I\'m by no means an expert and what I say applies only to Zeiss glass so it\'s not general.
1. Colors & Contrast
In my experience images taken with Zeiss glass are best processed as little as possible. When I use Canon glass, I need as a rule to bump contrast, vibrance and clarity in post to make the image usable.
Example, unprocessed & processed Canon:
What I\'ve done here is rotated the image a bit to level the horizon, increased vibrance (+30), black clipping (+7) and clarity (+100).
If I was to do that to a Zeiss image, the result would be very bad. Example (unprocessed first)
Apart from the overcooked colors look at the rocks in the foreground - the clarity added in Lightroom has completely killed off the fine detail.
So in my experience a good rule when dealing with Zeiss images is don\'t touch the clarity slider.
There is in fact only one major color correction action that I can recommend and that\'s setting the white balance. When shooting in evening light or at night it gives creative options without compromising the integrity of the image rendering.
Example:
In-camera white balance:
Custom white balance set so that white is really white:
Which one is \"real\" is debatable. The first one looks more like I saw it but the second one represents more the true colors of the scene as it would have looked under neutral light.
2. Sharpening & Resizing
While color correction is fairly irrelevant for PP:ing Zeiss images, sharpening when resizing is very important. Do not use Photoshop\'s default bicubic resize as it will destroy all the finer detail in the image by averaging surrounding pixels. Again this is a thing much more important with Zeiss images than for instance Canon. It\'s always good to sharpen when resizing but it is really critical when using Zeiss glass.
Example - I suggest opening the two images in different tabs in the browser and flipping between them:
Photoshop bicubic:
My script:
So what\'s my script? First, I want to advise against taking it as a solution for everything. I\'d say that it works perhaps with 75% of my shots. In the other shots some form of manual intervention is required as the image can get over or under sharpened. Bokeh and background blur can be a problem. Furthermore I have experimented very little beyond my initial script which was vaguely based on one by Samuli. I have not tried to improve upon it which I\'m sure can be done. I would more recommend Samuli\'s source material.
Also worth adding is that Makten has demonstrated that it can be beneficial to just sharpen the L channel in LAB mode rather than to apply sharpening in RGB mode. If you do it in the L channel you won\'t amplify color noise and CA.
That\'s it for my PP basically. I on occasion set the white balance in Lightroom, export the image to photoshop and do a resize & sharpen. If my script produces bad results I go in manually and see if I can fix it.
Sep 14, 2010 at 04:19 PM
Previous versions of denoir's message #8846510 « Post Processing Techniques »
