Preseli: which version?

[Davejl]

Because it is yet another day of rain and gale-force winds, I'm confined to messing about with a shot from last autumn taken in the Preseli Hills in Wales (the bluestones of Stonehenge came from near here) rather than taking new ones.

I shoot raw, but it does give a sometimes almost overwhelming amount of choice for processing. In all but one, I've just confined the changes to varying the W B preset (camera set or raw converter default) and colour preset. Just wondered which (if any!) others prefer out of these.

A


B


C


D


E

[Manfred M]

Dave - let me turn the tables on you and ask which one you like best and why?

After all you are the photographer and should have a vision as to where you want to go with this image. This really isn't something anyone else can do. Let me further suggest the likely source of your dilemma. I suspect the underlying issue is that you have not developed a standardized workflow. This is why you see so many different options.

I don't know which raw converter / editor you use, but assuming it is either Lifghtroom or ACR, a good place to start is the order presented to you by the software. It doesn't mean you have to touch every slider, but following an order is important. It also doesn't mean you have to stick with it as you gain experience, but it does at least provide a logical and systematic way of processing your images. To me that is an important step and I suspect some of the answers will pop out as you adopt a consistent approach.

[ionian]

To my eyes, I prefer B - the saturated greens give an alive feel to the image, and cut through some of the background haze. However there are loads of directions you could go with this in post, so Manfred's advice is (as always) very sage.

Note that I still have much to learn with processing, but I mainly use Lightroom and follow the slider order workflow with a couple of exceptions. Ironically this image may be one of those exceptions - I would try some split toning after basic levels, then go back and adjust as necessary.

[Davejl]

Hi Manfred,

Fair comment, but I really was looking to see what others feel. I do know which one I prefer and I do think that I pretty much know my way around the raw converter that I use. If my post was misleading in suggesting that I didn't know how to operate the raw converter, I apologize. I was really just commenting on the possibilities which are there in the software and trying to get a feel for what others think, e.g. is more saturated colour generally more popular. I haven't said which I prefer because I don't want to influence anyone.

Cheers,

Dave

[Davejl]

Hi Simon,

Thanks for your reply and comment on version B, which appeared before I finished posting my comment in post #4.

Dave

[Shadowman]

I don't think the impact is changed at all between each of the versions. Nice capture.

[Wavelength]

I see there is a gradual change of dark brown to light chocolate brown color, which i notice most... It is difficult to select one all are nice

[bnnrcn]

Hi Dave, I like A as it looks natural and warm

[Geoff F]

For me, Dave, A seems a bit too red in the clouds, while B looks a fraction on the blue side. C or D would be acceptable. E is just possibly getting a little too far on the brown side.

[Manfred M]

Dave - In that case, I will have to suggest none of them as I think the shadows need to be opened up more to give you an effective image. I think the contrast is a touch low as well and the colours needing a bit more punch. The sky needs to be brought out some more, etc., etc., etc.

That being said, unless both you and the viewer are looking at these images using a calibrated and profiled computer screen, you could be looking at two completely different images. The laptop I am using had a noticeable blue colour cast before I profiled the machine.

My style seems to be different than your style. My starting point would likely be something along these lines:




I tend to bright, punchy colours and these are probably not to everyone's taste, but if you look at the landscapes I've posted over the past couple of years, you'd likely see the image and recognize my style. I tend to bring out details in the image that others leave behind.

Picking on one of the other Moderators, Donald's landscape style tends to be moody B&W images.

I remember being at the d'Orsay Museum in Paris a few years ago. After the visit I was asked about what I thought and my answer was along the lines of "a beautiful building whose architecture is blighted by some terrible art". As a general statement, the Impressionists don't do much for me, with the exception of Seurat.


I guess I would feel better in answering your question had you allowed an option of "none of the above".

[Davejl]

Manfred, thanks. I did actually say "which (if any) others prefer" thus allowing you to quite reasonably say "none". The raw is processed in the Silkypix software provided with the Fuji X100. I take your point about the difficulty of knowing whether the colour on my screen looks like anyone else's and also I agree about the shadows, though on my screen the rock outcrop in E is less shadowed.

The really odd thing is that your version is considerably less saturated to me than mine! The heather looks almost de-saturated to me. A is using the Silkypix WB, the others use the camera (auto). Silkypix WB gives images a magenta bias to me.

I have noticed that quite saturated images seem to be preferred by people. I feel that I used to saturate my images too much and have tried to get away from it to something that feels more "natural" to me.

For what it is worth, I think I like E best.

Cheers,

Dave

Now I must go and see if the roof is still on. The wind outside is getting horrendous.

[Davejl]

Thanks for your replies. I think Manfred has hit the nail on the head with his comment that we are probably not seeing the same colours, given screen differences. I'd also be very interested to know whether anyone else sees Manfred's version as less saturated than mine.

Dave

[Davejl]

Manfred, that is crazy! I've just downloaded the bigger version of your edit and it now looks much punchier than what I was seeing in the smaller one. Apologies!

Dave

[Manfred M]

Thanks for your replies. I think Manfred has hit the nail on the head with his comment that we are probably not seeing the same colours, given screen differences. I'd also be very interested to know whether anyone else sees Manfred's version as less saturated than mine.

Dave

Dave - when you mentioned that, I went back and looked at the image I posted. I usually work in the wide gamut ProPhoto colour space, which means I need to convert it to sRGB when posting. Unless you are viewing in the FireFox web browser, which has the best colour management of any browser, the rest will produce a rather muddy looking image. So nicely said, I blew it and posted the wrong version.

I've fixed the image and you should see a difference now.

When you suggest that the images you see in SilkyPix have a magenta cast, I strongly suspect that your screen is not set up correctly. Unfortunately, this is a common issue and x-Rite (i1 and ColorMunki) and DataColor (Spyder) make a lot of money selling devices and software that do this. That is why I mentioned I am viewing the images on a calibrated and profiled screen, so the colours, contrast, brightness I see are accurate. In fact when someone mentions that they see a colour cast, the first question I ask is if their screen is calibrated and profiled. If the answer is "no", I can virtually guarantee that this is the problem.

[Davejl]

Manfred, thanks for that. I began to think I'd gone mad!

The Silkypix WB preset, which I don't much like, can be seen when looking at the green-magenta slider within it, to be slightly biased to the magenta side. Image A has a bit of pinkness in the cloud which doesn't appear in any of the other renderings and Geoff F seems to see this too. The WB as set in camera (where I tend to leave it on auto) seems to give good results. The home prints I do seem pretty acceptable to me with no obvious colour cast. The major issue I have with some home prints (from Fuji and Canon files) is that I find it very difficult to judge how light or dark the print will be. I presume that proper screen calibration would overcome this?

Dave

[Geoff F]

Yes, Dave, using a properly calibrated monitor should give a much better match with prints although you have to use the correct print profile for your paper. Also make sure you use the most suitable of the printer auto 'improvements' for your image; or turn off all the printer auto changes then make manual adjustments. Alternatively use the Adobe printer matching if you have suitable software.

Most of the better quality paper is fairly stable now; but you can still get a bit of variation between different batches.

Using third party ink can also give unpredictable results.

[Davejl]

Thanks Geoff

[marlunn]

Dave, I think on my calibrated monitor I am with you on E being my favourite, it looks more natural to me for the Uk countryside.

[Manfred M]

Manfred, thanks for that. I began to think I'd gone mad!

The Silkypix WB preset, which I don't much like, can be seen when looking at the green-magenta slider within it, to be slightly biased to the magenta side. Image A has a bit of pinkness in the cloud which doesn't appear in any of the other renderings and Geoff F seems to see this too. The WB as set in camera (where I tend to leave it on auto) seems to give good results. The home prints I do seem pretty acceptable to me with no obvious colour cast. The major issue I have with some home prints (from Fuji and Canon files) is that I find it very difficult to judge how light or dark the print will be. I presume that proper screen calibration would overcome this?

Dave

Dave - when I wrote about a colour managed workflow, the endpoint of this process is to create a print that closely matches what you see on the screen. As others have mentioned photo papers have specific icc (International Color Consortium) profiles that allows the software you use to print to produce a print that closely matches what you see on the screen. This is why calibrating and profiling your screen are so important.

Your computer screen uses an additive colour, transmitted light, RGB process whereas a bring uses a subtractive colour, reflected light, CMYK process, so the two can be made to look close, but never identical. By convention, we use daylight (6500K) light to evaluate prints and that assumption is built into the end-to-end process. By convention we set our computer screens to an output level of 120 cd/m2, so in a perfect world you prints should come out fine if you computer screen is set up that way. Unfortunately, the world is not perfect and I find a that setting my printer (as well as commercial printer that I have used) end up printing colours that are just a bit too dark. I determined this by making test prints and adjusting my images in post processing to make them appear a touch lighter. That way the final result is exactly where I want it. So long as I don't change the controls on my printer, this value is a constant and I apply it to every print I make on my printer.

With the magenta cast you are reporting and the dark prints you are seeing, you should consider investing in a calibration device and setting up you computer screen properly.

[Davejl]

Thanks Manfred. I think that possibly the main issue I have is that I only have a laptop computer. Depending on the angle of the screen, the darkness or lightness of the screen varies tremendously. I know that when I used a desktop computer at work to view any of my photos, this wasn't an issue.

The other thing is that I only have a cheapish A4 printer with 3 colour inks plus black. If I were to go down the road of getting into printing seriously, I'd certainly invest in a desktop computer, decent printer, and proper calibration. Unfortunately, financial constraints preclude this at present.

Just to clarify, as regards the magenta issue, I don't suffer a magenta (or any other) cast in the prints I make. The magenta cast was something that the default Silkypix white balance seems to add to the X100 photos, so (apart from shot A here), I never use it. I do in fact get a pretty reasonable correspondance between the colour on screen and that which I see in my print, excepting the lightness/darkness issue.

Cheers,

Dave