Good point, and there are programs to extract that jpeg, but it is a relatively low quality jpeg, I forget the exact level.
In camera or in Post
- Thread starter sonya580
- Start date
If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).
Good point, and there are programs to extract that jpeg, but it is a relatively low quality jpeg, I forget the exact level.
Reading this thread I'd say the other point of potential confusion is what it means when a raw converter 'applies camera settings'. As Steve and others have mentioned it really only means the starting point for raw adjustment sliders (and curves, etc.) are set according to the camera settings but the adjustable parameters are not really baked into the raw conversion at that time and those adjustments can be changed in post with no image degradation.
That's very different than what happens with camera settings to a TIFF or JPEG where those initial settings are actually 'applied' to the conversion in camera and any changes in post is working at a disadvantage compared to raw file processing. That's true whether you use a brand specific raw converter that by default applies camera settings or use something like LR and choose Camera Settings as your processing default in LR Preferences.
For folks shooting lower volume and choosing to get the most out of every image they process it can make sense to simply adjust each image from scratch without worrying about the camera settings. I would say that sorting and culling images can be easier when the image is closer to the final edits and pre-adjusting the conversion sliders based on camera settings can sometimes make that initial sort easier even though you can then go back and choose very different conversion settings without image degradation.
That's very different than what happens with camera settings to a TIFF or JPEG where those initial settings are actually 'applied' to the conversion in camera and any changes in post is working at a disadvantage compared to raw file processing. That's true whether you use a brand specific raw converter that by default applies camera settings or use something like LR and choose Camera Settings as your processing default in LR Preferences.
It really depends on the photographer and their working style. For folks doing high volume event work including things like wedding reception photos or sports photography it may not be practical to hand process each and every photo. A lot of folks doing this kind of work shoot jpegs and rely heavily on camera settings. But others still shoot raw, start with their conversion program defaulting to the camera conversion settings but they have a bit more latitude for particularly good shots that need some custom conversion help like pulling shadow details.
For folks shooting lower volume and choosing to get the most out of every image they process it can make sense to simply adjust each image from scratch without worrying about the camera settings. I would say that sorting and culling images can be easier when the image is closer to the final edits and pre-adjusting the conversion sliders based on camera settings can sometimes make that initial sort easier even though you can then go back and choose very different conversion settings without image degradation.
Makes sense. Lightroom is also shooting for a middle ground in recent updates where it can be told to automatically apply presets based on certain camera settings and iso that it reads from the raw.
Yes, many of the settings in the camera are relevant for JPG only - and they find theier way also in the JPG previews as mentioned by @kimball .
However, even the statements you find about which of those settings have an impact on your RAW files and which ones not is not consistent and to a certain extent contradictory.
I made some tests recently because I decided to move away form Nikon software because since Capture NX2 died there is no selective post processing by picture area any more. Also my friend needs to change software, but because he has tens of thousands of images processes with old Nikon software he wanted to know excactly what will happens with his existing work. During these experiments we also cxame across some aspects around the in-camera settings.
To write it down in one go would got too far but I try to give a summary:
- Tests were done with Capture NX2, Capture NX-D, Capture One and DxO PhotoLab 4
- The only programs that allow 1:1 access to in camera settings of Nikon cameras is Capture NX2 and Capture NX-D.
With full 1:1 access I mean that e.g. for the single parameters like sharpness, detail contrast, etc. behind the Picture Control Profiles like e.g. "Portrait". You can see in the software with which profile and which setting of its detail parameters an image has been shot and you can change them and you can see the effect right in the raw picture you are processing. - There are other things in the NEF files that usually get olst in thrid party software, e.g. the focus point that was engaged during caturing the image can be shown by the Nikon software but usually third party software doesnot use this information.
- As opposed to some statements I found in other forums the Picture Control Settings have an effect on your NEF's and the way they look and it is to some degree dependent on the software that you use.
- Example 1: White balance
- You can change manually adjust the white balance in the camera for each profile (e.g. Direct sunlight)
- Let's say you make three images, one with neutral setting (cursor in the centre), one with the cursor in the magenty corner and one with it in the green corner.
- When I opened the raw files with a Nikon program, you would see neutral previews for a short moment and after the software has done its initial read of the NEF's you can see them change color according to the setting in the camera. You start processing your picture based on these color settings coming from the camera.
- Opening the same raw files with another program - in this case PhotoLab 4 - I initially got a different result. The program ignored the setting in the camera and showed all three pictures looking the same.
- After the last update I installed a couple of weeks ago I discovered accidently that the behaviour had changed. Now the software is reacting to the white balance settings in the camera and shows a a neutral a pinkish and a greenish image when openting the raw for etiding.
- Example 2 : Picture Control Profiles
- Let's say you make a series of images on an object and change the Picture Control Profile for each of them
- Opening the Raw files with a Nikon software you will see the differences.
- Opening the Raw files with another software (in my case PhotoLab 4) the images looked all the same, even the image shot with Monochrome was shown with normal colors (!)
- Example 3 : Changing individual parameters of a Picture Control Profile
- I made two images with profile Neutral and set the sharpness to zero for the first one and to 9 for the second one.
- Opening the raw files with a Nikon software the differences are visible.
- When opening them with PhotoLab 4 there diferenvces were not visible, but the program seemed to aim for the optimal quality automatically, i.e. both pictures looked similar to the one with sharpness set to maximum n the Nikon program.
- Example 1: White balance
- Unfortunately there is no easy answer to the initial question. How a software interprets the native raw format of a camera manufacturer is in the hands of the software manufacturer and as I could observe, this can change with software updates. That said, I would expect that the changes go towards better and more complete interpretation. I think software manufacturers strive to use the infomation hidden in the rraw files the best they can as long as it does not collide with their software architecture or their functional strategy.
- Is it a problem, if software manufacturers do not use all information from a raw file ?
From my perspective the answer is no, as long as you don't loose primary image data.
The only question is how you process the image data (Raw file) with your software.
I changed from Capture NX2 to Capture-NX-D, tried out Capture One and ended with DxO PhotoLab 4 that I am really happy with. I never missed being able to access Picture Control settings made in the camera later on in the software one single time and up to now I haven't found anything that I lost due to the fact using another software. The most important thing to me was that the workflow based on the native raw file - in this case NEF - is non-desctrucitve, uninterupted in terms of forced format changes (to TIFF or DNG) before exporting the final results and allows random sequence workflow.
- Originally I was told to stay with Nikon software because at that time it was kind of common sense that the camera manufacturer and inventor of its native raw format has always the best chance to get the best out of the image data and thus his raw interpretation algotrithms should be the best. These days are over. I can't speak for others but I have all my cameras on one and the same Picture Control profile setting getting close to my preferences and leave them in this state all the time. Everything else is done in the post processing software.
IMHO there is no significant efficiency advantage in using in-camera picture control instead of doing it in post processing, because modern software supports things like presets and batch processing of large amountof images - IF YOU SHOOT RAW. For people that are working under high time pressure requiring to get "ready" shots right out of the camera, in-camera Picture Control still makes perfect sense. The same applies to to people shooting direct JPG for other reasons. As JPG processing often means quality loss it is better to get the best possible JPG out of your camera in the first place. And because the in-camera setting for picture control has an influence on the way the image is shown on the camera display, even raw shooters might find it useful to adapt certain settings just to get more out of the rear display, but that's it.
kimball
Brian Kimball
Great post; I hope you don't mind me quoting just the relevant parts.
I think it's important here to distinguish between raw image information and raw metadata information. The full and complete set of raw image information is always used by a raw processor. Not so with the metadata, such as white balance, chosen picture profile, and picture profile settings like sharpness and even B&W status, as you found. Slowly some 3rd party raw processors like Lightroom Classic are starting to honor the raw metadata the describes the state of the camera's JPG engine, like picture profiles. But I think that's the only one.
I think it's important here to distinguish between raw image information and raw metadata information. The full and complete set of raw image information is always used by a raw processor. Not so with the metadata, such as white balance, chosen picture profile, and picture profile settings like sharpness and even B&W status, as you found. Slowly some 3rd party raw processors like Lightroom Classic are starting to honor the raw metadata the describes the state of the camera's JPG engine, like picture profiles. But I think that's the only one.
Lots of good nuggets in there. I've definitely seen some Nikon raw shooters make custom picture profiles with really odd tone curves that help them better visualize their ETTR process. Makes the resulting preview image garbage, but it works for them...
Larry Shuman
Well-known member
I've seen it work too many times. I've shot an eagle and looked the preview and am I am amazed at the sharpness. When I download using Transfer 2 and open the nef in PS CS6 I see the same sharpness before I perform any edits.
Of course not, it is just personal thoughts around the topic that I thought might be of interest for somebody.
Good point !
That said though, using the term metadata might be a bit confusing depending on the context as the understanding of metadata usually looks at things like EXIF, IPTC and XMP data, but the border line is floating between data that just describes how exaclty an image was taken and information that can directly be used to interprete the raw image data in a defined manner. E.g. the setting of aperture, shutter speed, EC, etc. are part of the metadata but the raw image data already inherits it, because the settings direclty influence the brightness of the picture in terms of modifying the primary raw data. The information you refer to is not part of the primary raw image data, but is added to it as a kind of separate data package designed for instructing the raw processor about how to interprete the primary raaw image data. This is where the out-of-camera software comes into play and where the differences show up betwenn the different packages. As I stated ealier the behaviour of the software I use changed due to an update from ignoring white balance settings coming from the camera to interpreting it and showing the raw file accordingly after initial interpretation.The third category finally is meta data that is just there for supplemental information an has nothing to do with how the image looks, e.g. geo tag information or copyright.
Looking at it from the sioftware manufacturers side it is a difficult decision how far you will go. Of yourse you are interested in supporting multiple camera manufacturers' products, but because of there number and the evolution of their native raw formats the effort that has to got in the software to keep track with all that is pretty big and complex. If you wanted to provide this kind of functionality for multiple models of multiple manufacturers on the level líke the Nikon software allows to read and edit raw this kind of raw image meta data you must be able to change your user frontend in the software dynamically depending on make and model of the camera that the image has been taken with.
I know similar things from machine communication interfaces in industrial automation solutions and - to be honest - it is a pain in the a**
. The attempts to solve this problem with standardization did work just as "well" as the attempt to achieve a standardization by means of the DNG format - which in fact was an attempt of Adobe to get the camera and the software companies on a short leash making it easier to make money through licensing fees ... Nasim Mansurov has written two pretty interesting article about this (article 1, article 2) and although being a bit older already I think they are still describng the situation pretty well.This was the reason for me to change over to DxO PhotoLab. It works directly and purely based on the native raw, i.e. NEF in my case. Interestingly this program also gives an example of an alternative way to handle things that were originally handled in-camera:
If your raw processor has the functionality to use lens profiles and providing they are good and correct, the lens correction function in the camera might become obsolete.
Now the interesting question is, whether the in camera lens correction is stored as part of the primary raw image data or is it stored as meta data telling the raw processor how to correct the image. In my case (DXO PL4) the software uses a profile dedicated to a particular camera-lens combo and I can see the image changing when activating or decactivating the use of the correction profile. What I don't know yet is whether or not the pirmary raw image data has been altered in the camera by its lens correction function or not.
Great idea
, but if they had known what @Steve was telling us about "Blinkies" they probably would have been able to survive with that as well .