Pushing dynamic range in RAW images - is it cheating, is it wrong, is it just rubbish?

January 29, 2016  •  Leave a Comment

Firstly, apologies, I haven't had (or rather, spared) enough time to add content to this blog. I'll try and find more time, honest.

This topic started when another member of a motorcycle forum I frequent grumbled about the style of photography being used in a magazine he subscribed to. The following text is compiled from my responses to his grumble, so if at points the thread of the copy below wavers or is somewhat disjointed, then apologies, this wasn't written as an article, more a collection of thoughts.

 

Firstly the grumble, copied verbatim from the forum:

"Does anybody else dislike the photography filters they use ? If you've looked at the mag recently you'll probably know the style I mean, it's like the contrast has been pulled to emphasise black and white. Its almost cartoon and makes looking at details in the picture difficult.

It's really annoying"

 

It was then suggested that the image in question used HDR

 

This is the photo in question.

12593453_1015892198471368_67929712890839

This is a photo by, I believe, Chippy Wood who supplies Bike Magazine here in the UK with images.

 

This was my initial reply:

"Hmm, I'm guessing the answer to 'does this^ use HDR?' is 'kinda' (yes, in other words). It's not a straight out of the camera shot, by the looks of it, but then it doesn't look like true HDR.

HDR is usually a compilation of 3 images (taken in very short succession in a rapid burst, unless in a controlled light situation where more time can be taken), one taken with the light bit (sky in this case) metered correctly for light, the mid tones (the bikes) metered correctly, and then the dark bits (the wooded section, top left) metered correctly. The 3 images are then combined (either in camera or in post process in software such as Lightroom) to get the best of all three. Usually, in the hands of someone fairly ham-fisted, this results in a quite muddy image, usually with unrealistic and unnatural dark or light halos around objects. 

To my eye, it looks like this image has been taken just the once and then the image has been pushed and pulled in post (probably in Lightroom), it's quite a marmite effect, some like it, some don't. Some images it can work really well, others it can look a bit crap. It all depends on what effect the creator wants to make. It may just be that the photographer made a mistake on the day when taking the images and went to rescue the image in post, though I very much doubt it.

If it were me I'd pull in the sliders a bit and give the colours a bit more saturation. But that's me. There is no right or wrong way, only personal preference.

A good example of this is that one of the photographers I sometime tune into occasionally posts up an image of his and invites his audience to crop and process it to see how they would interpret his work. It's quite fascinating to see the varied results.

Here is a link to one of the RAW challenges. It's from the pages of Photo.gp, who amongst others supply images to the MotoGP Motomatters site."

 

Another forum member likened the image in question to the "image equivalent of a "smile" setting on an 80's Audio Graphic Equaliser"

I replied:

"That's the thing about post processing, it's down to the eye of the individual operating the software and how they want their creation to look, or perhaps down to how jaded they were when processing hundreds of images.

For example, the following two images I took last year at the Classic within a couple of hours of each other. Depending on what my camera settings were and if I'd got them 'bang on' or just 'near enough' I could push what the camera naturally records to pick out more of what can be seen. This also goes the other way in post processing in how far I want to push it all for the effect I wish to get.

This is Rutter crew trying to figure out an issue that Michael's just returned with.

p1459774485-5.jpg

Despite what you see it all happened really quickly, the crew being really quite animated, moving around quickly and in difficult lighting (going from bright sunshine into dark shadow and back again), so my camera settings were a little bit off and I had to rescue the image a little to get good overall lighting.

This on the other hand, is Maria Costello awaiting her turn on the start line on Glencrutchery Rd. I had a little more time to get this image, and quite a few opportunities to refine my settings. I'd got the camera settings absolutely bang on in the sweet spot of the lighting available. Which meant that in post I could push the processing in equal amounts in both directions in seeing more of the image in perfect light. It's not HDR, this is from a single image, there was no compositing. It clearly illustrates what the Fuji X-T1 I use can collect given the right settings in the right conditions. I understand it's not everyones taste, but it makes for a more unusual image and that's something I'm trying to use to stand out from the crowd."


p1459780993-5.jpg

 

 

A comparison, before and after post processing with RAW files

Screen%20Shot%202016-01-28%20at%2017.31.

Screen%20Shot%202016-01-28%20at%2017.18.

On the left is the original RAW, on the right is the pushed version.
These two images above are both taken on a Fuji X-T1 with a 
16-55mm f2.8 R LM WR Fujinon Lens.

Because of the Autumn light there are awkward harsh shadows and there are some spots where the highlights have been blown, processing retrieves the obscured information. 

The Canon 6D I use it takes JPG images that, for me, are a bit lacking*, it's RAW images, however, straight off the bat they require little or no intervention for the common palate. My Fuji on the other hand, can be brutal when it takes photo in RAW yet deliver very acceptable photos in JPG mode. But taking them in RAW on the Fuji means I can push the image wherever I choose.

I see my work mainly as image creation, I'm not really out taking 'photos' to please others, I don't have to deliver, I can play, so I do.

*more down to my lack of understanding getting the best out of the in camera settings, there's plenty to choose from.

 

To illustrate my point above about the difference between my Canon and my Fuji, here are a couple of Canon 6D shots from the same day using a Canon 70-200mm 2.8l is ii usm lens with a Canon x2 extender, suffering the same issue regarding harsh shadows.

Screen%20Shot%202016-01-28%20at%2018.16.

This one above, again, shows before on the left and after processing on the right. As you can see, the untouched RAW on the left looks fine, the one on the right has had the harsh shadows softened and the highlights restored through post processing, you can see it clearly in the Norton's number plate.
 

Screen%20Shot%202016-01-28%20at%2018.16.

The Canon 6D doesn't always get it right, a harsh shadow with a bright sky background is always a recipe for trouble. There are setting on the Canon that will correct this, but I don't have time (or the brainpower!) to hunt through settings just to counter it, especially when Bruce Anstey will move his head any second to grin at someone else, so it's shoot in RAW and fix later.

The benefit of shooting in RAW is that you see exactly what your camera sensor sees, and because of the glut of information it retrieves you can push the processing about a stop in either direction, making an otherwise useless photo useable if not perfect with a little tweaking. 

If however, you shoot using JPG then you're completely at the mercy of your own skills, good, bad or indifferent. Your camera will record exactly what you tell it to via your settings and because your image is saved as a compressed file, it will throw away all the good stuff. 

If you're shit hot with a camera and can flick through the many settings hidden in the depths of your camera at the flick of a wrist, then fill your boots, shoot JPG, but when your subject is moving in, out and through difficult light conditions, shoot RAW and fix later.

 

Conventional photography images contain harsh shadow and whited out highlights. With your eyes, you look at a scene in full and it appears balanced, but take a sunset for example, if you concentrate your focus on the setting sun (don't look at a full sun kids, you know the rules) you might be aware that the dark part of the scene will plunge into pitch black. Likewise, if you look at the dark part of the scene the sun and sky will appear much brighter.

That is your own aperture of your eyes, your irides (plural of iris) open and close to allow the correct amount of light in. So say you look at that very same scene and look for the most neutral part, some of that scene will still be too bright for you to look at and some of it will be plunged into darkness, even though you know full well that you can look at the brightest and the darkest parts and see them perfectly well, that's the dynamic range of your eyes. Cameras suffer exactly the same limit (though some are better ie have wider range, than others). Most photographs / images are shown with this same limited dynamic range, leaving some parts burnt out and white and some parts cloaked in darkness. Had the photographer pointed their camera with the correct settings at either the light or the dark then they'd have got those areas spot on as well.

By using RAW and tweaking a few sliders in Lightroom* you can extend the dynamic range and reveal the detail in the shadows and the overexposed (bright) areas. It's both a relatively natural and again unnatural way of viewing the world, your eyes telling you that a scene to the naked eye is all balanced, although you know full well that it isn't (using the sunset example above). This extension of the dynamic range hasn't been available until relatively recently in the grand scheme of the lifespan of photography (nearly 200 years), so it's no surprise that when someone sees an image that has had it's dynamic range widened past what would be deemed the normal range the image is deemed odd or weird in some way.

 

I choose to push the dynamic range of my images for the following reasons - I can, I like the effect, they stand out in a sea of bike images that are near identical in (none) treatment. 

 

*very few photographers would knowingly use Photoshop to correct images if they knew how powerful and easy Lightroom is in comparison


Comments

No comments posted.
Loading...

Archive
January (1) February March April May June July (1) August September October November December
January February March April May June July August September October November December