Yes, profiles are model-specific, and if you have similar-sounding models that is not enough to make the profiles show up in the popup. Namely, if you have a D300s, the profile popup will only show you D300s profiles, nothing else.
This "default" position of exposure slider is just a position where ACR doesn't make any exposure correction to raw data. In case of D90 this position is +0.55 and not 0. Since I calibrated a profile according to result from NX2 with no exposure correction, photos from both NX2 and ACR will also match in that default position of exposure slider. Of course, for real photos, you will move this slider from this position up or down if needed. For the same reason, I used blacks 0 instead of 5.
Since this exposure correction is positive, I could also make a profile that doesn't need any correction (with default exposure = 0), by raising the tone curve in the profile. Adobe profiles for D90 are made that way. It is not possible in case this correction is negative (D700 for instance) without clipping some highlights, that's why this default position isn't 0 for v3 beta profiles. But I prefer making it this way for some reasons
I had a chance to examine an aspect which this beta addresses...banding. Now I don't have a clue as to what and under what circumstances the banding fix is addressing, only references to "reduce banding and highlight color artifacts." I have a file of a foggy beach with the sun trying weakly to break through. Using Recovery, I brought the brighter portion of the fog to just slightly brighter than it's immediate surround with no highlight clipping. Then I engaged Levels, ran the highlight slider to brighten the fog, and banding set in. (I noticed this from an earlier editing session; I didn't use Levels but it seemed the most expedient here) and sure enough, a subtle rainbow banding appeared. Running this on both Camera Standard and Vit's "NX2 Standard T02". Same thing.
I realize that all I could validly summarize is that the editing steps I used facilitates banding. And I cannot assume that there is banding occuring using Camera Standard that NX2 Standard T02 fixes. I don't even know if the rainbow effect is color artifacts. (Feeling stupid about now! ).
So there it is.
I certainly understand from where you are coming. Testing is testing and standards need to be acknowledged. I am not suggesting that one or the other is "right" only that in nulling out the slider (and reducing Black to -0-) I get a match to Camera Standard, whatever "Standard" means!
There are two meanings to the word Standard, which are loosly related. I think the formal meaning for engineering when I see "Standard" as in :
When is a standard not a standard?
When there is more than one...
I think in this case, standard will be whatever the photographer says it is...for him(her).There are several ways on which to "Standardize" in evidence here.
So, what do I look for? What's beta fixing? Is "NX2 Standard T02" the beta setup to be recognized by D90?
Well, this profile for D90 is just my attempt to make a profile in similar fashion like v3 beta profiles that Eric published at the begining of this thread, so take it as is. It is not made by Adobe, but with my program that I made for profiling Canon compacts.
It can be shown that both this profile and v3 beta profiles are not as smooth as the profile in NX2, so there are chances that you will see more banding than using NX2, if you are playing with levels and other tools. Dng profiles in general use trilinear interpolation method for rendering using 3D lookup table, hence the difference. There should be less banding with this and v3 beta because lookup table is bigger (90x16x16), and probably calibration method is slightly better, and it could be further reduced by enlarging the lookup table, but that would also make the profile bigger. Also, even profile in NX2 is showing something like banding in some parts (mostly on the gamut boundaries), so it's not perfect either.
I took a look at NX2 (actually the program on the other computer is View NX, not NX2) and found that it operates suspiciously like the Nik software with control points set up the same way.
I'll have to download the free trial and check it out...again. I tried it when I purchased my first Nikon DSLR and decided I didn't need two editing programs.
At this point, I'll have to stay with the ACR setup as I do use all the profiles at one time or another so for consistency, that's where I'll settle until some time down the road.
I'll be lurking here though.
Thanks, Vit and Flyshow for the help.
All these profiles are "twisted" in Sandy's teminology, because lookup table is 3D. Tried "untwisting" one of these profiles, that actualy converts 3D lookup table to 2D, so rendering is the same for all brigtness levels (the same HueShift, ValueScale and SaturationScale). It didn't perform well in highlights, because camera profiles are intentionally heavily "twisted" in that region to produce smooth output (as much as it is possible) in blown parts of the picture and avoid various kinds of banding
Untwisting I do get awful orange hightlights similar to adding fill light and some recovery too.
Frankly I don't really understand the 'intention' because for me the profiles are useless for recovery. I have an alsatian dog picture which goes from a golden brown to a sickly magenta.
Seems to be worse with the standard profile than anything else but still it's horrible.
This annoying bug lasts so long time and will be until Adobe will change algorithm of processing files. So just now recovery slider completely unusable with skintones on pictures.
Use brush with negative exposure to recover highlights.
Brush can also lead to strange effects on blown parts of the picture as soon as you move exposure of the brush from 0, and it happens even outside brushed parts (already discussed here), so I prefer combination of recovery and exposure correction
Digging further in this profiles stuff, I found out that this banding issue, mentioned by Hudechrome, can be quite visible on some cameras, like D7000, which already has a kind of "v3 beta" profile (although not visible from it's name). I made my profile for it, but it's not any better in this regard. Will post a sample showing this problem in near future
I took a look at the skin tone problem mentioned by flyshow, and at this point, I definitely can see differences between v3, Vit's version and Camera Standard. Vit's version seemed the least damaging, and in any case, using Recovery at all is only a last resort, preferring Exposure first.
I had an image where the subject is 3/4 backlit so that the skin tones were uniform and the correct value but the white on the hat was a bit overblown, perfect for testing. None of the versions exhibited skin tone color problems when using recovery to try to control the blown whites, and Vit's version had the most interesting effect of all, tending to lower the skin tone somewhat to produce a more natural color while Camera Normal and v3 exhibited less width in it's range. On landscape images, Vit's went too far(dulled down high key values too much)so I would prefer the others when working there.
I also installed NX2 this morning. Nice that Nikon gives 60 days for the free trial (60 days Adobe! ) and a quick look at how they control highlight protection, as they call it, places it well behind ACR in all it's permutations/combinations. But that is a quick look.
Back to banding. Banding in any case has always been an issue, going back to PS5. I have a b&w photo of Mt.Hood perfectly reflected in Trillium lake I did in 1972 and still the best print of that neg is the analog, where, obviously, no banding can occur. This is the bane of digital However you look at it, a digital image is only an approximation of the captured image compared to analog.
Yes it seems the 'bug' has existed for a long time. It's a shame as I was happy to see the advent of dng profiles only to find that they aren't as useful as I thought. Exposure can also alter colours slightly I've found.
I guess I have to stick with NX2 as my main processor but to me it's so much slower than LR so processing a large shoot can be a pain.
Thanks for that link Vit! I hope it didn`t disturbed you to much to make the search for me;)
However I wanted to make a notice that I`ve posted a question there in case you are not following that thread anymore.
It is unfortunately not very helpful in this case that LR's Exposure steps in Quick Develop are 1/3. Quick Develop is LR's only place for relative adjustments and -0.5 EV cannot be obtained this way.
This shows that LR would really need a better implementation for relative adjustments. They need to be accessible in the Develop Module and they need to be able to go into Presets!
For any already processed job it's almost impossible to implement the v3 profiles for now. And if processing is outsourced, it's also a problem. Processing service providers are not likely to add in v2 or v3 Beta profiles and, as stated above, it's pretty hard to apply the v3 profiles after the xmps come back from the service provider.
Better relative adjustments and full keyboard shortcut implementation - but that's another story ...
1/3EV (.333) is more accurate than 1/2 EV(.5). It possesses better resolution, leading to better accuracy. The most accurate, of course, would be continuous (stepless) changes, such as found on aperture settings for lenses for large formats. However, with those lenses, resetting to an accurate EV is difficult, because the shutters are discontinuous. The advent of EV started when camera manufacturers such as Hasselblad adopted the EV system, which required a discontinuous system to be effective. It was very annoying, perhaps the most annoying aspect of using Hasselblad, about which I grumbled quite a bit. Some people even resorted to having camera repairmen file off the gearing so that the iris became continuous again! (I didn't). Instead, I had to accept that while the exposure now possessed less accuracy, the repeatability was better. Now you could function mentally in either shutter priority or aperture priority
So, rather than have LR become less accurate in nailing exact exposure by changing the increment to 1/2, rather have the entire process be set to function in 1/3.
Of course, the sliders are continuous within the native limits imposed by the digital process.
Could somebody be really kind and reupload the original v3 betas made by Eric as the link no longer works?
Thanks very much!
Can you easily explain how to do this so that I can create camera profiles for my Canon G11?
Would this help with matching colors between my G11 and my Nikon D700 if, let's say, I'm shooting with both cameras at the same location?
I don't know how many models gynngee has done this with (and had "perfect" results), but if it were always that simple, Adobe would not have different files to implement the same profile on different cameras. Different cameras may have different sensor response characteristics, so how well this works will depend on how similar the two camera sensor response characteristics are, and maybe some other things I don't know about...
Regarding more officially supported ways of doing it:
One easy way to get the same look on different camera brands is to choose an Adobe profile (i.e. Adobe Standard or ACR 4.X...).
The other way is to use the DNG profile editor (read the doc).
To match two cameras, at least to some degree, you use a color-checker and Adobe or X-Rite software to create a custom dual-illuminant profile, and/or set of profiles for different lighting conditions, based on the Adobe Standard profile for each of your cameras.
To create a custom profile you buy a Color Checker Passport from X-Rite which includes the software or use a larger, standard-sized Color Checker also available from X-Rite and then use the DNG Profile Editor from Adobe Labs.
DNG Profile Editor:
Datacolor / Colorvision also has a new camera profiling package called SpyderCheckr:
Just to be clear, the Datacolor doesn't provide a DNG Profile as a result, it produces a set of modifications of the HSL sliders in the form of Presets. Not the same thing at all and depending on your needs, less optimal than having a real DNG Profile. You must limit your adjustments of the color rendering to the HSL panel so it eliminates the use of HSL in other presets (otherwise you overwrite your HSL settings).
Most of the false profiles I did personally were done Pre DNG Profiler or CC Passport. In fact I refer to them as presets. They were presets using a Macbeth card and a combination of the Tone curve , camera profile and HSL tabs in ACR . Being a Nikon- Fuji user my main objective was to try to produce fuji -portrait -fuji - velvia look with my D200 -D2x cameras. I found reproducing the ACR interpretations of the Fuji more successful than reproducing the colors given by the Fuji software. None of them were perfect. But as presets I found it easier to address color problems on an image to image basis. The big problem colors from memory was Blue - Cyan and Orange - Brown . With skin tones on a Fuji the shadow is a lot redder and more saturated than the Nikon ( being more yellow and muddy). When this was addressed the oranges went way to red. Later after Dng profiler was released I had a couple of goes but from memory it had different issues. I still occasionally use these presets on my D2x's & D200 but if final output if within the confines of the S5 I will use one of my S5s, those files are like putty.
The DNG Profile Editor/X-Rite Passport software and the Datacolor HSL presets would seem to complement each other, so I'd likely use both.
I would suggest you look a bit further...dealing with a Preset vs a DNG Profile is different...
I would use Adobe Standard in place when computing multiple Datacolor HSL presets for various lighting conditions.
I would also Adobe Standard as the basis of a dual-illuminant DNG Profile Editor profile and have this custom profile in place when creating multiple Datacolor HSL presets for various lighting conditions.
Then I would use Imatest and my eyes to see which was preferable.
I realize the Adobe Standard profiles have hue-twists in them and the DNG Profile Editor-generated ones do not, so it's possible one scheme would be preferable over the other.
are this Camera Profiles usable also on a D3X. Maybe not, I tried to install in the camera profiles folder under camera raw but they are not shown with files made by my D3x...
I draged the "Camera v3 beta" folder to (Win7): C:\ProgramData\Adobe\CameraRaw\CameraProfiles.
There are 2 other folders: Adobe Standard and Camera
Within LR / ACR are just standard Adobe settings, nothing about "Nikon ...v3.dcp"
Tried different places, but still the same ...
What am I doing wrong?
PS: After unzipping filenames showed up in green instead of just black (Windows Explorer). Changeing this did not help either.
Given your mention of the color difference, is it possible you dragged a zip file into the CameraProfiles folder? Windows Explorer attempts to make zip files look like folders, so there could be confusion there, but Adobe will not read information out of zip files directly.
Try making a subfolder under CameraProfiles called Custom. Then try dragging the .dcp files specifically into that folder.
I'm assuming your camera is one of the ones noted.
It sounds like you are doing nothing wrong.
These beta profiles only support the D3, D300 and D700. I assume you are using RAW (NEF) files in LR, and that in LR > Develop Module > Camera Calibration, you are clicking the Profile dropdown list. The beta profiles use the same names as the Camera profiles, but the beta profiles have v3 after the name.
One more thing...
When I add profiles for my own camera, I don't put them in the location listed above, I put them here:
This is more flexibly expressed in the general case as the following, which can be pasted into the Windows Explorer address bar:
Oh, I guess the problem might be this: mine a D300s and this will not work?
I found somewhere else a dcp file for the D300s and this works.
Is there a change to "switch" the D300 files with the DNG-Editor?