Posted on

Evaluating Photoshop Lightroom and ACDSee Pro Photo Manager

I tried out the trial versions of Adobe Photoshop Lightroom and ACDSee Pro Photo Manager recently.  I was particularly interested in seeing how they would work for a photography workflow, such as basic image adjustments to curves, sharpening, dodging and burning, fixing minor problems and cropping.  For more serious manipulation I can always use Photoshop or the GIMP, but I kind of like the non-destructive process on these two new products I’m trying out below.

Last time I used ACDSee there was no such thing as ACDSee Pro Photo Manager.  Of course, there was also no such thing as Photoshop Lightroom.

Problems with Photoshop Lightroom:

  • The user interface is annoying slow and unresponsive.  It’s not ridiculously bad, but it is jerky enought to annoy.  I can tolerate that all the calculations required to apply filters to an image take time and processing power, but any significant delay or slowness in simply expanding, collapsing, or resizing panes or windows is quite unnecessary.  Is it due to their use of non-native widgets (ie, skinning)?  Probably, though that doesn’t mean an interface needs to be slow – take recent versions of Firefox on Windows for example.
  • It lacks the ability to correct for barrel/pincushion distortion, and to do perspective correction.  I need to do barrel/pincushion distortion correction fairly often – for example on any photos containing buildings or straight lines.  This means it would be necessary to bring images into Photoshop a lot of the time for what should be a basic correction, even though more advanced corrections like primary colour adjustments and chromatic aberration correction, which I would probably need less often, are included.  Note that I couldn’t find this feature in ACDSee either, but based on price I had higher expectations of the Adobe product.
  • The noise removal was not too useful to me.  On a scale of 0 to 100, where 0 is no noise removal and 100 is maximum, 0 is not enough and even 1 is too much – especially for chroma.  There is no setting in between 0 and 1.  Then again, noise removal was not too good on ACDSee either, but in a different way.
  • Using Lightroom to simply browse photos on the hard drive was not very intuitive to me.  I usually prefer browsing my existing file heirarchy, and there seemed to be no way to do that – I had to ‘import’ images into ‘albums’ or similar, where albums didn’t necessarily correspond to their locations on their hard drive.  It’s an extra layer of confusion and it means I can’t be sure that I’m grabbing the right images if I go back into an Explorer window and just drag and drop.

Problems with ACDSee:

  • In any other colour mode than sRGB, everything seems about ten times as slow.  Instead of dragging a slider and seeing the colour on the image change as the slider is dragged, now you start dragging the slider and wonder why the program seems to have stopped responding for ages.  Then a few seconds after you’ve let go of the slider and are starting to click randomly on the screen to see if anything you are doing is having any effect, the colour finally changes in the preview.  This means that practically, it’s not possible to do any adjustments with a colour space other than sRGB loaded.  No problem, that’s fine – except that it’s a bit of a hassle.  If you forget and leave it loaded, and make further adjustments, it goes all slow again.  It’s also slightly disappointing that a piece of software for professionals would assume its users will all be using sRGB.  Notwithstanding the whole sRGB vs colour management debate, there are a lot of photographers that do value wider gamut colour spaces.
  • The sharpening feature is decent, but not as good as in Lightroom.  The sharpening radius is only selectable as a whole number of pixels, so sharpening using a radius of 0.6 is not possible.  Furthermore, there is a ‘threshold’ for sharpening detail, but it represents a sudden cutoff – unlike Lightroom’s ‘detail’ slider which allows a smooth transition between the sharpening applied to high contrast edges vs smoother surfaces.  As a result, sharpening an image with lots of grain cannot look nearly as good.

Of the two, it is still hard to choose.  Lightroom has better sharpening and smoothing so for quality it would win for me, but I prefer the  file selection approach of ACDSee.  Then again, the ACDSee product is not really usable when doing adjustments using colour spaces other than sRGB; while I don’t expect to be doing this often, I might want to eventually.  And yet, the ACDSee product is still cheaper.

Posted on

Number of pixels is broken as a performance measure

I’m a few years too late to this one – the boom days of digital cameras started years ago. Still, manufacturers are still promoting the number of pixels in their sensors as a measure of performance for cameras. There are multiple reasons why pixel counts are misleading both for still images (megapixels) and for video (SD vs HD).

The first is a mathematical reason. The number of pixels increases with the square of the resolution. That is, if you double the resolution of a digital image, you are increasing the number of pixels by four times.

So, while a 10 megapixel camera seems much better value than a 5 megapixel camera, it may not be the increase in resolution you expect. A 10 megapixel camera is not 2 times the resolution, it’s only 1.4 times the resolution in the horizontal or vertical direction. That means that comparing 8 megapixel with 10 megapixel cameras, you’re only considering 11 percent improvement in resolution. While this difference isn’t completely insignificant, it is usually easily overshadowed by other factors such as how much you’re going to resize, process and sharpen the hell out of it once you get it into Photoshop.

This brings me to the second reason that number of pixels doesn’t work as a performance measure.

Simply, the perceived quality of an image at a given viewing distance in a given medium has more to do with the characteristics of the original image than at which resolution it is presented. In other words, a blurry image at a high resolution is still a blurry image. The quality of the image can be affected by the quality of the original lens, any image processing that was applied including sharpening, softening, interpolation, or scaling, which often all happens inside a camera as a crucial part of its operation, and the medium in which it is presented – LCD or CRT, paper or film.

If you bought a high definition television and watch a true HD picture in it, say Becker on channel 10 (which isn’t on anymore, this was a while back), you may just be slightly disappointed at the slightly soft, grainy appearance, because this was originally shot on film. This is more noticeable to the casual viewer than the approx 2x increase in resolution.

Some people who buy expensive television sets and cameras spend lots of time searching for material to play, or things to photograph, that shows off the resolution of their equipment. People even give reviews of DVDs online recommending a movie or not based on whether the DVD encoded quality is high enough to justify their expensive TV purchases. What if it’s a good movie but blurry, grainy and slightly oversharpened?

One the other hand, I could take the argument too far and become one of those trendy people who purposefully buy cameras that will take grainy, blurry photos in order to show off the fact that I don’t care about visual quality and, supposedly, I therefore care more about taking an interesting photo. But I think once you get to the point of paying more for something of poorer quality, you are a bit of a fashion victim.

It you are considering buying a digital camera, and have a megapixel value in mind, perhaps I can convince you to rethink – dropping those pixels by 50% is only a 30% decrease in resolution, and you may find that you never notice the difference if the camera you’re using has an autofocus that’s too hard to get right, resulting in slightly blurred images, or can’t take decent pictures indoors.