Image analysis and editing

[DanK]

Some time ago one of my instructors made an analogy to the film days. He said that using a raw convertor / parametric editor is very much like taking a roll of film to the local photo lab, where the printer operator would adjust a few sliders and print the image.

This analogy hasn't been remotely accurate for many years. Photoshop is still a lot more powerful and flexible than Lightroom/ACR, but Lightroom/ACR has included a powerful suite of editing functions for a long time. I have images on exhibit that have been entirely edited in Lightroom.

The original file out of the raw editor is generally saved (as is the actual raw data), but frankly is somewhat irrelevant as a "snapshot" of the image data exported to the pixel based editor and in many cases, this is the base data that is worked with.

This is an illustration of why I don't like terms like "best practice". I would say that there are "best principles", but they can often be operationalized with different practices. Manfred has a very careful workflow. So do I. They aren't the same. In the case of this example, I never discard the original file unless I am certain that the photo is a reject. This partly reflects the fact that I have a different workflow. I use Photoshop when it's helpful, but I don't always use it, so I don't always have a base "original" photoshop file.

What is best practice in my book is not to discard data as long as there is any chance you might need it. This is a vestige of my decades of working with statistical data. The raw file is my data. Ditto, backing up. There are various ways of backing up. Manfred uses a RAID array, if I recall, for his local backup. I use an external drive that I mirror by hand. Both are "best practice" because they both protect the data. As long as you keep two current backups, one local and one offsite, whatever you do counts as best practice in my mind.

For this reason, I would never treat a photoshop file as the original unless the base layer that contains the parametric adjustments is saved as a smart object (which makes the file huge) orf you put the parametric adjustments into a copy of the base layer. Otherwise, you can't get back to your original data.

[Manfred M]

This analogy hasn't been remotely accurate for many years. Photoshop is still a lot more powerful and flexible than Lightroom/ACR, but Lightroom/ACR has included a powerful suite of editing functions for a long time. I have images on exhibit that have been entirely edited in Lightroom.

I agree Dan, which is why I tried to qualify my statement a bit, but neither the description of the tradition limitations of parametric editors nor the pixel based editors are true anymore. Like you I have prepared high quality, exhibition and / or competition ready images using only a parametric editor. I will also have to say that, in my case, these were studio shots where I had excellent control of both the subject and the lighting so only very basic "tweaks" were required.

The cross-over point of where one tool is a better choice than the other is getting less distinct with almost every major update to the packages. The modern parametric tools are excellent for global and some area and "local" adjustments. The pixel based editing tools are strong in all these areas too, but take longer to master, but they still shine in making highly precise adjustment down to the pixel level (if that is required).

[xpatUSA]

The original file out of the raw editor is generally saved (as is the actual raw data), but frankly is somewhat irrelevant as a "snapshot" of the image data exported to the pixel based editor and in many cases, this is the base data that is worked with.

This is an illustration of why I don't like terms like "best practice". I would say that there are "best principles", but they can often be operationalized with different practices. Manfred has a very careful workflow. So do I. They aren't the same. In the case of this example, I never discard the original file unless I am certain that the photo is a reject.

For this reason, I would never treat a photoshop file as the original unless the base layer that contains the parametric adjustments is saved as a smart object (which makes the file huge) orf you put the parametric adjustments into a copy of the base layer. Otherwise, you can't get back to your original data.

Ah - the "original file" to me is either the raw file or the OOC JPEG, - hence my earlier question and possible misunderstandings.

For myself, I have no fixed workflow. It depends entirely on a) the purpose of the final output and b) the camera being used. I might keep a raw if I anticipate further work for some totally different purpose. I never keep SOOC JPEGs since my output is usually JPEG - (occasionally PNG and rarely GIF). I mercilessly delete all intermediate TIFFs but will occasionally save a GIMP .xcf which I believe is more or less .psd but sans Smart Objects.

Life is simple for me because I don't print, don't sell, don't offer my snaps for serious judgement and I often defy convention.

[Manfred M]

Ah - the "original file" to me is either the raw file or the OOC JPEG, - hence my earlier question and possible misunderstandings.

For myself, I have no fixed workflow. It depends entirely on a) the purpose of the final output and b) the camera being used. I might keep a raw if I anticipate further work for some totally different purpose. I never keep SOOC JPEGs since my output is usually JPEG - (occasionally PNG and rarely GIF). I mercilessly delete all intermediate TIFFs but will occasionally save a GIMP .xcf which I believe is more or less .psd but sans Smart Objects.

Life is simple for me because I don't print, don't sell, don't offer my snaps for serious judgement and I often defy convention.

Context is important and the world "original file" is not particularly clear.

Right now I am watching some finish carpentry work being done on the house next door. Baseboards are being installed.

If we write about the "original material", is it the tree that is cut down for lumber, the rough cut and kiln dried wood coming out of the sawmill, the unfinished piece of wood that comes out of the the planing mill (or MDF factory) or the finished lengths of material picked up at my local Home Depot? I suspect the guy working and installing next door would probably use the last item in my list.

Context is important...

[xpatUSA]

Context is important and the [phrase] "original file" is not particularly clear.

Right now I am watching some finish-carpentry work being done on the house next door. Baseboards are being installed.

If we write about the "original material", is it the tree that is cut down for lumber, the rough cut and kiln dried wood coming out of the sawmill, the unfinished piece of wood that comes out of the the planing mill (or MDF factory) or the finished lengths of material picked up at my local Home Depot? I suspect the guy working and installing next door would probably use the last item in my list.

I gave you my interpretation and you gave me your different one by analogy and the guy's too - hopefully neither were in rebuttal to mine.

Context is important...

I know that.

Especially when people don't qualify a term or phrase. For example calling a too-bright image "over-exposed" or thinking that "resolution" is measured in MP. But I digress, sorry ...

[xpatUSA]

I gave you my interpretation and you gave me your different one by analogy and the guy's too - hopefully neither were in rebuttal to mine.

Still mulling over the use or abuse of "Original", probably ad nauseam by now.

Considering the previously referred-to "best practice" from raw:

The first occurrence of a "file" is when the camera writes the image data to the memory card. For example as P1001234.rw2. Then comes a conversion, say to P1001234.tif. Then comes an edit becoming a master file, say P1001234.psd. And then, say edited further for the web, P1001234.jpg. Each one, retained or not, is a "file" by definition. Four separate files.

For the life of me, I can not imagine any but the first occurrence of these files as being the "original file".

original [file] : That from which a copy, reproduction, or translation is made

i.e. any copy, reproduction or translation is not an original file.

https://www.merriam-webster.com/dictionary/original

Another analogy being the descriptive term for the first humans indigenous to Australia ...

[DanK]

I consider only the raw file to be the original file, that is, the original data. Everything else is derivative. There is a reason why it's called a raw file (not RAW). It's the raw data.

For what it's worth, I maintain virtually no JPEG files. Given that I use Lightroom as my database, it's trivial to re-produce one when I need it. When I want to post to Smugmug, Flickr, or here (via Smugmug), I use a plugin that creates a JPEG to my specs, uploads it, and deletes it. Occasionally I need to keep them temporarily for a specific reason, e.g., to upload to some other location, but that's a rare exception. I also don't convert a raw file to a stored TIF before starting in Photoshop (unless I have detoured via Zerene to stack. LR and PS create the needed file on the fly, and I store only what I produce once in Photoshop.

[amywebdesign]

I enjoyed the video and agreed with virtually all of the edits. It is a helpful video which I will watch again.

[Wandjina]

I generally don't recommend a lot of websites and videos I see on the internet, but I ran into this one last night and it is particularly good for a couple of reasons; he takes images that were submitted to a well known American photo competition and then discusses what could be improved and then demonstrates how to implement those improvements. He happens to use Capture One, but that is really more of a detail as these edits can be done with any other editing tool.

The person doing the talking is Kevin Raber. Kevin used to run the highly regarded photography website, Luminous Landscape (LuLa), after its founder, Michael Reichmann died a few years ago. He seems to have founded photopxl.com and took a number of LuLa contributors with him. Unlike LuLa, the new website is not a pay site and is free. Reichmann / LuLa's claim to fame is that he popularized the concept of Expose-to-th-Right (ETTR) a number of years ago in an article posted on LuLa.

I've watched his first four edits and he nailed the issues I had with the image (and missed a few, in my view), but regardless, he made significant improvements to each one. The important lesson is how he identifies the flaws; that in my experience is where most people struggle a bit.

The link to the article and video are here: https://photopxl.com/roberts-gallery...h-kevin-raber/

It is a long (over 1hr 15min) video, but is easy to watch in small pieces.

I think most people who present competition images at Club, National and International can learn something from this video. Thanks for the link Manfred.