Apple brought out Aperture 2.0. A wonderful update to their photo management and editing app. I have been playing with the demo quite a bit and it does address most of the major problems it had in 1.5. I'll write about it more in the future especially about where I'd hope that the Lightroom folks learn from Aperture 2.0 - and there are quite a few areas. So this is NOT a review and it only addresses a tiny point. It might not be a problem for your camera also, so I suggest you check it out for yourself first. For now, I'll focus on image quality, which for me is of paramount importance. Apple's RAW library has had major moiré problems in the past, where the RAW interpreter would create small little mazes. I am happy to say that most of the worst examples of this have been solved. They were actually already solved in 10.5 Leopard BTW. However, there are some problems remaining. I will compare the quality of the RAW converter in Aperture with that in Lightroom. Here is an image from a frozen lake in the Colorado Mountains taken using my venerable (and very light!) D50.
The ice on the lake has lots of wonderful structure that you can see in this 100% crop from the Lightroom image:
Now compare this with a 100% crop from an Aperture export of the same image:
What is going on? I see anomalously colored areas, little mazes and a generally muddy image. This is not salvageable with any moiré reduction, sharpening or edge sharpening in Aperture. It just becomes more messy. To illustrate further, here is a 300% zoom of the image in Lightroom compared with Aperture:
Lightroom conversion:
You see all the detail and bumps in the ice very well.
Aperture conversion:
Ugh! I admit you have to be a pixel peeper for this, but that is terrible. Of course on a webimage, you will never see it as illustrated by the Aperture image below (compare to the first image on top of this post):
Looks fine at this scale of course. However, you will certainly see this in larger prints (i.e. 8x12 and larger). In all this I haven't even mentioned that Nikon's software conversion of the image is even slightly better than Lightroom/ACR! Also, note that I looked at only one image. I see it in other images too though also from different cameras, but your style of photography or your specific camera might make this less of an issue.
Of course, when I edit this picture further, I would probably brighten the foreground a little and crop it differently. Here is the result starting from the Lightroom image above.:
Friday, February 15, 2008
Thursday, February 14, 2008
Awesome tip at Luminous landscape
As I am pretty sick today, and cruising some photography fora, I was reminded of a tip I read about a while ago. The tip consists of making selective zonal adjustments to make elements in the image pop. The article shows some amazing examples, such as the steaming trees which can be made from a fine image into an amazing one by a few very simple adjustments. Take a look at it and I am sure you'll think of some places to use it. For me, it caused me to revisit an older Rocky Mountain National Park photo where you see a tree backlit by the rising sun. I remember the whole tree having sort of a shrouded halo, which never materialized in the actual capture. On the left is the Lightroom original and on the left is the best Lightroom rendering obtained by increasing fill light to 32, the recovery to 10, and reducing the blacks to 3, and I cropped a little. No changes to white balance, clarity, vibrance or saturation. This made it a lot better than the original already, but still lacks pop.
Sort of OK but pretty boring, which is why I never really looked at it I guess. Only now, scouting for images that might benefit from this treatment did I come accross it. I took this image and in Photoshop, I selectively brightened the glare in the tree, while keeping the tree dark, darkened the sky selectively with another curves adjustment layer and played a little more with the foreground to make some sparkles in the snow and give some detail in the trees.
This is the result:
Much better. Note that this sort of treatment is very similar to how the program Lightzone operates, which can do zone masks for all its tools too. That looks like a very interesting program. I should try it sometimes.
Sort of OK but pretty boring, which is why I never really looked at it I guess. Only now, scouting for images that might benefit from this treatment did I come accross it. I took this image and in Photoshop, I selectively brightened the glare in the tree, while keeping the tree dark, darkened the sky selectively with another curves adjustment layer and played a little more with the foreground to make some sparkles in the snow and give some detail in the trees.
This is the result:
Much better. Note that this sort of treatment is very similar to how the program Lightzone operates, which can do zone masks for all its tools too. That looks like a very interesting program. I should try it sometimes.
Sunday, February 10, 2008
Pete Caluori
A good friend, Pete Caluori just put up his website after working and mulling over it for ages. Pete works almost completely in film and uses amazing cameras that capture on film with sizes over 8x10 and even panoramic images. He also prints using resurrected alternative processes with Palladium salts or the Van Dyke process. His photographs are quite amazing and their radiance is something that cannot be readily expressed in the limited palette of the web. Check his website out over here.
Saturday, February 9, 2008
Redid sunrise panorama
I just redid a sunrise 360 degrees fully immersive panorama I made at Emerald Lake. I combined multiple sets sets of exposures for this result. The old version had the sunlit area blown out.
I turned this into a nice QTVR too, but I don;t have a good place to host that yet. I'll post it when I do. When I took these photos, I hiked up from the trailhead far before sunrise. Always fun, especially when it is snowy and icy ;-). For these 360 degree panoramas, I use the awesome 10.5 mm fisheye and a homebuilt device that can be seen here:
I stitch using the awesome open source/freeware hugin using enblend.
I turned this into a nice QTVR too, but I don;t have a good place to host that yet. I'll post it when I do. When I took these photos, I hiked up from the trailhead far before sunrise. Always fun, especially when it is snowy and icy ;-). For these 360 degree panoramas, I use the awesome 10.5 mm fisheye and a homebuilt device that can be seen here:
I stitch using the awesome open source/freeware hugin using enblend.
Wednesday, February 6, 2008
Exhibition
I hung some of my Photographs, many of them in frames I made myself, up at Standley Lake Library in Arvada. They will be there the whole month of February. This is a wonderful library close to a nice nature preserve - Standley Lake. Here Bald Eagles nest every year and you can find many photographers and bird watchers trying to catch a glimp of them. I don't own the requisite long lenses, so I have never tried. I hung the images in this gallery and a few more.
Monday, February 4, 2008
CSS frames around your online images
I am preparing a post (check back later) on how to matte and frame your pictures for maximum impact and the realized I might just as well explain how to create the fake mattes and frames I put around some images on this blog. The link to my smugmug page for example on the right has one of those. They are very easy to make. So here goes.
First you need to call your image as follows for example:
This results in your image with a thin white line, creating the reveal you would get from a beveled mat cut:
Than you throw a span tag around it:
The 10 pixels of black make the frame and the 25 point padding makes the grey mat. The color of which is determined by the background tag.
The complete code for this would be:
As you can see, I usually put a tag around the result and enclose it in a link to the website where the image can be seen larger. You can probably do the same thing with nested spans, but for some reason, browsers put a grey stripe at the bottom if you do that.
First you need to call your image as follows for example:
This results in your image with a thin white line, creating the reveal you would get from a beveled mat cut:
Than you throw a span tag around it:
The 10 pixels of black make the frame and the 25 point padding makes the grey mat. The color of which is determined by the background tag.
The complete code for this would be:
As you can see, I usually put a
Subscribe to:
Posts (Atom)