Homebrew Sabbattier Effect

December 29, 2009 at 2:57 am (Computers, Howto, Photography, Software) (, , , , , )

In photography, the Sabbattier Effect is a technique originally described by H. de la Blanchere in 1859 whereby some of the images tones are reversed. It is also commonly called solarisation. The technique seems to have been ‘invented’ many times over by many different people, the common thread being that it was introduced by inadvertently turning on lights during darkroom processing. After many years, the effect can be created while processing negatives, while creating prints, or in our case, digitally using a computer.

In any event, the effect creates a high contrast look that is especially well suited to architecture and other man-made subjects. In my opinion, the effect does not work all that well and looks rather unpleasant on human subjects and scenes from nature. Of course, given the right picture to begin with, there are always exceptions. Your best bet is to experiment.

Many photo editing software titles include a solarisation filter, however, if yours does not, or if you want more control over the final look I will describe a method for creating the sabbattier effect using only the curves tool and a desaturation filter (ie: convert to black and white).

The first step is to choose your subject, and to convert it to black and white. I choose a picture of the Erickson Building in Penticton BC, that I had cropped square. Here is the original shot:

Now after converting to greyscale open up your curves dialogue. Basically what you want to do here is create a ‘sine wave’ shape. Hard to describe, but easy to show…here’s what it looks like in digikam, including a preview of the result:

This step is wide open to experimentation. Try different shapes and levels on your curves. The more ‘oscillations’ you use in your sine wave, the more pronounced the effect will be as you can see here:

You’ll notice that abusing the curves tool in this way has introduced some colour artifacts back into the picture. Not a problem, just convert to greyscale once again, and that’s about it. Again, experimentation is the key here. Twist those dials and see what you can come up with. Now here’s the final results, first the mild version, and then the more extreme version:

So there it is. Please leave a note if you found this useful or not, or have any other concerns, questions, or comments. Thanks!


Permalink 4 Comments


December 8, 2009 at 11:48 pm (Camera Gear, Computers, Linux, Photography, Software) (, , , , , , , , , )

So I get lots of inquiries on flickr and other places where I post my pictures on how much ‘photoshopping’ (terrible verb by the way, Photoshop doesn’t exist on my chosen platform) I do if any, and people want to know what software I use and all that. So: rather than repeat myself every other day, I am just going to write a detailed account of my workflow here, and link folks who ask directly to it.

When it comes to hardware, it’s pretty simple. I built my photo editing and main desktop rig myself. A few main stats:

  • AMD Athlon 64 X2 Dual Core Processor 5200+
  • 4x 1GB 800 MHz DDR RAM for a total of 4096MB memory
  • 3x Hitachi HDP72505 500GB SATA HDD for a total of 1.5 terabytes of disk space

I have it running dual-head and have two identical Acer 20″ flat screen LCD monitors. As far as camera gear, I have a Nikon D40 body, and a Nikon D90 body. For more info on my camera gear please see the about page, there is more info on lenses and all that.

On the software side of things, I should mention that I don’t run Windows, no, not Mac either. I’m one of those crazy software hippies that use Linux. This has, of course, shaped my digital photographic workflow to a large extent, as most commercial software available to Windows and Mac users is not available to me. To be perfectly clear, I am just fine with that fact as I much prefer to use open source software if it is available.

So: say I’ve been out for a long hard day of shooting. I should mention here that I always shoot RAW format, in my case, producing Nikon’s ‘.nef’ files. Upon returning home the first thing I do is import all my files to my photo organizing software of choice: digikam. My directory structure goes like this -> ‘Raw Photos’ -> ‘year’ -> ‘month’ -> ‘day’. If I was shooting with both bodies, each will get its own directory suffixed with the body model. So pictures I shot on December 4, 2009 look like this:

My Albums/Raw Photos/2009/12/2009-12-04-D40/
My Albums/Raw Photos/2009/12/2009-12-04-D90/

I put the full date in the last directory name as the full path is not always visible, and I find the exact date to be useful information. I will immediately tag all photos in digikam with the geographic location of the shots, and with any other tags which may apply. Now I will begin with my inspection and culling of all the shots. I will ‘star’ any exceptional shots, and delete the obvious duds. All others, which make up the large majority, remain status-quo.

Next, after deleting the crap and identifying the pictures I wish to publish to flickr, I open my RAW editor, Raw Therapee. When I first started shooting RAW, I tried both Raw Therapee and UFRaw. Nothing wrong with UFRaw really, I just prefered Raw Therapee, so I standardized on it. You would do well to try both if you are looking for a RAW editor.

Raw Therapee screenshot

Above is a screenshot showing a typical view of Raw Therapee. Now as for actual editing, there’s a few things I do by default, and you can see from the capture on the left side that I have saved them as a preset called “Darren’s Default”. Basically it is only two things, and increase in contrast, and an increase in saturation (Raw Therapee calls this ‘color boost’). If the photo in question has some poorly exposed areas, I will try to lighten them using the ‘Highlights and Shadow’ tool in the ‘Exposure’ tab. Generally a small tweak here will lighten up the dark areas nicely. I find you don’t want to be too heavy-handed here, as it will make the photos a bit too tacky and HDRish. Yes, I have been guilty of this a few times. For some long exposures I will also do a white balance adjust at this point, for example, if street lights are creating an ugly orange/yellow cast on the photo.

That’s it for Raw Therapee. At this point I save the picture as a 16 bit TIFF file straight into the same directory in digikam that the RAW file came from.

So digikam has a few handy editing features built right in. At this point I will open the TIFF file for (lossless) editing. I use digikam to perform any black & white conversions and crops. I also use it to straighten the horizon if I flubbed it in camera. Obviously the previous steps only apply to a few photos. At this point however, almost every photo I take gets run through an unsharp mask. If the picture requires more drastic editing at this point such as cloning or selective desaturation (very few do) I would open the TIFF file with The Gimp at this point, and save back into digikam when finished.

So now I save each TIFF file I intend to publish to flickr as an 8-bit JPG to save on space and bandwidth. Here’s a screenshot of digikam, with the main window filtered to show only JPG files ready for upload to flickr:

digikam screenshot

So that’s about it. I use FlickrUploadr which is built right into digikam to upload my pics to flickr. This is also where they get resized to 1600px on the long side. I wish I didn’t have to do this but I was finding it took way to long to upload the original sizes. Perhaps if I ever get a superfast OC3 internet connection I may reconsider this 😉

To recap, the large majority of my shots, probably about 85% of them only have the following ‘photoshopping’ done to them:

  1. increase contrast
  2. increase saturation
  3. unsharp mask

That’s it!

Permalink 3 Comments

Goodbye Edmonton Alberta

November 22, 2009 at 1:50 am (Computers, Edmonton)

Well, I’ve been running my own website, badcomputer.org for quite a while now.  I rolled my own blog code there, but found myself writing hardly ever due to the fact I was always playing around with php code.  I’ve come to the conclusion that although I like to be fairly hands on, when you spend all your time writing code you have no time for writing about life.  Couple that with the fact that I am once again moving back home to British Columbia from Edmonton, and my server will be down for who knows how long, it is a no brainer to start using a blogging service.  I still want to maintain badcomputer.org … I still have all the computer code and documentation I’ve written on my server.  Perhaps when I get settled I will look into hosting this blog on that domain.

Interests change you know.  I used to like spending hours tinkering with my computers.  Not so much anymore. I still like to look under the hood of my computers every once in a while, but I prefer to, you know, just use them.  Besides, I am spending a lot more time on my photography now.  I think that’s what I want to focus on in my spare time, and on this blog.

So: I’ve said this before at a few of the forums I hang at … Edmonton has been good to me, I love the place, but it’s time to move on.  Like the cliche goes, we can still be friends.  So long beautiful, I’ll miss you 😉

Edmonton Skyline

Permalink Leave a Comment