This content has been marked as final. Show 41 replies
Well, have you actually tried the Adobe Standard for your camera? While it might be useful to build you own, only if it's better than the new standard one.
Jeff is being polite. There's just no way that a custom profile built on someone else's camera is going to be better than Adobe's, unless by sheer coincidence the other camera's response is identical to yours.
Are you saying you're not happy with any of the new beta camera-matching profiles?
Jeff, I have tried Adobe Standard yes. Of all my options it's probably the best, but it's still not trying to replicate the exact colours that constituted the original scene.
Ramon, I have also tried the beta camera matching profiles but all they do is replicate the Canon Picture Styles... none of which produce accurate colours either.
What I'm trying to achieve is colours that are representative of the colours I'm truly photographing. While I realise that someone elses 40D will not necessarily be identical to mine, I can't believe that every 40D produces completely different colours.
To followup on this, in principal you'd assume that the Canon Faithful and Camera Faithful Beta options in DPP and CR respectively would do the job. However in practice they produce washed out images with no contrast that don't represent the colours of the scene either.
Adobe Standard and ACR4.4 produce the best colours in terms of "general photo quality" but I also know that they're still not producing the type of colours that I saw.
And for reference, while I know that my LCD is not very good it IS colour calibrated so it's as good as it can be.
Just how do you think Adobe got their profile if not from at least one copy of the camera model?
How would a profile made by Joe Blow on his camera body be any better?
Come on, if your camera is that far off, shell out the $70 for the Xrite color checker and be done with it.
Try a different monitor.
I'm sure they tested them against a camera, however I've read on here that the Adobe Standard is not intended to be colourmetrically accurate and is just yet another "flavour" to RAW processing. Automatically generating a profile with a photo of the gretag IS meant to be colourmetrically accurate (again, so I've heard... maybe I'm working on a false premise).
I've already shelled out for a Colour Checker, unfortunately it's not one supported by the profile editor (but was supported by the calibration scripts). The Xrite colour checker is also not supported (unless it's a direct knock off of the gretag one). If it comes to it, that's what I'll have to do but I still don't see why there's a major issue with someone who has a 40D and a Gretag chart to send me their profile and it saves me the ~$130 for a Gretag checker I'll use once.
As for general colour satisfaction, it's not bad at all... but that doesn't mean I don't want it to be better.
An improved color profile is not likely to improve your contrast significantly. Images that are "washed out" and "with no contrast" are more likely a problem with exposure and contrast than with color profile. Check out the exposure, black, and contrast controls on ACR.
Remember that the dynamic range (difference in luminosity or lightness between the darkest part and lightest part) of a typical image is considerably greater than the dynamic range of a print. Professional raw converters generally try (by default) to maintain information across as wide a dynamic range as possible. In processing, you need to determine what part of the original range should map to what part of the range in your image.
Often, automatically produced JPGs bypass this. They have higher contrast and boosted saturation. This reduces the need for further processing and effort; however, it also reduces the control, and the photographer's ability to maintain shadow and highlight detail.
Do you have a method or software that produces JPGs that achieve the color that you're looking for? What is is?
Can you describe specifically what color problems you are seeing?
(You also have to keep in mind that saturated colors in the image may lie outside your monitor's gamut and hence may be clipped.)
I understand what you mean about the contrast, however I am comfortable that my images are exposed correctly on the whole. I guess it's just very hard to quantify, and very hard to prove, where my images are not "right" colour wise.
I also realise that white balance plays a major role in this type of subjective comparison... The issue I'm finding is that when I am flicking between profiles for an image there are times I just can't find anything that's right.
Maybe the biggest issue, at least in part, is perception. When analyzing an example image I've seen the following behaviour for each profile:
- Camera Standard is probably best, but while on this example it's okay, in some images it's over-saturated.
- Adobe Standard is close, probably slightly too green.
- ACR 4.4 is up there too, best in some cases but can have issues with skin tones.
- Camera Faithful is next on the list but tends to look flat and lifeless (even though no change other than profile).
- Camera Portrait makes faces too pink and skies too blue. In general it seems to over-saturate.
- Camera Landscape makes faces look pretty good, but skies and landscapes are WAY over saturated.
- Camera Neutral looks relatively good colour wise but has no contrast and looks under saturated.
The problem is, in this one example they've behaved in a certain way. In another example (even one where I've manually white balanced) they'll behave differently... suddenly Camera Standard will look terrible and ACR4.4 will look the best... or Adobe Standard.
I guess the issue is I'm just not confident that LR is reproducing the colours that the camera saw... but only because I don't think the profile is right. Once I know that the profile is right (and whether I'm being silly or not, I would be satisfied that Joe Blog's 40D is going to be almost identical to my 40D) then I'll know that the faces look a little red because I've got my white balance wrong. Right now there are three that are "close" to what I saw, but I'm not confident that any of them are spot on.
I hope this makes sense. I'm not attempting to be difficult, to be honest I figured that the creation of the Profile Editor would lead to a proliferation of downloadable camera profiles that were supposedly "colourmetrically accurate" for each camera... maybe I'm just a little quick on the bandwagon.
PS. I realise that colour gamut of the monitor plays into this also, but I am assuming that if I can SEE that the colours are too saturated then I'm not exceeding the gamut... I guess it's possible that my "washed out" profile Camera Faithful is exceeding the gamut and subsequently doesn't look very good?
As an experiment only, it is possible to convert one of your "skin tone portrait" raw images to DNG, and then to open it with the DNG Profile Editor (DPE). After you've read the tutorials regarding editor's use, you can play with the editor a bit to try and make colors better ... and then to describe here what you've done and Eric Chan may be able to say more.
Altho I do have to admit that something may be wrong with your color management if the skin tones look worse for profile 'portrait' than they do for profile 'landscape'(?) IE, it could be your monitor profile that is causing skin tones to "appear" off.
My camera isn't Canon, but I was very impressed with the color accuracy of the standard beta profiles.
In general if you're seeing a color cast (e.g., too green) the issue is white balance, not the profile. A simple tweak of the WB should fix it.
I'm not exactly sure what you're looking for, though. In general you don't want to get the exact colors that the camera saw because then you're talking about scene-referred color and that's downright ugly. You generally want something that's been tone mapped in a way that makes the image look pleasing, and the way you do this could vary image by image.
A single profile, no matter how "accurate" it is, whatever that means, is highly unlikely to produce desired results across all images because the colors you're looking for tend to be image-specific and content-specific.
Ramon; My apologies, I wasn't aware they'd been purchased. That's good to know, the Xrite cards seem cheaper than the Gretags were.
Eric; I don't believe my white balance is incorrect necessarily... different profiles GIVE the image a very slight colour cast because they interpret colours in a certain way. For example the Landscape profile intentionally saturates skies because everyone wants their landscapes to have bright blue skies.
What I am looking for is exactly as you say, "exact colours that the camera saw" which I wasn't aware was called "scene-referred colour".
I'm interested in my base starting point being 100% exactly what the camera saw, atleast in so far as a 40D can reproduce those colours. Once I've got that starting point where the colours are as accurate as possible (assuming the white balance is right) then if I find the result undesirable then I can tweak the saturation or tweak the HSL sliders etc but I don't want to be tweaking the profile.
Correct me if I'm wrong, but isn't that the whole point of the auto-calibration option in the Profile Editor?
Yes, PE should give you a relatively accurate and consistent starting point, within technical limits depending on your lighting conditions.
BTW, James, if you're really interested in using the auto-calibration feature of PE, you can get a miniature version of the ColorChecker. (I think X-Rite has a small version for less $$$. It'll work just as well as the larger, more expensive versions.)
Yes, I'm looking into these at the moment (thanks to Ramon for clarifying that they were compatible with PE).
I bought the Perfect Pixs Natural Colour Card (30 patches) because it had more skin toned colours for calibration, but it's no use to me now.
I've been fiddling for the first time with the 24 patch Color Checker trying to get my Pentax K100D PEF's to reproduce exactly what I saw (not what the camera saw) when I trip the shutter. I've been studying and researching this through trial and error for some time now and I've come to the conclusion that one profile can't fit all scenes. I have to tweak each one for subtle variances.
I've compared two converters ACR 3.7 (also 4.0-4.5) using Tindeman and For's ACR Calibration scripts and Raw Developer (Mac only) using its matrix based ICC profile generator in combination with its Lab curve feature and I run into color errors due to different spectral reflectance properties of real scene objects and lighting situations.
I'm very picky about getting exactly what I see that the differences wouldn't be noticed by most folks. For instance shooting up close light gray brick wall with yellowish beige mortar lit by 10AM 45 degree direct sunlight renders all one warmish hue in both converters where the brick should be cooler and the mortar should be warmer. Color temp adjust just makes everything cooler or warmer. This is where the DNG Profile Editor would come in to fix this.
But also I found changing the angle of the camera to the subject and underexposing by one stop fixed this instead of tweaking software. I mean there's no way of getting any consistency unless you're shooting under controlled studio lights and I'm not even sure about this since I haven't tried it.
The thing I've discovered with my Pentax about getting exact numbers with the Color Checker is the exposure needs to be set so that the white square is already close to being between 230-245 RGB when first opened using the default raw converter contrast curve. I've noticed that after both scripts are run that the scene the CCchart was shot in has now lost quite a bit of contrast. The numbers are dead on to the published data and so are the look of the colors but the rest of image doesn't look like my eye saw the scene in regards to contrast. I believe this is due to the linear properties of the instrument used to measure the chart to arrive at the published numbers. When I correct the contrast, all the numbers are off looking noticeably saturated but not by much.
What these scripts and the PE wizard do is make sure your hue's are as they should be separate from density adjusts which affects saturation. I've also noticed that lightening saturated colors by applying a slight curve tweak will noticeably reduce saturation without noticeable change to density and contrast. PE can be used to fix this as well.
I think I understand what you mean about the saturation not directly affecting the hues and certainly I don't like what some of the profiles do to saturation for this reason.
When you say that one profile can't meet all needs, I assume for the most part this comes down to limitations in our cameras and their inability to replicate the colours we see? Do you think though, that it's possible to get a profile that's "most accurate" and then just tweak HSL sliders on a per photo(s) basis rather than end up with numerous different profiles? Or maybe a tweaking of the contrast curve?
My end goal is to apply a single profile on import to all my images... after that I don't mind changing specific things, but I want to import all my images with the best possible base settings.
PS, looks like I'll be buying another colour checker and doing much the same tests as you have.
Yes, James, I do believe you can get a very accurate profile that can meet most normal lighting and exposure driven captures, but there's always going to be that one sunset with those beautiful magenta to coral to yellow color combo's in the clouds and sky that may all render as one yellowish orange. This is where DNG Profile Editor becomes most useful.
An article linked over at the BableColor site shows a visual spectrum analysis display captured by a very accurate and expensive instrument showing how a digital camera distinguishes all the colors of light. It seems it does very well with just red, green and blue and fills in the rest of the rainbow of colors that make up light which can be seen by our own eyes by defraction through a prism. The digital camera only sees red, green and blue where our eyes see a rainbow of color. I've also read that scanners have even a narrower spectrum bandwidth. No wonder folks have so much trouble getting decent color off those devices.
Take a shot of the shiny mirror side of CD angled to the light so it shows that rainbow and see how well it renders it compared to how your eyes saw it. Some camera's seem to do better than others in capturing more varied color transitions.
From my short time studying raw conversion workflows, the pickier you are with color, the more time you'll have to spend on each image. This is the burden given to us by digital technology. It's like at Walmart where you check out your own groceries at the cash register. You're now doing the someone elses work and not getting paid or receiving a discount for the labor Walmart and other store chains no longer have to pay for. But at least when the eggs are broken, you've got no else to blame but yourself.
I love technology. Don't you?
Interesting discussion, and not at all surprising to me. I have never been happy with the out-of-the box color from the Adobe profiles either.
Having spent my life in imaging and having now spent literally hundreds of hours myself with the profile editor, I CAN say that I have finally achieved a measure of satisfaction.
I started out, as many of you have, thinking the color could/should be more "accurate".
After quite a while I finally came to realize that "accurate color" is a fantasy, if for no other reason than the display/print media we use are INCREDIBLY limited as compared to what we take into our eyes. But there are other reasons as well.
Sure, green things need to look green, and red things red, but if we're expecting "nuance perfection" we're going to be continually disappointed.
There is a process, involving not only the eyes but also the brain, in interpreting a scene represented in a photograph into a mental model of reality, so that we not only perceive but understand the subject material. This process is not by any means simple, and involves - much as the engineers would hate to admit it - the artistic centers of the brain as much as (or more than) anything else. This is why we describe the "feel" of a photograph, why some photographs have more "impact" than others, and why specialized films such as Velvia found popularity.
It is also why even our modern cameras don't make what some folks would call "accurate" color.
Only when I changed my mindset, and put aside "accurate" instead to pursue "pleasing color" and "generally good input to the editing process" - when I finally "saw the light" - did I make progress. Only then did I find I could begin to overcome some of the shortcomings of the out-of-camera color and all the various profiles.
Notably I did *NOT* succeed by paying a lot of attention to numbers and color charts.
At this point I've developed a Canon EOS-40D profile that accomplishes these goals:
- Pleasing color, not unlike the Canon out-of-camera color but with some significant areas improved. Reds are not overblown as Canon's, blues are lighter than Adobe's, oranges are not oversaturated.
- The conversions make good input for further editing in Photoshop or Lightroom.
- The conversions have a good bit more dynamic range than the Adobe profiles without looking washed out. Skies and clouds are not blown out.
James (the OP in this thread) has contacted me and I've sent him a copy of my 40D profile to try it out. If he finds it to his liking, I may choose to go ahead and offer the profile online.
Noel, is your "profile" a set of Camera Raw settings, a .dcp profile created by the DNG Profile Editor, or something else?
It's a .dcp profile, usable by Camera Raw and Lightroom. It does require an altered set of default settings to be saved by the user as well, as it produces lighter conversions than the default profiles, as well as expecting a blacks level of 0 and a centered Brightness slider.
Once set up, the results are consistently very pleasing.
Here are some results you might find interesting:
Adobe Camera Raw Default (profile ACR 4.4):
My 40D Color Match Profile:
The in-camera Canon JPEG:
Are those screenshots really in sRGB? I actually prefer the ACR default for this shot, but I guess that is very personal. The shady areas get a little too much of a blue tint I think in your profile. That is probably due to the zero black setting which lowers the contrast.
If those are screenshots, I hope you assigned your monitor profile before converting to sRGB and posting on the web.
All three of those shots show very subtle differences. Have to agree with Jao, ACR 4.4 is the most accurate to how I've seen outdoor color in similar scenes. Shots I've taken a couple of days ago in the park with the sun 30 degrees off horizon (the golden hour) show quite a bit of yellow similar to the ACR 4.4 version. The dried palm leaves in the background show too much magenta and noticeable blue in the greens on my display in the other two shots.
This may be due to how you took the screenshot and if you assigned your monitor profile.
Those are not screenshots but a composite of the original image and the Photoshop Histogram display. They are being properly presented here and look as they do on my monitor when viewed in the sRGB color space.
My point in posting them was not to evoke critique on the content of the images themselves, but to show these two things:
1. The image developed with my new profile looks more like the Canon color than ACR 4.4 color, which I feel is consistently more "muddy". It is important to note that they were developed with the "As Shot" color temperature, which I neglected to mention before.
2. The histograms show I've managed, with the default settings, to retain more dynamic range in the image while still keeping it looking natural than either ACR or Canon's in-camera processing.
It's also worth noting that I feel Canon's color is more pleasing than that delivered by Camera Raw's profiles - e.g., ACR 4.4, and that having more or less one expectation for the color delivered by either camera setup (Raw or JPEG) instead of two enhances the use of the equipment. I choose to use both Raw and JPEG settings, depending on the need.
Thanks for your feedback.
I'm fairly new to raw developing and grapple with the same issue as you, but I've come to the conclusion that as long as pleasing color is the goal, tweaking each image is always going to be required no matter the custom profile, camera used or raw converter used and I've tried nearly all of them.
Establishing some kind of efficient workflow relying on a new profile and expecting it to render a perfect image that is pleasing to the eye is from my opinion based on limited experience impossible when shooting outdoor shots similar to yours.
I like to shoot the variety of geese at my local park under a late afternoon sun and I get so many variances in exposure and color temperature between each shot I end up having to tweak every image to get it to look pleasing to me or at least correct looking.
Using a calibration script only gets certain hues correct for some images where sometimes the ACR default profile makes the image look perfect and the cal script makes things too red and oversaturated. Add definition and contrast and the color temp is off.
It's always a fight between a color temp that makes earth tones look too greenish yellow and finding where to slide the red hue to the left in the calibration tab to get that balance between the color temp appearance. I love the look of earthtones lit by late afternoon sun, but it's such a fight because these colors are close to gray and need to appear less saturated compared to primary colors making it tricky on the eye.
See for yourself from my screenshot of images I've processed in Bridge:
It is my observation this is simply not an "all or nothing" issue, but really a study in subtleties. There is certainly no "right" or "wrong" color, and in my opinion no such thing as "accurate" color. There seems to be "good" color and "not so good" color, though even then the answer differs from person to person.
Of course one can tweak all the converter settings; that's the idea. :) I especially like the settings that overcome lens deficiencies.
But what I like best is being happy with the results, and when the default settings deliver good looking images a little more often, then the happy results come easier.
Nice geese, by the way. Just keep them away from the airplanes.
Thanks for the comments on the geese. BTW that's an Egyptian goose/duck? in the main panel. I'm using it to calibrate my earthtones correctly. I had to make a mental note after leaving the park of the noticeable amount of brownish yellow in the dirt and limestone rocks so I could render it as I saw it. I have a poor memory about these things.
I think I spoke too soon about the inadequacies of the ACR profiles. I remember downloading the latest camera profiles from the DNG profile site and the Adobe Standard Beta 2 deals with this greenish yellow issue between color temp as shot and the saturation/hue shift with earthtones. At first it makes the dirt look slightly magenta but when adding definition, contrast and beefing up saturation the inevitable hue shifts that occur makes the dirt and feathers look correct. The saturation is more even throughout the image so I can now just apply Vibrance and I finally nailed it.
I got a better rendering of the red brown feather without adding red to the rest of the image. Saved the settings and applied it to the shot of the two geese on the far left and it got rid of that greenish looking dirt.
This is so much fun learning about this stuff.
Check it out. I got the brown I was looking for and I didn't even have to to touch the calibration tab or use the calibration script. A lot less work getting there.
Any way you can post the raw file?
I have been following this thread for some time now and find it quite interesting and would love to try those profiles of yours. Could you make them available?
If I can find an ftp file delivery site that works I'll give it a go and post the link in this thread.
I tried YouSendIt and all it does is just show a progress bar within the browser that never progresses. My At&t DSL modem upload lights don't even show activity. I've never had luck with that site unlike others here who seem to have no problems. I may be doing something wrong. I just don't know what that is right now. I've never had to upload anything other than jpegs which are no problem.
Why do you need the original raw of the Egyptian duck?
I used Firefox and was able to upload. Mac's Safari seems to be the problem. Here's the link to my Pentax K100D PEF.
If it doesn't work let me know. I have another link that was delivered to my email that's different from the one that shows up on that site's page after upload which is the one I've posted here.
>Why do you need the original raw of the Egyptian duck?
Thanks for going to the uploading effort. I'm sorry; I thought your image was done with a 40D, because of the thread title, and I was going to try a conversion with my profile and post the results. I just read back and noticed you mentioned you're shooting with your Pentax, a fact I completely missed before.
That duck image responds very well in ACR 5.2 with a -30 exposure adjustment, +25 Fill Light, +33 Vibrancy and the ACR 3.6 profile.
Sorry for not mentioning the camera model, Noel.
With numerous posts on this subject with mainly Nikons and Canons, thought I'ld show how off they can be with another camera to rule out subjectivity and personal taste in color. There doesn't seem to be a quick way to know when an image should look finished since most digital raw captures and even jpegs already appear quite good to begin with using ACR's default settings.
There's so many tools now within ACR to get me to where I want to go that I end up on quite few of my images coming full circle after long sessions of tweaking to arrive pretty much close to what the default settings provide excluding bad exposures and color temps. A change in profiles add only subtle changes that add an extra kick to the image that makes me question my perception of color. I end up asking myself if it really is the profile's fault or just me and the way I see things that affect my color adjustments.
I tried your settings in ACR 4.6 and the duck's brown tail feathers are too caramel-(yellow brown) using ACR 3.6 profiles when they should be a deep red brown as seen in similar shots of the same duck here...
This can't be fixed with color temp adjust. This time a change in profile was in order. The Fors and Tindeman calibration scripts would keep adding magenta into the yellowish brown of the ground soil trying to get that red brown in the tail feathers.
I started carefully studying what ACR's color temp sliders were doing to the RGB channels to get that warm yellow brown without getting a sappy saturated greenish yellow and found the hue of blue to the left and orange to the right with the top slider was fighting the calibration slider hue adjusts.
When red hue cal slider is pushed to the left to get the red brown instead of caramel, the yellow brown dirt takes on a magenta hue, so I have to adjust the bottom color temp tint slider to the left which introduces green and the tweak session begins back and forth between the two adjustment panels.
I see the same issue with Noel's posted image samples and in other similar threads just not as pronounced as mine and I'm finding most prefer this magenta/bluish tinge to there renderings which points to the fact that all of this is very subjective or maybe my eyeballs are getting long in the tooth.
Here's what I got using your settings, Ramón, compared to switching to the Adobe Standard Beta2 profile.
Did you get the same results? Thanks for the input on this.
>tried your settings in ACR 4.6
I used ACR 5.2, as stated in my post.
I'll post a JPEG in a few minutes.
ACR 5.2 with a -30 exposure adjustment, +25 Fill Light, +33 Vibrancy and the Adobe Standard profile. <br /> <br />c <a href="http://www.pixentral.com/show.php?picture=1IpIOuaupJuGEL4o3M1N2MaXmZh1dk" /></a> <img alt="Picture hosted by Pixentral" src="http://www.pixentral.com/hosted/1IpIOuaupJuGEL4o3M1N2MaXmZh1dk_thumb.jpg" border="0" /> <br /> <br /> One correction to #35: I used the Adobe standard profile, not <s> <i>3.6</i>.</s> Sorry.