvo to do so t that time. So using other more efficient formats kan really save a lot of money and even in that respect uncompressed TIFF could be a pretty heavy burden.

… I don´t have any other opinion than you when it comes to the application compatibility of TIFF. It´s excellent and for DNG it´s not all the time and especially not with Photolab.

… and I personally don´t see any need at all to use Photoshop when using the last version of Capture One for anything I do. It has all I need even for retouch and the relatively new concept of style brushes has really been a game changer. Before there was a process where you always manually had to create the layer first and then pick a tool. Now you just pick one of 20+ style brushes and the start to apply it since the layer is crated automatically in the background. I believe most digital photographers would be perfectly happy if we could reach that level even in Photoware. I prefer an approach like that before the inflexible Control Points in Photolab.

This will also become more and more obvious since the value of all images now have been severely depreciated by AI. More and more professional photographers will find that they just can´t spend as much time as before on postprocessing images and that´s also why there will be more and more important with acceptable technical quality on the JPEG-files that comes right out of the camera.

The only case I still use TIFF lately is with NIK Collection (Efex Pro) or when scanning old historical images to my relatives in America. The big thing with DNG is that it´s absolutely fantastic as a workflow format especially in Fotoware DAM-systems since they have developed a few processes that really can take advantage of that DNG kan host both RAW-data, embedded JPEG in full size and even embedded RAW-files if you like. So it can contain and keep together both the original RAW, The DNG RAW-data, full size JPEG as the photographer wanted the JPEG to be developed. It has properties very useful for handling the authenticy. No other format what I know manages that.

I can tell you that some of the really heavyTIFF-files we had at the City Museum managed to get the automation hub to stall and that because a combination of the weight of the file and the lack of a well defined XMP-metadata header in TIFF-files that caused the system to scan the file for a place to write the metadata. So, we still have to live with TIFF in some DAM-environments of historical reasons but sometimes it´s a real pain to need to have it in the work flows. That´s why I really hate TIFF. It will still be my last choise.

One thing I didn´t understand was that you still seem to be fine using dull general tools in Photolab as far as it is letting you do it and eventually having to export a TIFF in order to finish the job in Photoshop instead of having a set of appropriate tools that let you do it all in Photolab. There isn´t really all that much missing if they don´t have to rewrite a large part of the application to achive this.

That’s because as I’ve mentioned before, they are tools. Not every tool is goo for every job. So I tend to pick the tool that either I’m most comfortable or is best suited for the job.

In the case of masking tools in PhotoLab, they are quite useful for luminance/color based selections that feel more organic than hard edge masking. If I need to adjust something that requires tonal or color adjustment selectively and other tools like various sliders in PhotoLab are not enough, I would use Local Adjustments, which are quite powerful. If I need a really hard edge precise mask or do more complex masking or retouching I would jump to Photoshop where I have all the other tools and plug ins.

For example, take this casual shot of a cat. So, I’ve processed it in DXO, and all I need to do is recover some of that fur details in the white areas. The best and easiest way to do it is in the RAW format, with control point. Because it will use luminance and color info to create a nice luminocity mask that feels organic and recovers details. If I was in Lightroom I would have to manually paint which won’t produce as smooth results and select subject would select entire cat, which is not what I want. Here its just a control point.

But than again, I would still take it to Photoshop because I want to take this further.

The tools are fine if you use them for what they are meant to be used.

But just for fun, I’ve tried to use the Auto Mask brush with edge detection and control points for more precise masking and works well too, but I would do it faster in Photoshop or lightroom probably.

Changing the two things into orange for example.

The one on the left is done with auto mask brush the one on the right is done with control points.

2 Likes

… and I tend do use Capture One when I see that Photolab will not be the right software to use but still I hope that I would be able to do it all in Photolab soon because 9 times of 10 I get a result I´m pleased with faster and more efficient than with Capture One. The Control Points can often be very efficient - as long as a high level of precision is not required. That´s why I call them QD-tools.

Yes. I don’t think control points were meant to provide precision masks as much as lose masks for localized corrections which feel like it could be normal tonality of the image. Basically it is more similar to so called luminosity masks we sometimes use in Photoshop. Once someone explained it to me as using the image to select itself. Meaning its not about selecting objects in the image per se, but using color and tone of the image as base for selection. That is what was meant with using image to select itself.

Similar to how you might go to channels pallet in Photoshop and CTRL mouse click on one of the channels like red channel and you would get selection of 50-100% luminosity value of the red channel. Brilliant for certain types of adjustments to the image that feel natural. On the other hand if you wanted to selection a particular person in the image, than these types of methods do not work.

Control points fall in the middle ground because you are using color and tone or brightness for selections like you would in Photoshop, but you have a bit more control over where in the screen do you want to put the control point. Also remember, you can use “-” control point to subtract for areas you don’t want , letting you target quite precise ares, but still based on color and tone information.

When it comes to hard edge selections its less appropriate tool, but we can manually use auto mask brush , although its less efficient than C1 or Lightroom tools for that kind of work. On the other hand all these programs offer so many sliders for color and tone manipulations that local adjustments may not even be needed sometimes.

1 Like

Here is another example.

Lets say I want to change the color of that well camouflaged little bird. This one.

Lets say I want to make it pink to show it well in this example. Like this.

I can simply use few brush strokes from a Auto Mask Brush and viola!

PhotoLab has some pretty good masking tools that are not outdated, they just require different approach and you can do a lot with it. .

Also I can easily duplicate the mask I just made and invert the duplicate and make everything but the little bird underexposed. Also quite easy. I would not want to do this, but its for making a point that tools are quite powerful, if you know how to use them and you use them for what they are meant to be used.

I am sorry to say this, but it feels like a good label for our conversations recently. It was an illustrative example based on an image I searched for online. I do not own such a chessboard.

I leave my point standing. There are images (not shown) where existing masking tools cannot make out what some AI tools can. That’s it. That’s the point.

1 Like

Thank you very much for clearing the copyright question :grin:. 33 posts after posting an image which was not yours. My fault for not having guessed. I don’t disagree with the point you just made, but as I made fun of the artifacts, my point is JPGs are not the best material for some masking tools.

When Adobe introduced “context sensitive fill” or other “AI”-labeled features and I tried them (I’m far away from being Photoshop or AI expert) I found it an interesting alternative to the more traditional “dumb” tools. It appears that these tools have just other weak spots. Like a hammer not to be a good tool for knitting.

I’m rarely using Photolab now. Lightroom’s masking has become so fast and easy that it’s now my editor of choice. Adaptive custom presets take it to a new level. The new noise reduction is also close to DXO’s.

2 Likes

Not every tool. The brush and selection masks are user-controlled and don’t depend on colour or tone. And they have been there forever. I agree the magic brush selection tool is not great compared to Photoshop but it is there for the times Control points/lines don’t work. And the ultimate fallback is the simple brush.
But Photolab has never pretended to be a single solution for photo editing. It is above all a raw converter and assumes you will go into Photoshop or Affinty or whatever for final “creative” editing. That’s why it doesn’t do panoramics, or HDR brackets or focus stacking etc.

1 Like

… which needs to be improved
(straight lines, option to fix/secure the mask against accidental movement,
better tablet/Wacom support and …)

2 Likes

Of course professionals delivering originals for prints in glossy magazines needs the best of tools and need to put in a lot of work doing their best and I also hate sloppy work and poor technical image quality and Photolab usually delivers a fantastic quality superior to the most but sometimes even the media used limits what can be expected. Especially old digitized analog color film can be pretty demanding.

… but my point is that you don´t really need Photoshop even for som basic retouching nowadays and haven´t done so for at least a decade. Already in Lightroom 5.X I remember I used SLR Lounge presets for Adjustment Brush. SLR did not just add a lot of common presets but also added som nice and very useful brushes to the Adjustment Bruhes in Lightroom (or what ever the corresponding tools are called in Lightroom). This is many years ago (5.x was released 2013 or so).

I used it for retouching a wedding I shot at that time and I used it for a baptising too I remember.

Many years I have read that Lightroom didn´t do layers but that wasn´t really true. Lightroom has had layers for ages but not like in Capture One or Photoshop, It has been layers with zero administration. Since Lightroom then and now always have been about lowering the bar for the users, almost all that has been hidden behind the interface but everytime you open Adjustment Brush and start to use a brush a new layer is created in the background and a little grey dot is added on your motif where you start using it.

Nowadays I use Capture One mostly when I make my series of “animal portraits” that is still in progress. I don´t think Photolab has been up to masking hairy edges or feathered edges on birds. Capture One with its much more sophisticated “refining” tools is far more suitable for that kind of work than Photolab. I work on these series of animal portraits from time to time but animals are really not my main thing in general, so it´s not my main focus.

I am interested in animals though but more as individuals - I am convinced that even animals have souls and it´s really mostly the images I have that reflect that, I find interesting.

The animal portrait thing has just happened since I have been many times mainly in East Africa. I´m really more interested in “people doing things” - social life, ethnical differences and architecture, history and religion (despite I´m not at all religious). I have a lot of old color slides I digitize.

Street life in Katmandu 1976:

Below an image of an old monk taken in the old kingdom of Ladakh, Kashmmir, India taken 1978. Color slide film is about the worst media to handle. Lost colors, and impossible to sharpen properly in Photolab but that can be done in Capture One! This image processed in PL 6. No need for any special masking tools there.

Religios motif fron Kandy Sri Lanka 1976

The reason I have used a brownish tone on these images is that many of my old Agfa CT 18 slides (ISO 50 or DIN 18) have lost the whole green RGB channel. I had hard to justify the time it would take to try to restore the colors properly.

Nice photos. The crop is a bit strange to me, as if something is missing from the frame, but other than that, good stuff. I love the sepia ones, nice warm colors. I also like dark mood in-general, so kudos for the Elephant image.

About Lightroom and retouching. While the tools have been expanded over the years, it is really more for people who are not really retouchers than for those that are not. Yes, you can apply selective color and tone adjustments and yes you can do some basic clone and heal things, but it is more of a crutch than replacement for dedicated tools. The main pain point being that it slows down Lightroom considerably as you add more brush strokes or masks. It also lacks precision considering the speed and resources. Adjusting a pupil, should be fine, doing more than that, things get slow.

There are some good AI powered tools for specific retouching jobs that have gotten pretty good, its called Retouch4Me, made by a guy from Estonia. I think it can be used from Lightroom as a plug in.

The plug ins themselves are quite pricy for casual users, but worth it for people who offer retouching services.

  • Retouch4me Heal $124
  • Retouch4me Dodge&Burn $149
  • Retouch4me Eyes Bundle (Eye Vessels + Eye Brilliance) $124
  • Retouch4me Portrait Volumes $124
  • Retouch4me Clean Backdrop $149
  • Retouch4me Skin Tone $124
  • Retouch4me White Teeth $124
  • Retouch4me Fabric $124
  • Retouch4me Skin Mask $124
  • Retouch4me Mattifier $124

It is a one-time purchase of the perpetual license with free updates. 3 keys for different devices of one user.

Reminds me of a project I did many moons ago with an elephant in the city, composite as an ad.

01_0000_Layer Comp 1

01_0001_Layer Comp 2

Obviously, its a different kind of work and Lighroom can be used for prep work, but not compositing. Today with AI maybe that can be automated, but this was from 2009 I think.

There was another one, also went with sepia.

My point is that, these are juts tools, and we try to express ourselves in our own way, using the tools that allows us to do so, and hopefully enjoy the process. Some might only need Lightroom or Capture One or DXO for everything. For others is just one more tool in the workflow and not every job is the same. Sometimes you need more and something you need less. I don’t think either of it is wrong.

Its about being creative. And if the tools you use do the job you want/need great. If not, find a tool that does the job.

Art Quotes"I have always believed that a large part of the beauty of a picture arises from the struggle which an artist wages with his limited medium." — Matisse on Art By Henri Matisse, Jack D. Flam

Robert Genn: Many a fine style has evolved from a decent handicap.

"The higher the obstruction, the more single-minded the problem, the more the creative mind is challenged.”
—Leon Jacobs

The greatest enemy of art is the absence of limitations.

“Progress in art does not consist in reducing limitations, but in knowing them better. It is the limitation of means that determines style, gives rise to new forms and makes creativity possible.” ― Georges Braque (1882–1963), major 20th-century French painter, collagist, draughtsman, printmaker and sculptor

“I have always believed that a large part of the beauty of a picture arises from the struggle which an artist wages with his limited medium.”
— Matisse on Art By Henri Matisse, Jack D. Flam

“We demand too much of technology and not enough of ourselves”.

"Great art is the outward expression of an inner life of the artist, and this innerlife will result in his personal vision of the world.”— Edward Hopper, prominent American realist painter (1882 – 1967)

Cheers!

@MSmithy
Lovely!
I really liked your elephant!

You are absolutely right about the elephant. That image is in fact not cropped at all it´s standard 3:2 but 4:3 would be a better proportion in that case and often is. … and other time a square might be the best:

Thanks for all the quotes and the take on limitations. It took me about two years to create a process to handle my Afga images and the only tool that really could solve it I think was Photolab and especially the presens of three flavors of contrast and especially the total absence that I created of Microcontrast and the possibility to meet up with Fine Contrast instead. That solved a lot of my earlier problems with unclean images.

By the way, terribly expensive retouch tools. I think SLR Lounges were far more inexpensive.

1 Like

Changing the color on the dark green pillars in the image above might work even with the pipette selection tool in the Color Wheel but maybe not in this case since the color of the window frames are pretty much the same. Maybe one can limit the selection scoop in the Color Wheel. That is the problem with color-based selection methods.