Dull images with orange tint

Somewhere along those threadslinks i posted above, a forummember explained why LR and DxO is completely different in there aproach of the sliders.
Something about the use of luminance layers/mask’s.
He demonstrated it with that tiff gray ladder image.
So without judgement or choosing sides of dxo vs adobe (i stepped away from LR4.0 i think.) it’s clear that the selective sliders in LR8? Are much evolved and probably “more filled with AI”.

DxOPLv1 frustrated me also at first, i used a quirky Silkypix 7pro and there excelent knowledge of panasonic color rendering. They got even a camera style readout and preset.
There dodge n burn approach was what i was used to. So that Smartlighting was way to agressive all above 50% was rubbisch in my eye’s. (i was used to bump d n b to max and then fidle the rest around.)

So i started to digg in there system, the how it works what it does.
Ask around here on the forum why things happend ascit happend doing things as i tried.
Played around in the smartlighting toolset and did different approaches on 1 image.
Discovered that SL as best works if you use it as a form as DR fencing.
Even a severe underexposed image and white bulb lamp spot can be fixed whit this approache.
I use exposure compensation as shadow lift wile my box of SL on a bright place suppressing this white place to lift also. Just slidge presentage 25%max and the heavy lifting with exposure compensation.
The selective tone and selective contrast same thing, coming from SP i was used to contrastbalance and presets as low mid high strong, now i needed to decide myself what contrast level was a good point of customising start. Contrast and selective tone need to be worked on together. (filmpack elite needed)

Then i dived into the world of controlpoints, something completely different in way of working. Luckaly i had non zilts experience with luminance layers or layers as in photoshop to begin with so it was something new.

Well to cut the story short. DxOpl is come a long way to v3.2 much is improved as in inverting layers, copy layers, repair and cloning , opacity , feathering and such adjustments in the tools.

Ok for a new user it’s overwhelming or a feeling with your hands in a muddy pool to find the right screwdriver. BUT with some elbogrease it’s a very powerfull handtool. The control you have to attack a image from all kinds of ways in different manners is in my eye’s what the tool makes.

Agreed, it’s not finished yet but dxo staff and developers are working hard to give us every time a new step up.

I break a lance, ok, and maybe i am precidisch but it’s not as bad as you think.

Regards,

Peter

I believe we’re talking about the same thing. Yes, it is the color rendering / camera profile. Here’s a demonstration. First, from an out-of-camera JPEG - an underexposed group of faces in an image with very high dynamic range (a bright sky in the background, dark foreground):

Now the RAW file processed with Smart Lighting = Uniform, Medium and Color rendering set to Generic / Generic defaults:

You can see that the colors are oversaturated, particularly reds and yellows - typical for my Olympus RAWs. And the underexposed faces have that orange look and the image overall looks washed out and flat. So let’s see what happens when Protect saturated colors is raised to 90:

Much better! It can be better still with Smart Lighting changed to Spot-Weighted, with three faces selected plus a region of the sky that includes part of the landscape (not visible in this screen grab):

Better still! And now if we change the Color rendering to Generic / DxO Portrait:

My next decisions are to raise highlights, midtones, and blacks. If I raise shadows, the contrast drops too much - I don’t want that.

For years and years and years, this “best RAW Converter” discussion has been raging. Why not pick your best photos in the camera manufacturers software for your camera, tweak your best photos little in that software, and then export tiffs or even adjusted RAW files to the EDITING or CATALOGING software of your choice?
Some may find this shout me down list helpful:
Reason 1: Reason 2: Reason 3:
Reason Bazillion:

There is an appropriate article on Thomas Fitzgeralde’s site today which tackles exactly the subject of this thread except C1 vs LR.
First paragraph:
Whenever I read people struggling to get to grips with Capture One, one of the most common sources of frustration is generally from people trying to use Capture One like it is Lightroom, and expecting the same results. I can not emphasise this enough, but Capture One is not Lightroom. It’s a different application, and it behaves differently in a great many ways.

Understanding this is the first step to acceptance.

See article at:

Ian

I agree. Still, it would be nice if the range of the global shadows slider in PhotoLab had less affect on mid-tones. The same is true of the highlights slider. I can work around it, but the range of those two sliders overlap other tonality too much, IMHO.

Mark

It’s a bit of reading, but I think this thread (especially the linked post) is a good place to learn more and understand how to preserve midtones while adjusting highlights and shadows/blacks in PhotoLab. Note that there are other consequences to tonal adjustments in both Lr and PL, depending on the method you choose.

For more precise control of tonal ranges in PL, maybe the tone curve tool is the way to go? It may be more clumsy than a handful of sliders, but has no side effects or ambiguities.

Peter;

Thanks for the answer.

I’ve seen this gray ladder and long explanation. I have an idea how those sliders interact each other but I doubt this is the best way to implement Selective tone tool. Personally I wish those sliders would be more separated.

Thanks for the tip regarding Smart lighting. I was just playing – it works for some photos but it doesn’t work on other photos and I get this lifeless look. I have a mixed feelings about this Smart lighting. They should fix this and implement some sort of AI to control contrast on faces. I’m sick of using Microcontrast because it affects whole image with bokeh.

No; DxO is not bad but sometimes it works and sometimes it doesn’t. I can’t say everything is perfect and I wish some things would be done more conventional. I also think some things could be done better but …. Let’s wait and see what Photolab 4 brings.

Like you say; you need to dig in but honestly; how far should I dig.? Till I come to China through Earths core? I’m far from dumb and I managed to learned ACR, Photoshop, Corel, Capture One, Lightroom, Creo (ex Pro Engineer), Excel, Word, Portrait Pro, Topaz, Helicon Focus, Imagenomic…etc, You name it. I built myself DAC converter, A class single ended amp, AB class amp, a bunch of loudspeakers, subwoofer, boomboxes, USB to S/PDIF converter but I haven’t yet figured out this DxO logic why are things as they are.

I will tell you what’s one of the most annoying things…I put a photo in DxO and I click on ‘’Fit on screen’’ and image is not sharp. If I transfer this same image to Photoshop its sharp. So; DxO doesn’t show me sharp image on screen and Photoshop does show the same image sharp.

I think it’s time they fix that. For God’s sake; any other RAW converter shows me sharp image on screen and I don’t have to magnify it to 75% or 100% or whatever.

Hi Greg;

I was just exploring those camera body profiles and Samsung NX1 profile looks exactly the same to me except it has less saturated red and orange color. I need to play some more but looks like it will help in some cases.

DxO is just a challenge. I need to master it. It’s personal. Then I can move on.

i think that was the final conclusion of some more experienced people also. :slight_smile:

I agree that SL is not always play nice and it can be a frustrating slider to begin with.
As far as i remember and understand:
Spotweighted is : centre weighted average unless you dictate it with those boxes.
You basicly tell SL the upper and lower part of the to adjusted exposure difference in precentages.
(the greater the difference between the boxes the more it “flattens” you image. (bit like idyn from Panasonic. that lowers 1EV in three steps 1/3 2/3 3/3 and if that’s not enough lowers contrast in tone curve line by lifting shadows and lowering highlights in ooc-jpegs. (it effects your raw only due EV adjustment so it works as a auto ETTR :sunglasses:)
By placing more boxes you can control the line of leveling of the tonecurve correction: say you use three boxes two on shadows and 1 on highlight the shadows got more attention in the correction part.

Personally i have SL on 25% max which is just enough to extend my m43 images DR abit (pulls in the raw DR inside the sRGB DR i think.) (this colorspace theory is always foggy for me. ).
all other exposure/tonality/contrast i do with other tools. (i use exposure compensation in auto mode just to see what the “engine” thinks i like. which can be wrong by the way , see the step 04 EC image.

I never would dare calling you dumb and most people here are far more skilled in photography ánd postprocessing in which application of your choosing then me i learn much more from them then i can teach. :slight_smile:

Yep same as the “blob” or “plop” kicking in of the optical module. (effect of sharping and clearview and CA and micro contrast is only seen in preview at 75% or higher. Sorry i think it’s a computing speed thing and i find this also irritating not having a button to manual "preproces the image in a “temp export jpeg.” a onscreen how it can look like if i would do a export. (it’s mentioned here as wanted improvement.) (i just export and watch that jpeg back in my DPL folder, lame but effective, to see what i need to change. (suffix temp and reshoot it with folder select in my raw folder so it pops up when ready. )

I promised a go on you file:
i followed only your steps and did nothing to enhance it to my own interpretation, like local adjustments.


my boxing skills :wink:
(you can move or add if needed but i mosly do so if my other adjustments are in conflict with SL.(toggle tool off on to see that.)

take the shadows and highlights inside the "clipping warnings as much as possible without destroying the “look” of the image, (tonality/contrast balance)

(i think LR does this “equal adjustment of Tone and Contrast” in one slider.)
effect of the unbalanced adjustment:

The balanced methode is more then twice as powerfull as the single methode. (that’s why Filmpack is so important to have.)
see the mishit of auto exposure correction:

And the effect of SL inside the corrections:


so the basic cat looked like this:

i would go a bit like this:
colorcasting removal: controlpoint is great for that:


and some cat color enhancement, (it’s your subject)

Tip use a ALT click to make a negative and move this arround to see if the original mask is effecting things you don’t want: video

it was a quick one but this i think my first export jpeg of the cat:


Again i don’t know your goal of look and certainly not question your skills in postprocessing but i am working up my skills in this program for several years now and i am not skilled in processing as a certified adobe photoshop artist or think knowing the trade of photographer just a techy with more persistence/endurance then most people. (if something is bugging me i keep digging until i am satisfied. LOL)

Sorry for the rant/long post.
Edit: got some blue on rightside and some too yellow on the teiles becasue of the casting correction. few extra points under the counter and a negative or three on the teiles:


hmm, redisch glow on the counter right side edge. another controlpointgroup to push that to “green” i think it’s CA which is peeping through.
done:

Yes the tone curve is more precise but that doesn’t address the issue of the highlight and shadow control ranges.

Mark

What do you mean, Mark? It’s just another option, isn’t it? It can give more precise control over highlights and shadows, midtones, and white and black point. The thread I linked to addresses Selective tone’s highlight and shadow control ranges.

Peter;

Well; that’s an excellent tutorial. This trick with double action (highlights + highlight contrast) is a good one. I didn’t know that. Thanks a lot.

Equal adjustment – I will remember this.

But there is lot of variables. So I need to double all my actions? We have Selective tone tool with Highlights, Midtones, Shadows and Blacks slider and then Contrast tool with Highlights, Midtones and Shadows sliders. How to control all this to get a good image?

Why there are no such DxO tutorials on Youtube? You know; a famous photographer or someone skilled (like PiXimperfect for Photoshop) is sitting behind computer and showing step-by-step how he develops his RAW files. I can find tons of such videos for LR, C1P and Photoshop but nothing similar for DxO.

Are there no such DxO users?

I think Smart lighting sometimes also does not have enough separation. Sometimes I put the box on highlights but midtones are also affected. Sometimes when I put the box on shadows this also affects midtones.

It’s somehow complicated. Where to put those boxes? Move them around – which box where? How big should they be? Lots of variables. For me more complicated then sliders.

‘’ (i think LR does this “equal adjustment of Tone and Contrast” in one slider.) ‘’

Yes, I think so. I guess this is part of ‘’behind the curtain’’ AI thing LR is doing. I hope DxO will implement something similar in the future to upgrade Smart lighting.

Highlights slider all the way to the left….

I know; you get this nasty look. I think I have seen somewhere – if you move Highlight slider to the left then you need to compensate and move Midtones slider to the right.

The same logic I tried with Shadows and Blacks slider but Shadows slider also affects midtones so I guess I need to do double compensate? Move Shadows to the right and then move Blacks and Midtones slider to the left to compensate?

That’s why I’m confused. You need to do this compensations because sliders interact. Then why they need to interact? Odd logic if you ask me.

And about sharpness;

It is about final time they give us sharp image on screen. When the image is ‘’Fit on screen’’ it should be sharp. It’s easier to work with sharp image. How can I adjust and not see sharp image?

It would be even nicer if we get a preview, how the image will look after ‘’export to disk’’. I do export to disk and look at jpeg. Sometimes is too sharp and sometimes not enough. So I just export to Photoshop and there I do resize and smart sharpen and export to disk.

Odd sofrware. Some things are excellent like Prime noise and some other things but sharp image on screen is not possible. Is this part of ‘’Dxo is not Lightroom’’ thing?

Every other RAW software can show sharp image when fit on screen.

Oh, now you mentioned it i used the tonecurve lowered the highlight tip and curved it a bit t get contrast, made a blackpoint dip for clarity.
and i tought it where a dryer andwasingmachine but it are glass tables, probably blueisch. :persevere:
Any way, contrast centre balance that’s what the tonecurve can give you and some control on the shadows and highlights and everything in between.
Me i am not an expert in that curvetool, but ive seen some youtube video how people are completely turnover a image with that single tool. If i remember well tou have the possibilty to select RGB and the channels separetly.
So if you change the Rchannel somewhat from the others you could suppres the red and thus the orange but if that control is possible i don’t know.
You have lrgb and l, r, g, b. Control so colorbalance is posibletone%20curve
just check you can :slight_smile:

no only with the “twins” highlight, midtones and shadows.
ive played with blacks because i was thinking it worked like SP7pro’s “blacklevel” which create "clarity"dehaze. it isn’t.
But dipping down botompart in tone curve is. i find it a difficult to master tool but using it to support clearview is workable.

this one: he is a expert in NIKcollections
landscaping images.
and


home site of dxo

bit of playing with the tiff, video
See the effect on the edges of the rows.

Don’t know, i think it’s philosophy you need to zoom in at 75/100% to tune the sharpening and see what your doing, same as HQ denoising.and CA. and it safes processing power and thus time.
The optical Module is a powerfull part of DxOpl, much more then just a distortion correction, lenssharpening is also applied which works together with PRIME.
every tool has some secrets which you can find if you look in to it but its frustrating at first but along the way you find the manual control and multiple ways of tackle a problem quite good.
Watch some webinairs and other stuff to get a glance of the tools in use of experts and repeat on your own speed and images. Your will be fine. :slight_smile: And remember it’s a tool not “God’s hand”. :wink:

Edit, highlight slider plus contrast version adjustment, the contrastslider adds or deminisch the detailing in a bright plane, so fine adjustment after the start of equal can help to give a white blob some texture.

1 Like

Thanks again Peter.

I need to find this image with stripes so I can play with sliders to get a feeling how exactly they work.

I also use Curves tool. I usually don’t do some complicated things. Sometimes I do S curve to enhance contrast. Sometimes I use them to adjust white and black point (like with Levels tool in Photoshop); this gives the image more shiny and punchy look.

You can do a lot of work with Curves but it’s not a simple tool so it needs time to get used to.

I’ve got lots of information from this thread – lots of knowledge to digest. I guess I’ll have to dig in and try. Practice makes the master. I hope I’ll find the way how to develop RAW files with the same speed I can at the moment with Lightroom.

There is one thing I STILL DON’T UNDERSTAND FULLY and this is exporting RESIZED jpegs directly from DxO to hard drive.
Here is my workflow:
-First I develop a Raw file. For example: I crop if needed, then I adjust Smart lighting, then I move some Selective tone sliders, then I adjust Micro contrast …etc. Now the image is just the way I like it.
-Then I click on ‘’1:1’’ to zoom on 100% to see the sharpness. I adjust the sharpness if needed and now the RAW file is developed.
-The last step is to export this RAW file to hard drive as RESIZED jpeg. I click on’’Export to disk’’ and export window opens. I adjust all the settings but …

HERE IS MY PROBLEM:
There is ‘’Interpolation’’ setting. If I set’ Bicubic’ exported jpeg is NOT sharp. If I set ‘Bicubic sharper’ exported image is OVER-SHARPENED.

HERE IS MY QUESTION:
Do all of you have the same problem and what you do?

export


look for: Luminosity 255-0 in steps.tif (

The export settings is backlogged as looked for improvement .

I am not using those different settings and don’t resize to special use so don’t know if it’s bad or not.
I do screen viewing only, FHD or on the future 4k tv screen, so i store as good as possible.
Like old homevideo clips which deminisch in modern viewing screens to a thumbnail or blown out of sharpnes by tv software. So store as big as possible resolution and i work from there if i would resize for any reason. So can’t help you with that.

This has been under discussion here and here. Also here by you, so again you are spreading problem discussions across multiple threads which only slows down the work and risks reducing forum participation by customers. Why not keep each problem in its own forum topic?

Egregius;
Looks like I forgot about that.
I hope somebody is working to fix this.

‘’ The export settings is backlogged… ‘’
What means ‘‘backlogged’’?

In there “to do list”

Peter; I forgot;
thanks for the file.
Regards

1 Like

I was just playing with ‘Luminosity 255-0’ file and looks like Highlights slider affects all midtones and also affects shadows in small amount. It’s hard to see on those two jpegs I’ve attached but I can see in DxO when moving Higlights slider that it also affects shadows. I find this somehow strange.
First image: no adjustments


Second image: Highlights slider -60