Over-processing

[TonyW]

As a relatively inexperienced photographer compared to most of the people who contribute to this forum, I hesitate to be critical of the pictures that I see here because I don't feel qualified. However, there are a few places where I generally seem to have a different opinion to the majority.

Many of the images that I see seem to me to be over-processed. Often, the saturation seems to be too much, giving a rather garish look to the image. This is subjective and could depend on the monitor that is being used, of course.

Another issue is sharpening. This is another subject that I am trying to understand and I tend to not sharpen my own pictures enough or properly. However, when I look at others and sometimes my own, I see that sharpening has produced some unwanted effects. There is the halo that is commonly acknowledged but another that I have not seen mentioned is with textured things. For example, sharpening the image of a tree with small leaves seem to make the texture too granular where a softer look would be better.

[Mike Buckley]

You seem to have keen observation skills, Tony. There are lots of different post-processing styles and you have to decide what works best for you.

You mentioned that the monitor could have a lot to do with how an image appears. You can't do anything about other people's monitors but you can always use a calibrated monitor yourself.

My approach in the matters that you brought to light is to determine as much as possible whether a particular image seems to be accomplishing what the photographer hoped to accomplish, even if (especially if) his or her goals are different from mine. When the look of a photograph is consistent throughout, I assume the photographer has achieved the desired results. In that case, I probably won't mention when I think something is overly saturated or sharpened. However, if a particular aspect of the post-processing seems to be inconsistent with the rest of the post-processing, such as the presence of artifacts, I'll bring it to the photographer's attention. I appreciate it when others take that approach toward critiquing my photos.

Having said all that, there does seem to be a trend in digital photo forums to use a relatively large amount of contrast, sharpening and saturation. That's a style that "pops." The photos that don't have those characteristics do seem to stand out as a minority. Fortunately, those minority photos are usually recognized and praised for what they are and what they accomplish at least here at CiC.

[Colin Southern]

Many of the images that I see seem to me to be over-processed. Often, the saturation seems to be too much, giving a rather garish look to the image. This is subjective and could depend on the monitor that is being used, of course.

Hi Tony,

Are you viewing them on a calibrated and profiled monitor?

Another issue is sharpening. This is another subject that I am trying to understand and I tend to not sharpen my own pictures enough or properly. However, when I look at others and sometimes my own, I see that sharpening has produced some unwanted effects. There is the halo that is commonly acknowledged but another that I have not seen mentioned is with textured things. For example, sharpening the image of a tree with small leaves seem to make the texture too granular where a softer look would be better.

If sharpening is adding artifacts then it's being applied incorrectly. In most cases optimal sharpening requires at least 3 passes (often more). I wrote a little about it some time ago - you might find this interesting.

It's also important to evaluate sharpening with the image displayed at the correct size.

[Dave Humphries]

I agree on all points Tony.

Often the over processed look is what they wanted.

Really all you can do is ask in the relevant threads, so you can begin to get an impression whether there is a consensus agreeing with you (but no-one liked to say), or if your monitor might be over saturated.

Assessing sharpening is an area fraught with misunderstandings and often the viewer's browser usage/habits can render their opinions invalid, although they (and we) don't know it. Many just don't understand which factors are important and which are irrelevant.

Many will use the "Save for web" feature and either let that sharpen, or don't bother. In my experience, only those with a thorough understanding who do all the stages of PP themselves (and know why) and don't rely on time saving features, will get it right.

Unfortunately, even people that do all the correct things with the image, may upload it OK, but then link an image here which is not the size they uploaded, but one of the alternative sizes their web host generates (either without sharpening, or with too coarse sharpening) and thus "shoots their image in the foot"

[Manfred M]

A profiled and calibrated screen is of course the first step in evaluating someone's work, and while that ensures that the colours are as close to being correct as are possible with the screen, there seems to be no tool to really ensuring that the appropriate level of brightness is being used.. I find that LCD screens tend to be set too bright and the colours tend to look a bit oversaturated. The other issue is low end TN screens; to make up for technical limitations, the suppliers seem to have increased the saturation to make up for other limitations.

Sharpening is another issue that if fraught with challanges. I work on a larger (27") screen and when images look good on it, they look oversharpened on my tiny laptop. This is just a fact of life. I do the multi-step (3 steps minimum) that Colin refers to (we read the same book, I suspect). My native 7360 px x 4912 px camera resolution image gets downsampled to something less than the 1920 x 1280 native screen resolution in post (to see the whole image) , so the image looks sharp when I open it up, but at 100% I can see some softening from the capture process; that I correct for. I do local sharpening during post (as required, again working at 100% native resoluton). I do my output sharpening after i downsample to display or print size (again at 100% scaling). If I post a 1920 x 1280 image on the web and view it on my laptop, it screams oversharpened.

On top of all of the technical issues; photos tend to look less satuarated than what we see with our eyes / brains. Our screens are only capable of displaying dynamic ranges of 7 or 8 stops. Our cameras can be in the 13 stop range and our eyes (and brains) are over 20 stops. This means the on screen image does not have the "pop" that we saw, so working the image in post to show people the scene we remember will look too contrasty and over-saturated to some.

[Colin Southern]

our eyes (and brains) are over 20 stops.

People often say this, but I don't think they're comparing apples with apples in that I think the "over 20 stop thing" also includes DR after pupil diameter changes, whereas folks don't allow camera aperture (or shutterspeed) variations when comparing.

All I know is that when I walk down a dark path at night I can't see the surface of the track unless I shield the street light ahead with my hand.

[Mike Buckley]

This discussion should also take into account that viewing the same image file using different browsers results in different appearances.

[Manfred M]

People often say this, but I don't think they're comparing apples with apples in that I think the "over 20 stop thing" also includes DR after pupil diameter changes, whereas folks don't allow camera aperture (or shutterspeed) variations when comparing.

All I know is that when I walk down a dark path at night I can't see the surface of the track unless I shield the street light ahead with my hand.

The eye / brain combination do not work like a camera, over and above the iris providing variable exposure. The sensor is totally different and the zone of the eye that provides a sharp image is tiny when compared with a decent camera sensor. I can see highlights, where the camera blows out and see shadow detail where the camera does not. I don't know if the 20+ stop is right or not, but I do know that the camera does not give me anything nearly as vivid as I see, whatever the reason.

Trying to interpret what I see given the limitations with the camera, computer screen and sensor means certain shots (fall colours is one instance) will be worked to bring out more vivid colours. A portrait, especially a studio shot, will get a more subtle treatment.

[jcuknz]

People often say this, but I don't think they're comparing apples with apples in that I think the "over 20 stop thing" also includes DR after pupil diameter changes, whereas folks don't allow camera aperture (or shutterspeed) variations when comparing.

All I know is that when I walk down a dark path at night I can't see the surface of the track unless I shield the street light ahead with my hand.

The problem gets worse with age as the eye slows in adjusting .... why when I wore glasses I had clip-on polaroids with their bottom area cut off so the brightness of the road matched the inside of the car to check my speedo. Now without glasses [ post-cateract ops] I have graduated sunglases for the same reason.

Years ago when working with fewer camera Mp and I over sharpened an image and got the halo I went around with a 'one pixel brush' and cloned adjacent background material to cover it .... tedious but maybe OK if you consider the image worth it

[ucci]

Most of the above is beyond my rather simple brain. But it all rings true, especially as it reflects the opinions of people whose input I respect and value highly. Damn! Now what was I going to say? Oh yeah. Now I remember.

Read a little while ago an interesting comment. Some scientific wallah made a claim that colour per se doesn't really exist. What we see and interpret as colour, in terms of how we normally describe it such as hue, saturation etc., is really just a function of our individual minds. It is how our neuronal pathways transmit and interpret the various wavelengths that strike our rods and cones. (So then, does this mean if I get angry and I see red and then I bung on a bit of a blue, it is really just all in my mind? I just don't know as I am rather green in these matters.) If his theory has any credence, this would imply that the actual colour seen is all in the mind of the beholder and is not a hard and fast end point seen exactly by all which can be defined scientifically and with precision by the actual wavelength of the incident 'light.' So the perception of colour on a monitor is more to do with individual brain interpretation and processing of wavelength input rather than the calibration of the monitor.

This leads all me to a point that has always fascinated me over the years. And it is how can you easily sex people? ( No, Alphonse, now there is no need to be so disgusting! )What I have noticed that there is a particular shade of blue green in the turquoise range that if shown to women they will declare positively that it is ''blue. Men on the other hand will equally aver that it is 'green.' So whose brain is right?

And further to support this I recently saw a TV programme on mind trickery where different coloured squares were placed on a board. People were asked to nominate the colour of a yellow square centrally placed. The square was then removed and placed against a different background and coloured squares. The same people were then invited to nominate its colour. It was quite eerie and a bit unsettling for those oldies of us who cling to our perceived status quo. In both cases the colour of the square was quite different and neither looked yellow. Far from it. It was only when the square was removed after each viewing from the grid and its background that it could be seen to be a sort of daffodil yellow. And those who took part in this little experiment just couldn't believe their eyes, as they say.

Maybe there is a bit more to this colour perception bizzo than a highly tuned monitor firing on all cylinders?
And that is about it from the peanut gallery, aka me.

[Glenn NK]

This might be interesting - totally off the original topic.

http://clarkvision.com/imagedetail/eye-resolution.html


Back on topic: I agree with the OP, and at one time at least, it may have been a cultural thing.

Case in point: When I started photography in the early sixties, I pretty well used Kodachrome or Ektachrome.

There were a few European photographers where I worked, and they suggested I try some European films - Agfa comes to mind.

I was accustomed to the intense colours of transparency films made in the USA, and was disappointed in the somewhat muted colours of Agfachrome - particularly when I shot some fall colours.

This is not to say one is "better" or "right", but rather to suggest that tastes do vary, and they can be related to cultural exposure. (no pun intended).

Glenn

[Colin Southern]

Here's a "food for thought".

What would the doppler effect have on colour, if the subject or photographer were moving?

[ucci]

Here's a "food for thought".

What would the doppler effect have on colour, if the subject or photographer were moving?

Can't fool us with that one Colin. We know it would come out covered in spots, looking like the coat of a dopple horse.

[Colin Southern]

Can't fool us with that one Colin. We know it would come out covered in spots, looking like the coat of a dopple horse.

Close

I remember watching a program on what happens as we approach the speed of light - and weird colour changes were part of it. That's about all I know!

[Manfred M]

Close

I remember watching a program on what happens as we approach the speed of light - and weird colour changes were part of it. That's about all I know!

And time slows down as one approaches the speed of light (a.k.a. time dillation), so selecting the correct shutter speed is going to be even more challanging. Does one base this on the traveler's perspective or the subject matter perspective?

On the other hand, neither phenomenon is going to affect us real-world photographers...

[TonyW]

Here's a "food for thought".

What would the doppler effect have on colour, if the subject or photographer were moving?

According to astronomers it's a blue or red shift depending whether they are approaching or retreating. But what if it is illuminated by a by a monochromatic flash from the camera? What's the colour of your picture?

[DanK]

Tony,

Good eye. And I agree with you.

Re this:

For example, sharpening the image of a tree with small leaves seem to make the texture too granular where a softer look would be better.

This granularity is a classic sign of oversharpening. However, it could also arise for a different reason. An image has to be sharpened more for some output formats than others. In particular, one needs considerably more sharpening for an ink-jet print than for web display. If someone sharpened for a print and then posted that version, you would get the 'crunchiness' you described.

Re oversaturation, etc.: I agree. This seems to be far worse with landscape photography than other types. For example, if you look at macro forums, you will see much less of it. I think this goes along with the popularity of overdone HDR images, which sometimes remind me of paintings that people used to sell out of the back of vans in the parking lot of a shopping center we used to live near. (Sorry, I am being frank about my tastes, not tactful.) Software has made this trivially easy to do. This is one reason I rarely post any of my own landscape shots. By design, I try to make them more natural and avoid the pop from oversaturation and other artifacts, which makes them seem dull in the eyes of many viewers. One way to avoid this whole problem is to process the image in black and white, where one can't hide behind liberal use of the saturation slider.

Sorry to vent...

Dan

[Mike Buckley]

process the image in black and white, where one can't hide behind liberal use of the saturation slider.

Instead, we can hide behind the contrast or levels slider.

[Brev00]

It is good that you are approaching photography with a firm sense of your own taste. That will become increasingly important as you are exposed to more processing tools that will make the right balance of color seem like an almost random choice on your part. It can be so easy to overdo processing, it is a wonder anyone stays close to a subtle, natural approach. You see, even 'subtle and natural' can be an effect produced by an app or action. Flexibility is a guide word for me. Flexibility means I can go completely over the top with one image yet rein it in for the next. Sometimes, if I am two minds about one image, I will duplicate it and do two versions. Then, compare them later. Also, giving my image a few hours and revisiting can help. I can get immersed in the image and lose track.

I think it is important to find photographers you do admire and want to emulate. So, try to find a few that you like. Buy their books, visit their websites, become contacts. Learning from others is a major shortcut in all of this so don't let those you do not like take you off your path. I don't think there is a 'majority.' Just you and the other you are viewing at this time.