I've read many posts on post-processing workflow, but I am interested in information on in-camera processing settings.
I've read that it is best to sharpen jpgs at the end of post-processing and that using a lot of sharpening in the camera is detrimental since sharpening can not be reversed. Most of the picture controls in my D5100 are set low to a 2 or 3. I've read suggestions up it to 6 or 7, but I assume that is good only if one is not doing any post processing.
1) A blog stated that the only sharpening in-camera should be to eliminate anti-aliasing. I wonder if that is why the D5100 is set to 2 instead of 1?
2) I think that many users think a low in-camera sharpening setting causes the camera to soften (blur) the images. I am thinking that a low sharpen setting merely indicates a lack of sharpening, and not a softening of the original image. Is this true?
3) I am using PSP X4 right now. I am shooting some very contrasty subjects (see my gallery). I find (with my limited experience) that keeping the sharpening in-camera to a 2, and using ADL at its highest setting gives me a flat image that retains shadow and highlight detail. I boost the contrast post-process, as well as color intensity and finally sharpening. I probably have overdone the post-processing, but hope to normalize this with experience. Can someone suggest a workflow that also considers what to do in-camera to keep it all compatible with the later post-processing in software?
4) I have a feeling that using RAW simplifies this but please explain. I wonder what in-camera processing is done to RAW images if any (such as anti-aliasing perhaps?)
5) Also info on any books or web sites that discuss this is appreciated.
#1. "RE: Pre-Processing In-Camera" | In response to Reply # 0pdekman Nikonian since 17th Nov 2005Wed 29-Aug-12 02:35 AM | edited Wed 29-Aug-12 02:37 AM by pdekman
Hi Steve -
Workflow questions can be tricky because different people have different needs, but I'll share my thoughts...
In order to create a viewable RGB image, digital cameras must take the raw sensor information (bayer array data) and render an image using a particular recipe for white balance, color space, sharpness, etc. If the image is stored as a jpeg, further processing is done to compress the image for reduced file size and reduced color information.
Shooting a jpeg is basically doing post-processing, but using the in-camera settings for the recipe. Your questions regarding how much sharpening to apply to the initial image assumes that further post-processing is expected to be done on a computer. While this can be done successfully, recognize that multiple edits to a jpeg image is not preferred because of the compressed format and loss of the original raw information. If I were to shoot jpeg, I would want to get the in-camera settings as close to the final usage as possible, leaving small or no edits to be necessary later on.
This is why shooting in raw is considered advantageous. The raw file keeps the bayer array data to which an endless number of image recipes can be created to render a viewable image. White balance, color profiles, saturation, sharpness, etc. can all be changed without any impact to the original photo information. This post-processing occurs outside of the camera which demands the user have computer skills but you also have the luxury of working on a large, color calibrated monitor.
Lastly, your sharpening information eludes to the common workflow of capture sharpening (to just remove image capture blur due to sampling) leaving creative sharpening (image dependent, perhaps local sharpening) and output sharpening (resharpening the image based on particular print size or web viewing) until later.
All of this probably opens more questions than answers. A forum search or web search should bring many good articles and tutorials.
#2. "RE: Pre-Processing In-Camera" | In response to Reply # 1Wed 29-Aug-12 03:15 AM
Paul, Thanks for the info and your insight. I think I might have discovered an advantage then to Lightroom 4 versus Paintshop Pro, and that is the raw data seems to be much better in Lightroom than in PSP. I have trials of both on my PC right now for comparison purposes.
I have a D5100. I am wondering the best way to set it up if I am going to use RAW instead of JPG. I am thinking that it may not matter how I set up the in-camera sharpening if I output only to RAW. Most of the picture controls have the sharpening set to 2 out of a 0 to 9 scale. I wonder if this is done for capture sharpening or does it only apply to JPG output. Does the capture sharpening occur without any user input or does one have to set it in a menu. Once I learn enough about the settings for best RAW output I can concentrate on properly exposing images, and worry about post processing until later.
Again thanks for your reply. And I have been reading as much as I can find on the Web, but the information about this is scattered.
#3. "RE: Pre-Processing In-Camera" | In response to Reply # 2esantos Nikonian since 10th Nov 2002Wed 29-Aug-12 03:49 PM | edited Wed 29-Aug-12 03:51 PM by esantos
As Paul points out a RAW image has no permanent or "baked in" adjustments to the file. I call it a file because in its purest form it is not an image file. With that being said the differences you are seeing in Lightroom and PSP have to do with the program presets or what is commonly referred to as camera profiles (that is the term used by Adobe which most here have adopted for simplicity's sake). I am not a PSP user but I can tell you that over the last few years Adobe has made great strides in improving their camera profiles - today they are as good as what you can get out of the Nikon DSLRs themselves. You should also know that Nikon Capture NX 2, Nikon's RAW processor, is able to read ALL your camera settings. You can also download this program on a trial basis and see if it works for you. All other RAW converters are only able to read the white balance setting you set in-camera.
About in-camera sharpening - the sharpening you can set in your camera is capture sharpening and unless I am mistaken is not applied by third party RAW converters, only when using Nikon Capture.
#4. "RE: Pre-Processing In-Camera" | In response to Reply # 3Wed 29-Aug-12 09:12 PM
Paul & Ernesto,
Thanks both for the support. I have been studying the on-line Lightroom 4 tutorials. Today I shot 110(RAW+JPG)of one dog. One shot included the owner. I decided to work with the RAW file in Lightroom. Although I had a little trouble with the library module I loved the develop module. The original file had a bluish light from the shade but the final came out very nice
I think I will continue using the RAW files and am ordering a copy of Lightroom.
#5. "RE: Pre-Processing In-Camera" | In response to Reply # 4
#6. "RE: Pre-Processing In-Camera" | In response to Reply # 0
I believe it is sometimes forgotten that in-camera processing produces by no law of nature a more truthful image than the output from post processing software such as Lightroom. Naturally, camera makers want to give an impression of having smart settings, etc. but there are no definite formulas for this complex task. Having said that, today's cameras perform a surprisingly good work on this subject.
The most important setting of course, is the white balance. It is not possible for the camera to know exactly how the color of the light in e.g., sunrise should be set, any more than the post processing software is able to know it. Indoor spaces often have multi-colored lights which adds its own challenge to the photography. Eventually, it is always up to the human brains to analyze the image and adjust the image -- or adjust the eyes to the image. With this capability, we do not see the colors absolutely.
In this "light", using any of the in-camera processing modes is generally speaking as accurate as using some predefined settings in post processing software. Lightroom, for example, has a number of different settings. However, they are only approximations but sometimes good places to start finer tuning.
I work solely with raw images and always adjust the colors later. Depending on the subject and image type, I usually estimate the color either visually (by adjusting the temperature slider) or by using a gray card or something I know is white or gray. I have noticed that the default processing of raw images in Lightroom often is very close to what I am looking for, especially in pictures taken in normal daylight. However, I do understand if others have different opinions on this.