Fare thee well, 2010

Colorize the Winter (FinePix S5Pro, 1/13s @ ISO 200; f/8, 153 mm (in 35mm)

Last year on December 31st, Shuwen and I were taking a walk with Toni, near Haiming in the floodplain forest of the Innspitz area (at the confluence of the rivers Salzach and Inn). The above photo was made on that day. A lot of things have changed since then. :)

2010 was a busy year, and personally, I'm glad that it's over. (should you ever go through a fiancee visa process and move from Germany to America, you will know why:-)

I just recently managed to publish my personal "Best of 2009" album on PicasaWeb, and I hope it won't take as long to publish the 2010 version of that. I also want to make a selection of only 10 best shots of 2010 and publish it here on the blog to maybe take part in Jim Goldstein's blog project. Let's see if I'm able to do that. (only 10!!! yikes!!!)

Have a great and lucky 2011 everyone!

A/B - Welcome to the Jungle

I recently updated the About Me page of my blog, and the "Philosophy" section explains my take on photography as an art form, and post processing. Quite a while ago I said that I do not show "alternative" versions of my photos, but only the "finished product".

At the moment, my personal mission is to advocate the freedom of post processing. I more than likely repeat myself like a broken record, but if I can use a roll of uber-saturated Velvia film and a softening beauty filter to a traditional film camera, why shouldn't I do the same in digital, and more, with post processing? After all, what I want is to show beautiful pictures. As a person said in a comment to one of my photos in Picasa Web Albums: she recognized my "style" because I'm showing an elevated version of nature that is slightly over the top. Supernatural, in a way. :-)

So today, I start a new category of posts: A/B standing for "after and before". I show the final image, and when you hover your mouse over it, you'll see the plain data that came out of the camera. Since this is done with a script you'll have to visit the original post in my blog to see it work (so please clickety-click on that "Read more" button.)

Welcome to the Jungle (NIKON D700, 1/100s @ ISO 1250; f/8, 65 mm (in 35mm)
Mouseover will show the cropped, but otherwise unedited image with the "Camera Standard v2" profile of ACR/LR applied to the raw data.

I picked this photo because it has received a good amount of praise when I originally showed it in my Picasa Web Album (link goes to the photo there.)

And from my past experience when I took part in similar projects I know that someone, somewhere will be saying "...but I prefer the original!" - that is to be expected and well, so be it. It's all about taste, isn't it? I decided to show the photo in it's final version, and as I'm writing this, more than 300 people looked at it in my web album, and 13 of them could actually be bothered to click on the "Like" button. It can't be all that wrong. :)

I decided to NOT show screenshots of the processing steps in Lightroom. That would be boring. Instead, I just give you a short summary of what I did so it should be more inspiring, and less teaching. :) So what has happened in post processing?

The most dramatic change comes from a lot of negative Clarity. It gives the image this hazy, glowing, dreamy look. The loss of contrast has been compensated by a boost of the highlights (a very good trick with negative clarity) and by switching to the "Camera Vivid" profile for raw conversion. I countered the effect on the tree's trunk and branches (and also in the foreground) with some local adjustments so that the tree's trunk and branches would stand out. I reduced overall contrast and used fill light to compensate for the darker shadows of the Camera Vivid profile. Shifted the yellow and green hues a bit to keep these two colors farther apart from each other. Split toning for the shadows with some magenta tone. That's it.

Off topic notes

A couple of more general notes regarding the blog, photography-wise off topic:

• The blog is back to it's two-pane layout with the photos in the posts being no wider than 640 pixels. The 1000 pixels wide single pane layout would stretch the text pretty far across the screen, making it harder to read; and the photos were really kinda big in the feed, or in Google Buzz. :P
• As you've seen in the last few posts I've started using a mouseover script to show the final image vs the original, unedited data (as far as that is possible with raw data, anyway.) Because it's not possible to use that feature in the feed, I will make use of the jump break feature of Blogger for these posts. They only recently added a feature that allows the posts to appear in the feed right until the jump break. The feed will still contain all other posts in full.
• I find myself using, reading, linking Wikipedia almost every day. Time to support their cause with a donation and a banner add in the right pane of the blog. We ALL use Wikipedia very often. Make a little donation, no matter how annoying you found the "Click here for a very important message..." box in the past couple of weeks. So come on!

Vignetting - defect or effect?

Vignetting is defined as light falloff at the corners of an image. In other words, the corners of an image are darker than the middle of the frame. This is a "defect" that surfaces for example when you're using a lens "open", and can be more or less pronounced. Common lens tests include testing for vignetting, and if a lens has vignetting, the world is coming to an end. At least that's what I'm tempted to think when I read some tech geeks' statements in photography forums. :) (and in reality, it can be fixed quite easily.)

It is also known by now that lenses that used to work fine on film do show more vignetting on digital sensors. The reason for that might be that unlike film, the sensor is not really a plane, but the photons have to "fall" into "light buckets" into which they're directed by micro lenses. Once more, an excellent tutorial at Cambridge in Colour illustrates it nicely: Understanding Digital Camera Sensors.

If you stop the lens down a little bit, the vignetting will become less and less visible. When you're using a full frame lens on a camera with a crop sensor, you'll probably won't notice much of it, anyway. And modern lenses are built in a way that they will show less vignetting even when used wide open. It's quite an amazing amount of engineering that went into that!

At the same time, a number of fixes for vignetting do exist of course: the cameras are equipped with JPEG engines that automatically correct for vignetting when you're storing your photos as JPEG. Raw converters of all brands have features and/or lens profiles to automatically correct vignetting or manually adjust it.

So at first it sounds a little bit surprising that one would want to ADD vignetting to their photos, but of course vignetting can be used as a creative effect - deliberately making the corners of an image darker to draw more attention to the center of the frame. Lightroom's original Lens Correction vignetting tool always had the option to darken the corners instead of brighten them, too. But the effect would apply to the whole, uncropped frame - it was really meant as a lens correction feature originally. Only in version 3 a new vignetting tool was added to the Effects section of the Develop module. It allows to apply vignetting after cropping:


Suspicion (4+1) (NIKON D700 with 70-300mm lens, 1/250s @ ISO 1600; f/8, 300 mm). Original.

In the Googleverse, Picasa still doesn't have a feature like that. The only thing that comes close is the hidden "Radial Tint" effect. You can enable it by pressing and holding the CTRL key while selecting the "Graduated Tint" from the Effects menu - by carefully selecting a color you can mimick a vignette effect to a certain degree, but it cannot really be used to darken the corners of an image. Google's online editor Picnik does have a Vignette effect, but it's somewhat square-ish and the roundness can not be configured - therefore, it's usability for accenting the center of the frame is somewhat limited. In addition to that, both work only with the 8bit JPEG data - posterization is much more likely to happen here.

Unfortunately, sometimes people start to use vignetting too regularly, or with an amount that can only be called absurd (the question "oh, is your lens broken?" comes to my mind if I see something like that). If the composition of your image doesn't clearly point out your main subject, adding a little vignetting might help with that - but it's not a cure-all, for heaven's sake. But even media sites like "The Big Picture" sometimes feature a whole series of photos that have been mangled with vignetting to no end (just scroll thru the start page and it will jump right at you; is it really necessary to add this kind of "drama" to documentary photos? I don't think so.)

I do use vignetting myself. Heavily for some photos even (see below.) It's a creative effect that can more or less add to the overall impression of a photo. But if the whole style of a series of photos only consists of vignetting it's not creative. It's boring, and annoying.


Hagengebirge (NIKON D700 with 70-300mm lens, 1/200s @ ISO 200; f/8, 300 mm). Original.

So please... let the effect stand out from the few photos where it really makes sense and adds to the image. In other words: keep the overall amount of photos with vignetting at a sane level. Subtle usage of vignetting is probably still the best. I think it's somewhat similar to the HDR technique: only if you cannot notice immediately that it has been used, it's a winner.

Using Android as a GPS logger

Yesterday I tried my Android phone as a GPS logger for the first time. There's this nice little app from Google called "My tracks" and it can export the track it logs as GPX files, so when we made a little hike in Daley Ranch, Escondido (blogpost location shows the parking lot/staging area) I gave it a try. On the "Bobcat trail" we walked thru a beautiful little oak grove with this wonderful wonderful huge old oak:


Old Oak (NIKON D700, 1/15s @ ISO 500; f/11, 24 mm (in 35mm)

So what is it good for, anyway? A GPS logger tracks your location and logs it. Like, it puts down your location every 5 minutes or every 5 meters that you move, whichever happens first. You can use this track later to "geotag" your photos. The location information is stored in the photo itself (as coordinates). If you upload the photo to Flickr or PicasaWeb, these services will find the geolocation stored inside the file and put the photo on a map. You know where exactly the photo was taken, and others can see it too.

That's not exactly new, and a lot of devices exist for that purpose. You can buy a stand-alone GPS logger that just starts logging whenever you turn it on. You can get a logger with a display that will also show you the coordinates (like, when you're hiking, or if you're into geocaching). There's also devices specifically for the camera - you attach them directly one or the other way and they will write the coordinates to the photo as soon as you release the shutter (those are the most expensive ones I think, but the most convenient ones for this usage of course).

I knew that using the phone for that purpose could be difficult - because the GPS is somewhat heavy on the phone's battery. But to my surprise, logging the 2.5 hour hike with My Tracks on the Android drained only about 10% of the battery (which was fully charged when we left home). I think that's partly because of Android 2.2. And I've not been using the phone for anything else while hiking - it was in my backpack, logging. (And making the occasional sound when an email arrived.)

At home, I exported the recorded track to the phone's SD card as a GPX file, connected the phone to the computer via USB to download the track, and plugged the camera's memory card into the card reader. Using Nikon's free "View NX 2" application I browsed the photos, selected all of them, and used the "Log matching" feature to select the GPX track log and add the coordinates to the photos. That was easy! Too easy, of course. :-)

Because, since I did this for the first time, I ran into some (small) obstacles. First there's the critical issue of time. That's two parted: the timestamp in the GPS log is in GMT (Greenwich Mean Time), so you must select the correct time zone. In addition to that, the time in your camera must be correct, too. Turns out that mine wasn't - I forgot to adjust it after daylight saving time ended! Duh.

This resulted in the majority of photos not being correctly geo-tagged. And while tracking down my mistake, a bug (?) in Lightroom caused additional confusion: when you click on the small little arrow next to the coordinates EXIF data, it will open the browser with Google Maps to show you the location - but it does not pass the coordinates correctly! Maybe it's lacking precision, I don't know.

I have to praise Jeffrey Friedl once more: using his "GPS support" Lightroom tool works as expected and shows the correct coordinates for the photos. For the Lightroom users, Jeffrey's tool is much more versatile than using the two step-method (first matching the coordinates with View NX, then importing the photos into Lightroom) that I described first, and well worth a donation (like most of his Lightroom goodies). Most importantly, Jeffrey's tool has a verbose error log that helped me to find the mistakes that I made - there's no such thing in View NX.

EDIT/UPDATE: Lightroom 4 has built-in support for geotagging files from GPX files (I still find Jeffrey's plugin more versatile though), making it unnecessary to fiddle with ViewNX or the likes, like I did.

PS: I hope those shrubs underneath the beautiful old oak were not poison-oak. So far, I'm still fine... :-P

Auto-ISO lacks intelligence

Beginning with the S5pro, I started to always use Auto-ISO when using the camera hand-held. And with the D700, it's a dream. And other than in the linked older post, I nowadays let the D700 utilize the ISO range up to a maximum of 6400. If you can expose to the right, then even high ISO is not a problem (Jao has a nice technical explanation on his blog about that.)

However... I think that camera's are still not intelligent and configurable enough to fully utilize the benefits of the Auto ISO feature.

When I'm using my camera with a zoom lens, the best I can do with Auto ISO is to operate the camera manually. I might need an exposure time of 1/160s at the long end (135mm) of my Tamron all round lens, but 1/40s is easily enough at the short end (24mm) of that lens. And would it be a VR lens, I'd dare to go to 1/40s at the long end, and 1/10s at the short end (and practice the methods of getting a steady shot even in difficult situations.)

[Those exposure times are based on the "general rule" that you get a steady shot at 1/(focal-length * crop-factor) seconds, with a little bit extra safety added.]

But I really think that Auto ISO should cover that. It's longest time should be configurable as a RANGE depending on the zoom range of the lens one's using. If you're an Aperture priority person like me (and care about getting a sharp image... duh) then Auto ISO today really works satisfactory only with prime lenses - I'm only talking about Aperture priority mode. With a zoom lens, you need different "shortest time" configurations depending on the focal length you're actually using.

So to get to the "sweet spot" of getting a sharp image and not using a higher ISO than necessary, there's only one option with Auto ISO: use fully manual control. And I think that's lame. Camera's are small computers today, and they don't offer something as trivial as taking the zoom range of a lens into account for a great feature like auto ISO.

Thinking of that, Nikon's knobs and dials (at least on the pro and semipro bodies) could be utilized much better to configure Auto ISO. The front dial could be used to configure the longest time, the rear dial the highest ISO. Something like that. And while I'm at it: it's STILL not possible to override the configured "high" limit of Auto ISO manually either (and since I'm already in rant mode: I would like to have a bigger exposure stepping for bracketing as well, please. The D700 is limited to +/- 1EV - how lame is that?!)

Too bad that DSLRs are not exactly the target of nerds that would hack the firmware to enhance the camera's features. "Yup, I jailbroke my D700 with cy4nsn0w and now use Auto-ISO+" - something like that. :-)

What film teaches us

Before you read on - I admit it, I never shot film with an SLR. I used to have a compact camera with me during vacations or something, filling one 36 roll of slide film with some snapshots, but other than that? Nope. So why do I dare to write about what film teaches us? (with "us" I mean all the digital photographers out there. Like me.)

I've more or less written about this topic here before, and this is yet another post that is based solely in my opinion and point of view. I wrote about developing your own photos from raw data, about film vs digital, about the charm of Hipstamatic, etc. etc. - and if you put it all together you can't help but notice:

Film has something that digital is missing.


Autumnal Forest (out of camera) (NIKON D700, 1/80s @ ISO 3600; f/11, 35 mm (in 35mm)

And that's character. A good photo could be made of more than good composition, good timing and all the artistic qualities that you may or may not be able to learn (and I think that, with the right amount of obsession/dedication everything can be learned.)

Sometimes, what really makes a good photo is also the appearance. And in todays mostly digital photography, there's nothing that teaches us so nicely about appearance than traditional film does. Every type of film has it's own characteristic. That could be the color hues, the contrast, the grain. It could also be the processing, one of the best known (and digitally emulated) effect at the moment surely must be cross processing.

Unfortunately, with the advent of digital, a lot of hobby photographers have turned their cameras into "most neutral reality seeking devices" - why? Because they can. Aiming at the perfect white balance, the most natural reproduction of color and all that, while photography itself cannot be an accurate reproduction of reality (I repeat myself.) And very often, when you "dare" to show a photo where you also worked on the appearance in an online photo community, this type of person will come crawling out from under a pile of theory and technical specs to state "I've been to the forest, and that green is just not natural." I wonder what this type of persons would say when you told them that the green is just like that when you shoot Velvia film.

Why bother about the color accuracy and whatnot when it apparently has so little do to with the overall impact of a photo? (and yes, of course you want the skin color of a person in a portrait to be somewhat neutral and realistic) Indeed at some point some digital photographers actually buy a film camera that they can use with their lenses and shoot some rolls of film. Just to experience the character of film themselves, to experience the "it's out of your control" feeling (I described that in the Hipstamatic article I linked above already.) Hey, I'm active in some photo forums online, I've seen it happening! :)

The funny thing is what happens with the negatives or slides after that: they're being scanned to make them available in digital on the computer! Maybe for further post processing. ;)

I often find myself struggling with the endless possibilities of post processing. But when I think about film and appearance, I realize that it is actually a good thing to have these endless possibilities (once you free your mind from striving for the perfect reproduction of "reality" of course). I can create just the appearance that I want for my photos. After my Hipstamatic excursion (which has ended also because I'm not using an iPhone anymore), I can safely skip trying a film camera and film - and just enjoy the artistic freedom of digital processing. And enjoy it without worrying. Don't listen to the haters. :)


Autumnal Forest (final) (NIKON D700, 1/80s @ ISO 3600; f/11, 35 mm (in 35mm)