Our Great Photo Processing Shootout!

No1 easy, through the burning to highlight the trees feels OTT.

No2 has been hit by a processing Tsunami.


Any chance of putting the RAW up for download (don't as k me how) so we can have a go?

I voted 1 because it looks warmer, not because I think the processing is any better/different.

It is clear to me that the same care has not been exercised to bring out the best of this image using the different editing systems.

The blue hue over the trees makes the whole image look totally unnatural, but I think the sky colour in 2 is much more natural looking than it does in 1.

Many thanks to all for the interest so far. Not sure how far we'll go before revealing the origins of both images! The choices of colour casts and toning has obviously made a big difference in how people react, and is certainly the prominent distinguishing feature.

However we were keen to keep the focus on those aspects of the final output which were common to both in quality assessment terms : I can think of a number such as noise levels in different parts of the scene, for example.

I am sure I speak for David as well in saying we appreciate your input and thoughts.

For me the most striking difference beyond the colour casts it the detail in the cloud, Image 1 has much sharper distinctive shapes in the cloud, image 2 looks much more wispy. I'd be interested to know how they actually looked when shot. Difficult to tell at these resolutions but the cloud in image 2 also appears slightly noisier despite being softer.

Share the original and let us all have a go ... lol

Cannot see that at the small image scale
Matt

I don't see this as a test of the relative capabilities of different RAW processors. To achieve that the same look should have been attempted by one person for both images.

Thats been suggested - the problem we find is that we all have different commitment to and expertise on different software. Even with equal skills it would be hard not to let your love of one particular package shine over the other.

Mmmn... I'm not viewing it on a full-sized screen, but Image 1 has kept a better distinction between different colours -- see the front of the boat eg. And it is the more pleasing result -- though whether it's 'accurate' is anyone's guess. It's possibly safe to say that some of the colour rendition in image 2 is off-key. But I do think the premise of the exercise is a bit cock-eyed. I'm not sure what we're meant to deduce -- or what it's safe to deduce -- about the different software being used. As someone has already said, it's likely to be as much about human decisions as anything else.

No 1 for me too - but only because the mist over the water is too blue. If that element could have been controlled, and kept more greyish, the rest would probably look better- if still a little surreal.

The comments added have made very interesting reading, especially for me as I was present when the picture was taken! The colour balance is actually far more accurate to the real-life scene in Image 2 than it is in Image 1! This contradicts the majority of comments placed about Image 2 looking unnatural, and is possibly because not many people are about at this time in the morning when the light is very blue!

Image 2 was processed in ACR and Photoshop, and has not had the blues emphasised at all (there has been a tweak to vibrance and a very small saturation increase across the board, but absolutely no adjustments to the relative blue levels). The blue tinge on the trees is consistent with the original RAW File, and is a consequence of the light at that time in the morning.

I fully agree with the comments that Image 1 has far greater detail in the mist and is overall a more pleasing / detailed image which does look more natural if you don't have the benefit of knowing what the scene actually looked like in real life. For example, on the bottom right where the reeds are, Image 1 is more brightly lit here (as if the sun is on the reeds) and also there is detail of different colours in the water near the reeds. This makes the image more pleasing, however in real life, this was not visible; the reeds were in near darkness and there was very little definition in the water surface, and this is more accurately rendered in Image 2.

I think this test does demonstrate there are massive differences between software for processing, but also that human perception and decisions play a huge part. A better test would be for the same person to process the image on 2 different program flows (Nigel, could you do this?) and then compare the results.

If I had been asked to judge between the two, without having the benefit of taking the image, I would also have chosen Image 1. My sincere thanks to Nigel, who did a superb job of processing the file and to everyone who has commented on the results.

Regards
David

I'm not so sure about that. Surely, it would be possible to make both images look the same whatever the processing software was, provided of course that we were skilled enough in that particular software?

Well if that was the goal then the test has 100% failed. Unless you do exactly the same thing with each set of software then there's no way you can compare technical quality.

As for an artistic comparison, like everyone else, I much prefer image #1. Streets ahead of image #2

In what way? You could get either result in pretty much any image processing software.

I wouldn't claim there are no differences in output from software. Indeed there are differences between Lightroom's PV2010 and PV2012 process versions, but they are slight.

You certainly can't tell by offering up two images that have been processed very differently

I think is is just common sense isn't it? Every piece of software is written differently and so all can produce different outcomes from the same inputs and I don't agree that you can get the same results regardless of the image processing software used. If this was the case we would all be using freeware programs and not purchasing Photoshop & Lightroom etc. I work in software and know only too well how differently results can be depending on the code written, even if two sets of code purport to do the same thing.

Regards
David