Sep 2010
Tram- and Bus Shelter Münchner Freiheit, Munich
Wednesday, 22 September 2010 ■ Stored in: Photography
The “Tram- and Bus Shelter Münchner Freiheit” is a rather controversial steel construction in the traffic center of Schwabing, Munich. Münchner Freiheit is one of the more important hot spots in Munich’s public transportation system, and gateway to Old Schwabing and the “Englischer Garten”.
Opened to the general public in winter 2009/10, the shelter for streetcars and busses is a 408 tons of steel roof carried by 18 pillars sitting on top of a subway station. Painted in green and white, its public perception with the locals is heavily divided.
It makes for a great location for photographers, though, and I have seen art students from Munich Universities here, sketching…



Opened to the general public in winter 2009/10, the shelter for streetcars and busses is a 408 tons of steel roof carried by 18 pillars sitting on top of a subway station. Painted in green and white, its public perception with the locals is heavily divided.
It makes for a great location for photographers, though, and I have seen art students from Munich Universities here, sketching…



What is iPad? — The iPad is for Photographers…
Note: This is one part of a multipart installment about my personal experiences with the iPad. If you are interested and haven’t read the previous parts, start here:
The iPad is for Photographers…
Using the iPad for photography is a mixed bag. Let me start with the positive aspects. If you use the iPad you start to see where digital imaging is heading eventually. It just feels natural to “touch” your photographs and manipulate them by direct interaction with the image itself, not having to use some device “in the middle” such as a mouse or a pen tablet. I can imagine that professional solutions such as the Wacom Cintiq pen displays may offer an experience similar or even superior in respect to direct interaction with the image, but those solutions are expensive and require a more complex setup (inherent to any “multiple devices” solution). And, though I haven’t used such a pen display myself, I have a hard time to believe that, e.g., running Photoshop on a classical desktop computer with such a pen display feels nearly as fast as the interaction with the iPads user interface.
The iPad allows to import JPEG as well as RAW images from your camera or SD cards using the Camera Connection Kit. The workflow is a no brainer, just connect everything and turn on your camera. Ether all photographs or a selection will be imported, with reasonable speed. iPad offers you to automatically skip images you may have imported before. The iPad’s iOS comes with a variant of the Mac OS X RAW converter creating JPEG previews, and you can browse your images as natural as it gets, with swiping gestures for switching to the next image or pinching gestures for zooming in or out. The experience of these interactions are unrivaled by any user interface I have experienced. Performance is incredibly fast, there are no delays or “hiccups”.
Apple has brilliantly designed a system that feels much faster than more powerful hardware. Many aspects of the systems design come together in the creation of this perceived performance, and it is not all UI design. One design decision obviously was to remove virtual memory management from the Mac OS X core to create iOS. That means, any application running gets to use only the amount of RAM that remains after iOS itself gets its share. With only 256 MB of RAM, that is not too much.
Though this clearly contributes to the creation of the incredible user experience, it is a trade-off that currently limits the iPads use for digital imaging. There are many Apps available that provide a rich environment for your image processing needs. However, any App can only keep a certain image size in memory. For most Apps, that’s an image size of roughly 4 megapixels. Larger images will be scaled down automatically, but don’t start to sweat — you will always work on a copy of the original file. In fact, Apps cannot directly access the original files of the images imported, they get access only through an API that makes available the image data and not the actual image file.
Though this “sandbox” approach greatly increases security (something that cannot be underestimated for a system that handles your precious photographs), it further restricts possible use cases. Any photographers workflow that involves on-set retouching of images and/or submission of the high quality, full size images via the internet to clients or stock photography agencies are currently left out in the rain. What works, though, is quick retouching and upload of web-ready images to Facebook, flickr, etc., depending on the image processing App you use. My personal recommendation would be Photogene, but there are many more good ones. Once you sync your iPad with your desktop computer, iPhoto or Aperture will import the originals as if you had connected your camera or a card reader.
That said, recently a very interesting article has been published by Rob Galbraith on his blog, that explains how to overcome many of the restrictions of Apple’s Photo app “vault” approach. Rob Galbraith is a digital pro photographer that came to my attention in 2007, while a heated public discussion about the autofocus system of Canon’s flagship DSLR EOS-1D Mark III was going on. Essentially, he started the whole discussion with an article he wrote…
In his recent article, he discusses his wireless workflow during photo shoots, which involves the iPad as an essential component. Running the app ShutterSnitch, he uses the iPad as wirelessly connected FTP server collecting — and at the same time presenting — the photos taken right during his photo sessions. Key strengths of the iPad he points out are, among others, its display quality that “…offers a degree of colour accuracy that isn't far off the best laptop screens we've tested…”, and the iPad’s battery that “…just won't quit…”.
There’s much more to his elaborate setup, I recommend to work through his eight (!) pages going deep into the details of wireless transmitters, routers, and the correct configuration of the whole setup. The bottom line of his approach is not to go directly into Apple’s Photo app through the camera connector, but to intercept the photo stream with a 3rd part app by using an alternative transmission method.
Conclusion: For me, the iPad is the perfect vault to keep an immutable copy of my photographs while traveling. The originals are kept safe, and in addition you have the possibility to review your days work and send out emails or publish on the web for quick communication. For more demanding usage scenarios, there are ways how to meet those requirements as well.
Next installment coming soon: The iPad is for Readers…
The iPad is for Photographers…
Using the iPad for photography is a mixed bag. Let me start with the positive aspects. If you use the iPad you start to see where digital imaging is heading eventually. It just feels natural to “touch” your photographs and manipulate them by direct interaction with the image itself, not having to use some device “in the middle” such as a mouse or a pen tablet. I can imagine that professional solutions such as the Wacom Cintiq pen displays may offer an experience similar or even superior in respect to direct interaction with the image, but those solutions are expensive and require a more complex setup (inherent to any “multiple devices” solution). And, though I haven’t used such a pen display myself, I have a hard time to believe that, e.g., running Photoshop on a classical desktop computer with such a pen display feels nearly as fast as the interaction with the iPads user interface.
The iPad allows to import JPEG as well as RAW images from your camera or SD cards using the Camera Connection Kit. The workflow is a no brainer, just connect everything and turn on your camera. Ether all photographs or a selection will be imported, with reasonable speed. iPad offers you to automatically skip images you may have imported before. The iPad’s iOS comes with a variant of the Mac OS X RAW converter creating JPEG previews, and you can browse your images as natural as it gets, with swiping gestures for switching to the next image or pinching gestures for zooming in or out. The experience of these interactions are unrivaled by any user interface I have experienced. Performance is incredibly fast, there are no delays or “hiccups”.
Apple has brilliantly designed a system that feels much faster than more powerful hardware. Many aspects of the systems design come together in the creation of this perceived performance, and it is not all UI design. One design decision obviously was to remove virtual memory management from the Mac OS X core to create iOS. That means, any application running gets to use only the amount of RAM that remains after iOS itself gets its share. With only 256 MB of RAM, that is not too much.
Though this clearly contributes to the creation of the incredible user experience, it is a trade-off that currently limits the iPads use for digital imaging. There are many Apps available that provide a rich environment for your image processing needs. However, any App can only keep a certain image size in memory. For most Apps, that’s an image size of roughly 4 megapixels. Larger images will be scaled down automatically, but don’t start to sweat — you will always work on a copy of the original file. In fact, Apps cannot directly access the original files of the images imported, they get access only through an API that makes available the image data and not the actual image file.
Though this “sandbox” approach greatly increases security (something that cannot be underestimated for a system that handles your precious photographs), it further restricts possible use cases. Any photographers workflow that involves on-set retouching of images and/or submission of the high quality, full size images via the internet to clients or stock photography agencies are currently left out in the rain. What works, though, is quick retouching and upload of web-ready images to Facebook, flickr, etc., depending on the image processing App you use. My personal recommendation would be Photogene, but there are many more good ones. Once you sync your iPad with your desktop computer, iPhoto or Aperture will import the originals as if you had connected your camera or a card reader.
That said, recently a very interesting article has been published by Rob Galbraith on his blog, that explains how to overcome many of the restrictions of Apple’s Photo app “vault” approach. Rob Galbraith is a digital pro photographer that came to my attention in 2007, while a heated public discussion about the autofocus system of Canon’s flagship DSLR EOS-1D Mark III was going on. Essentially, he started the whole discussion with an article he wrote…
In his recent article, he discusses his wireless workflow during photo shoots, which involves the iPad as an essential component. Running the app ShutterSnitch, he uses the iPad as wirelessly connected FTP server collecting — and at the same time presenting — the photos taken right during his photo sessions. Key strengths of the iPad he points out are, among others, its display quality that “…offers a degree of colour accuracy that isn't far off the best laptop screens we've tested…”, and the iPad’s battery that “…just won't quit…”.
There’s much more to his elaborate setup, I recommend to work through his eight (!) pages going deep into the details of wireless transmitters, routers, and the correct configuration of the whole setup. The bottom line of his approach is not to go directly into Apple’s Photo app through the camera connector, but to intercept the photo stream with a 3rd part app by using an alternative transmission method.
Conclusion: For me, the iPad is the perfect vault to keep an immutable copy of my photographs while traveling. The originals are kept safe, and in addition you have the possibility to review your days work and send out emails or publish on the web for quick communication. For more demanding usage scenarios, there are ways how to meet those requirements as well.
Next installment coming soon: The iPad is for Readers…
What is iPad? — Introduction
The public discussion about the iPad has been marred by hype, hyperbole, hate and hope — opinions ranging from ‘epic fail’ to ‘savior of the publishing industry’ could be found next to each other even in the same print or web publication. For the informed reader however, more often than not it became apparent that the respective authors had never “touched” an iPad to begin with. Lately, the discussion about iPad became more tamed — public vigilance concerning Apple first found its new target in Antennagate and nowadays seems to subside further with people expecting Apple to crumble under the advent of a flurry of iPad knock-offs powered by Android (though Google itself thinks using the current version of Android for tablets is a bit premature).
The one thing no one can deny any more: The iPad is a runaway success. Why is that?
Three months have passed since my dear wife got me an iPad. Enough time for the iPad to find its niche within my daily routine. Novelty has worn off but still I find myself using the iPad on a daily basis in areas where once my laptop has been indispensable. Reflecting on my usage patterns, I would like to try for myself to solve that enigma puzzling people since the beginning of 2010: What is iPad?
Surprising to me, I don’t use my MacBook Pro that often any more. At home, that is — at work, I still need that truck. With my personal computing requirements, however, I find only few tasks that remain predestined for running them on my laptop.
Those would be more or less the following: my digital RAW image processing workflow, re-encoding video, and writing this blog… The critical prerequisite for the first two tasks is just processing power: fast CPUs crunching numbers, and large amounts of RAM. However, writing this blog only depends on the web design software I had chosen initially, and that does not run on iPad yet. Wouldn’t that be nice, though? RapidWeaver is written using Cocoa, after all… There have been good examples of complex Mac OS X software that quickly found its way on to the iPad.
For now leaving behind the stuff iPad can’t help with yet, I would like to take a closer look at typical use cases in everyday personal computing. Surfing the web, reading, occasional gaming, writing emails and much more of the tasks I used to do on my laptop computer I now do solely with the iPad. Sometimes it took some re-thinking the ways I used to do stuff, but with a bit of flexibility you may find the benefits of iPad clearly outweigh “old school” computing.
When I started to summarize my experiences with the iPad, I thought that I might be able to write a short, comprehensive overview and be done with it. Turns out there is much to tell, even though I promised myself to keep it short. So, I decided to publish a number of short installments, each focussing on a particular usage scenario. The first one will focus on using the iPad as a tool for digital photography and will be published right after this blog entry:
The one thing no one can deny any more: The iPad is a runaway success. Why is that?
Three months have passed since my dear wife got me an iPad. Enough time for the iPad to find its niche within my daily routine. Novelty has worn off but still I find myself using the iPad on a daily basis in areas where once my laptop has been indispensable. Reflecting on my usage patterns, I would like to try for myself to solve that enigma puzzling people since the beginning of 2010: What is iPad?
Surprising to me, I don’t use my MacBook Pro that often any more. At home, that is — at work, I still need that truck. With my personal computing requirements, however, I find only few tasks that remain predestined for running them on my laptop.
Those would be more or less the following: my digital RAW image processing workflow, re-encoding video, and writing this blog… The critical prerequisite for the first two tasks is just processing power: fast CPUs crunching numbers, and large amounts of RAM. However, writing this blog only depends on the web design software I had chosen initially, and that does not run on iPad yet. Wouldn’t that be nice, though? RapidWeaver is written using Cocoa, after all… There have been good examples of complex Mac OS X software that quickly found its way on to the iPad.
For now leaving behind the stuff iPad can’t help with yet, I would like to take a closer look at typical use cases in everyday personal computing. Surfing the web, reading, occasional gaming, writing emails and much more of the tasks I used to do on my laptop computer I now do solely with the iPad. Sometimes it took some re-thinking the ways I used to do stuff, but with a bit of flexibility you may find the benefits of iPad clearly outweigh “old school” computing.
When I started to summarize my experiences with the iPad, I thought that I might be able to write a short, comprehensive overview and be done with it. Turns out there is much to tell, even though I promised myself to keep it short. So, I decided to publish a number of short installments, each focussing on a particular usage scenario. The first one will focus on using the iPad as a tool for digital photography and will be published right after this blog entry:
Lens Correction with Aperture 3
With the release of new versions of Adobe Photoshop and Lightroom, both including Adobes 6th version of their RAW engine, automated and profile-based lens correction has become “mainstream”. This is a relatively old technique, though, with several commercial offerings specializing on this task (e.g., DxO Optics Pro).
The very fist application to automate this process was PTLens, released in 2002. Its developer Tom Niemann was searching for a tool analyzing a JPEGs EXIF data, matching the photographies properties to a database of lens profiles and automatically correct for distortions. There was none, so he developed one himself. PTLens became a shareware add-on for Panorama Tools — these days it is available as stand alone program, command-line utility and as plug-in for a variety of image editing applications. PTLens has received lots of very favorable reviews over the years, and has always kept pace with more expensive commercial products.
Its price is very reasonable, the PTLens database of lens profiles is huge, and it is available as 64-bit plugin for Aperture 3. The results one can achieve with PTLens are great, and smoothly integrated as it is with Aperture, it has become a vital part of my digital editing workflow. I mentioned PTLens before, and I got a few emails asking about my experience with this tool.
One should carefully consider the application of lens correction. It is always better, of course, to use a superior lens distorting a photography as little as possible, than to rely on “tricks” after the “damage” has been done. Unfortunately, one can spend a fortune on lenses and often you just want to travel light, not carrying a wide collection of lenses each optimized for a very narrow range of use cases.
That’s where lens correction can help. My travel lens is a Nikon AF-S DX VR Zoom-Nikkor 18-200mm f/3.5-5.6G IF-ED. I am very happy with it, it has an extremely fast and silent autofocus and its VR system really works well. Image quality is very good overall — with such a wide zoom range you have to live with trade-offs, though. At its low end this lens is very sharp. However, there is noticeable distortion of the image at that range. At its long end, distortion is much better, but then this lens becomes somewhat soft. With this lens, distortion correction for images taken at its low end really improves things. The source image is sharp enough to “survive” the rather radical image transformation and the end result is still sufficiently sharp.
The following example illustrates the effects of lens correction with PTLens. Mouse over the thumbnails to view a larger image, moving back and forth between the thumbnails allows to compare the photo with and without lens correction. The left thumbnail represents the photo with lens correction. I have marked a few prominent lines in this example, taken from the distortion corrected image.
Figure 1: Mouse over to compare images
With PTLens available as plugin for Aperture, usage is as simple as one can hope for. Select one or multiple images to correct, and select the plugin menu entry from the main “Image” menu. The modal dialog of PTLens opens, already presenting a preview on the end result of the correction. Typically, your lens is automatically recognized, and the correction is applied taking into account the focal length the photo was shot at. Click-and-hold the preview image with your mouse pointer, and the image temporarily switches back to its unmodified state — so it is easy to compare the “before” and “after” images. PTLens has a number of parameters you can play around with, but I won’t go into the details…

One last word of caution on using PTLens — in fact, on using any plugin in Aperture 3… The great benefit of Aperture 3 is that any image modification in Aperture 3 does not actually alter the original image. Any change you make becomes an instruction how to change the image upon “publishing”. If you view images in Aperture 3, these change instructions are applied on the fly. If you export images, their change instructions are applied on the exported image, the original photography actually remains untouched. If you duplicate a working copy of an image, you just create a copy of the instructions set, which still refers to the unchanged original image.
However, once you process an image with any plugin, Aperture 3 writes a 16-bit TIFF with all currently existing change instructions applied. The plugin then modifies the 16-bit TIFF. You can continue to edit the image with build-in Aperture 3 tools, but they will become “change instructions” on the plugin-processed 16-bit TIFF, not on the original RAW file. Therefore, my recommendation is to apply plugins at the end of any image editing process. If you correct lens distortion, make sure to recover highlights and retouch the image before going into PTLens or other plugins. That way, you make sure the Aperture 3 tools have the biggest leverage — use them on the original RAW file, not on a 16-bit TIFF copy already missing some information otherwise available with the RAW file.
The very fist application to automate this process was PTLens, released in 2002. Its developer Tom Niemann was searching for a tool analyzing a JPEGs EXIF data, matching the photographies properties to a database of lens profiles and automatically correct for distortions. There was none, so he developed one himself. PTLens became a shareware add-on for Panorama Tools — these days it is available as stand alone program, command-line utility and as plug-in for a variety of image editing applications. PTLens has received lots of very favorable reviews over the years, and has always kept pace with more expensive commercial products.
Its price is very reasonable, the PTLens database of lens profiles is huge, and it is available as 64-bit plugin for Aperture 3. The results one can achieve with PTLens are great, and smoothly integrated as it is with Aperture, it has become a vital part of my digital editing workflow. I mentioned PTLens before, and I got a few emails asking about my experience with this tool.
One should carefully consider the application of lens correction. It is always better, of course, to use a superior lens distorting a photography as little as possible, than to rely on “tricks” after the “damage” has been done. Unfortunately, one can spend a fortune on lenses and often you just want to travel light, not carrying a wide collection of lenses each optimized for a very narrow range of use cases.
That’s where lens correction can help. My travel lens is a Nikon AF-S DX VR Zoom-Nikkor 18-200mm f/3.5-5.6G IF-ED. I am very happy with it, it has an extremely fast and silent autofocus and its VR system really works well. Image quality is very good overall — with such a wide zoom range you have to live with trade-offs, though. At its low end this lens is very sharp. However, there is noticeable distortion of the image at that range. At its long end, distortion is much better, but then this lens becomes somewhat soft. With this lens, distortion correction for images taken at its low end really improves things. The source image is sharp enough to “survive” the rather radical image transformation and the end result is still sufficiently sharp.
The following example illustrates the effects of lens correction with PTLens. Mouse over the thumbnails to view a larger image, moving back and forth between the thumbnails allows to compare the photo with and without lens correction. The left thumbnail represents the photo with lens correction. I have marked a few prominent lines in this example, taken from the distortion corrected image.
Figure 1: Mouse over to compare images
With PTLens available as plugin for Aperture, usage is as simple as one can hope for. Select one or multiple images to correct, and select the plugin menu entry from the main “Image” menu. The modal dialog of PTLens opens, already presenting a preview on the end result of the correction. Typically, your lens is automatically recognized, and the correction is applied taking into account the focal length the photo was shot at. Click-and-hold the preview image with your mouse pointer, and the image temporarily switches back to its unmodified state — so it is easy to compare the “before” and “after” images. PTLens has a number of parameters you can play around with, but I won’t go into the details…

One last word of caution on using PTLens — in fact, on using any plugin in Aperture 3… The great benefit of Aperture 3 is that any image modification in Aperture 3 does not actually alter the original image. Any change you make becomes an instruction how to change the image upon “publishing”. If you view images in Aperture 3, these change instructions are applied on the fly. If you export images, their change instructions are applied on the exported image, the original photography actually remains untouched. If you duplicate a working copy of an image, you just create a copy of the instructions set, which still refers to the unchanged original image.
However, once you process an image with any plugin, Aperture 3 writes a 16-bit TIFF with all currently existing change instructions applied. The plugin then modifies the 16-bit TIFF. You can continue to edit the image with build-in Aperture 3 tools, but they will become “change instructions” on the plugin-processed 16-bit TIFF, not on the original RAW file. Therefore, my recommendation is to apply plugins at the end of any image editing process. If you correct lens distortion, make sure to recover highlights and retouch the image before going into PTLens or other plugins. That way, you make sure the Aperture 3 tools have the biggest leverage — use them on the original RAW file, not on a 16-bit TIFF copy already missing some information otherwise available with the RAW file.
Remember the Smartbook?
In June 2009, Qualcomm and Nvidia introduced a new category of laptop computers not based on Intel CPUs, but running on ARM-derived chip designs — the same hardware used with modern Smartphones. Thus, those Netbook-style devices were aptly named “Smartbooks”…
Smartbooks were supposed to solve the main issue of any ultra-mobile computer, including Netbooks: Short battery life caused by Intel’s relatively power hungry CPUs (to be fair, looking at general efficiency Intel’s current lineup is much better today than it was during the days when Intel got stuck with the NetBurst architecture). Based on the ARM-design, Smartbooks using e.g. Qualcomm’s Snapdragon CPU were supposed to fulfill on a promise made by virtually any vendor in this industry: Availability of an always-on, Internet-connected mobile device with all-day battery life, which doesn’t weigh much and is small enough to be carried around while still being more productive than a typical Smartphone.
Sound familiar? That definition, sure — but anyone still waiting for the first Smartbook to hit the market? (I think, HP has released one officially — have yet to see it in any store or on the streets, though). Even before that market was shaped, one vendor single-handedly killed it. Today, you can buy Internet-connected, small and light devices that last all day and provide reasonably good productivity (if you don’t want to tinker with the innards of your system). That would be the iPad.
The introduction of the iPad got the whole crowd of computer vendors hurrying to catch-up with Apple. Head over heels, Smartbook designs were canned, or refitted into a tablet form-factor. At least, that’s the impression one gets. And the CEO of the guys having introduced the term “Smartbook” in the first place admits as much.
But, Google’s Chrome OS was supposed to power Smartbooks — it is still due this year. Maybe we will see a revival of this comatose device category. More likely, though, Chrome OS will power tablets. But, what happens then to the soon-to-be-released Android-based bunch of tablets? If you bought one of those, will you be happy once Chrome OS has replaced those “temporary” devices?
Smartbooks were supposed to solve the main issue of any ultra-mobile computer, including Netbooks: Short battery life caused by Intel’s relatively power hungry CPUs (to be fair, looking at general efficiency Intel’s current lineup is much better today than it was during the days when Intel got stuck with the NetBurst architecture). Based on the ARM-design, Smartbooks using e.g. Qualcomm’s Snapdragon CPU were supposed to fulfill on a promise made by virtually any vendor in this industry: Availability of an always-on, Internet-connected mobile device with all-day battery life, which doesn’t weigh much and is small enough to be carried around while still being more productive than a typical Smartphone.
Sound familiar? That definition, sure — but anyone still waiting for the first Smartbook to hit the market? (I think, HP has released one officially — have yet to see it in any store or on the streets, though). Even before that market was shaped, one vendor single-handedly killed it. Today, you can buy Internet-connected, small and light devices that last all day and provide reasonably good productivity (if you don’t want to tinker with the innards of your system). That would be the iPad.
The introduction of the iPad got the whole crowd of computer vendors hurrying to catch-up with Apple. Head over heels, Smartbook designs were canned, or refitted into a tablet form-factor. At least, that’s the impression one gets. And the CEO of the guys having introduced the term “Smartbook” in the first place admits as much.
But, Google’s Chrome OS was supposed to power Smartbooks — it is still due this year. Maybe we will see a revival of this comatose device category. More likely, though, Chrome OS will power tablets. But, what happens then to the soon-to-be-released Android-based bunch of tablets? If you bought one of those, will you be happy once Chrome OS has replaced those “temporary” devices?
Apple iOS 4.1 now available
Apple iOS 4.1 is now available, you can download it from iTunes right away.
I just installed it on my iPhone 3G, and on a quick first assessment I can confirm that the performance of the user interface increased noticeably. Though I did not experience the severe slowdowns some blog writers reported after upgrading their iPhone 3G’s to iOS 4.0, my 3G does feel “snappier” after todays update nonetheless…
I just installed it on my iPhone 3G, and on a quick first assessment I can confirm that the performance of the user interface increased noticeably. Though I did not experience the severe slowdowns some blog writers reported after upgrading their iPhone 3G’s to iOS 4.0, my 3G does feel “snappier” after todays update nonetheless…
