“Bush Go Home” vs “Allah Akbar”

This just in: Many people in Austria don’t like George W. Bush!

Posters against George W. Bush

George Bush was visiting Austria yesterday so there were a few demonstrations against his presence here. I went to the biggest one in the afternoon both as a photographer as well as to state my opinion that the Republic of Austria should not spend sizeable amounts of taxpayer money to entertain a war-monger and cause inconveniences for many of its citizens.

The thing with demonstrations, though, is that they are attended by lots of people who don’t share your own particular political and moral view of the world – in my case they usually tend to be a bit more extreme. It also seems to me to be that the more extreme people are the louder they tend to shout.

I’d rather not be associated, for example, with somebody who screams “Allah Akbar!” into a megaphone, nor with someone who’d like to get George Bush shot. I also don’t share the view (greeted with applause by the crowd) of one of the speakers that George Bush is the biggest human rights violator since the Second World War. Atrocious as his record is, even for a president of the United States, it seems almost petty compared to some other guys. This view, however, would not have resonated well with a large portion of the crowd, namely the communists. I’ll take a capitalist democracy over a communist dictatorship every time of day, thank you very much!

Nonetheless, it was a very peaceful demonstration and overall a mostly positive experience. See more photos from the demo in this Flickr group!

The Dynamic Range of Digital Cameras

How much dynamic range can a typical DSLR capture? The consensus on the net (1, 2, 3) seems to be that it’s about 9 f-stops. This figure is usually arrived at by photographing gray cards or black and white targets and checking if there is still a discernible difference between the different levels. I wanted to see what level of detail there is in those darkest f-stops and, in addition, check out if there’s any difference in the shadows between shooting RAW versus JPEG and between RAW processing in 16 versus 8 bits.

To this end I did the following: I set up a tripod in front of one of my bookshelves, determined the perfect exposure and then underexposed it by an increasing number of f-stops, shooting every exposure both RAW and JPEG on my Canon EOS 10D at ISO 100. All of them were taken at night with artificial light which did not change. This is how the scene looks like with the correct exposure (click for bigger version):

The scene correctly exposed

Supposing we have 9 f-stops of dynamic range we should be able to underexpose by 8 stops and still be able to make out the scene, at least its highlights. Of course, that last f-stop needs to be amplified, otherwise it’s too dark to see on a monitor. Here’s what it looks like when shot in RAW and processed in 16 bit:

Underexposed by 8 stops and then amplified

Quite impressive! We can even go one step further and underexpose by 9 stops:

Underexposed by 9 stops and then amplified

Noise and stripes become quite apparent here. It’s more or less a matter of taste whether what you see in that photo still constitutes detail or is mostly noise. If you say it’s detail then the 10D has 10 f-stops dynamic range. If it’s noise, it only has 9. There is no agreed-upon definition, so you’ll have to decide yourself.

The next question was how JPEG compares to RAW. Using the same techniques here’s what I got for an 8 f-stop underexposure with the default JPEG settings of the camera:

Underexposed by 8 stops and then amplified

Not so hot! In the 9 f-stop underexposure there’s practically just green left, with lots of noise. JPEG becomes usable when underexposed by 7 f-stops:

Underexposed by 7 stops and then amplified

I think it’s safe to say that using the default JPEG settings we have between one and two f-stops more dynamic range when shooting RAW. I don’t know if the JPEG dynamic range can be improved by fiddling with the settings and, to be honest, I don’t care. There’s already enough stuff to take care of when shooting, and RAW has other advantages as well.

My last question was whether it makes a difference if I import a RAW file with 16 bit depth versus 8 bit. Here is the scene shot with 10 f-stops underexposure and then readjusted in Photoshop, imported with 16 bit (top) and 8 bit (bottom):

Underexposed by 10 stops and then amplified Underexposed by 10 stops and then amplified

I would have expected to see significant differences but there are none (apart from the slightly higher contrast in the 16 bit image which doesn’t reveal any detail we can’t see in the 8 bit image). The difference is even less (as is to be expected) when the underexposure is less severe.

To conclude: The Canon EOS 10D has a dynamic range of about 9 to 10 f-stops at ISO 100 when shot in RAW. In JPEG, it loses one to two stops. Importing RAW in 16 bit probably doesn’t make a difference.

As a bonus I also did this procedure with my Panasonic Lumix LX-1. The LX-1 does have a RAW format, but it’s a pain to use – the files are huge, take very long to write and are not buffered, so I only used JPEG, as I do in practice.

Surprisingly, the LX-1 performs better with JPEG compression (using the standard settings) than the 10D. Here are the shots underexposed by 7 (on the top) and 8 f-stops (on the bottom), both shot at ISO 80 (the lowest ISO setting on the LX-1):

Underexposed by 7 stops and then amplified Underexposed by 8 stops and then amplified

I’d rate the LX-1 at close to 9 f-stops dynamic range at ISO 80 with JPEG compression. I had expected much less.

The Ancestor’s Tale

I’ve just finished reading Richard Dawkins’ latest book “The Ancestor’s Tale” and I found it to be a wonderful and educating reading experience. In short, it is a book about Evolution, but with a twist. Instead of starting at the origin of life Dawkins starts with us, humans, and goes back through time, stopping whenever an evolutionary branching point, or rather, since we travel backwards, a merging point, occurs in our ancestral line. His reasoning for this reverse chronology is that the usual account, starting at the origin of life and ending with homo sapiens, makes it appear as if Evolution was somehow directed towards us, which it wasn’t – today’s bacteria are just as evolved as we are, in the biological sense.

Dawkins stops at each of those “rendezvous”, of which there are 40, and introduces us to the band of species that we meet there, usually telling a “tale” about one of them. The Fruit Fly’s Tale, which is recounted at rendezvous 26, for example, tells us how segmented bodies – which almost all animals, including us, have – form and evolved. The Salamander’s Tale, told at rendezvous 17, explains why the term “species” is not as clear-cut as one would wish it to be, even in sexually reproducing creatures.

Evidence for Evolution is presented en masse in this book, but still Dawkins is careful to point out that there are lots of things we don’t know yet about the whole story. In some cases, for example, the exact order of the rendezvous is still uncertain and awaits further molecular evidence. Molecular evidence, by the way, is a key tool in evolutionary biology these days, and Dawkins explains to us where it comes from, what it tells us, and what we can do with it, all in layman’s terms, drawing analogies to literary research.

As I said above I’ve enjoyed this book very much. It is well written, entertaining at times, and very educational, as Dawkins’ books usually are. For an in-depth introduction to Evolution you cannot go wrong with it and even if you know the basics, you’ll be astonished at all the detailed discussions you’ll find in there. I highly recommend it!

And while you’re at it, be sure to check out Zachary Moore’s Evolution 101 podcast and blog!