Pixel peeping is the practice of examining a digital photograph at the lowest level of detail to determine its quality or lack thereof. Pixel is short for picture element. Digital photographs are composed of various numbers of pixels. Early digital cameras had a few hundred thousand pixels; now most have a few million and some have even more. Usually the pixel count is expressed as the number of pixels across (horizontal) X number of pixels up (vertical). For example, a six million pixel (6 megapixels) camera might be described as 3000 X 2000.
In film days, people who wanted to peer at the detail of photographs in a serious way used a loupe (magnifier) to examine negatives. Everybody knew that most prints had much less detail than was actually found in the negative; something that anyone who has ever ordered enlargements from a print will know. Looking at negatives was a serious business and required skill and patience.
Looking at a digital photograph up close is easy with a decent sized monitor and image editing software.
Now that we can pixel peep we may do so and find that our photographs do not impress up up close. When people look at pixels they are seeing their images as bunches of square blocks, each one a single colour; something like building an image with pieces of lego. It looks fine from a distance but not so great up close. What this implies is that we need to have an idea what things look like up close before we judge; we need to realize that there is no such thing as a diagonal line or a perfect circle in a world of rectangular pixels.
Having much to do and no wish to do it tonight I took some time to do some pixel peeping. I wanted to see how well the Canon EOS 7D digital single lens reflex camera did at reproducing a subject. Some of my 7D pictures have given me pause and lots of people on the internet are expressing (to say it mildly) their concerns about the 7D which is new 18 megapixel camera.
I decided to do my best to test things under favourable circumstances: a good lens (Canon 300mm f2.8 IS), tripod, mirror locked up and timer to allow the camera's vibration to settle down before the picture was taken.
For a target, I taped a stamp from an envelope on some books about 12 feet from the camera. I focused with the centre point using spot auto focus (this is when the camera uses a smaller focus point).
I took a series of RAW shots at ISO's from 100 to 12800. I processed the RAW shots in Adobe Camera Raw 5.6. I used default settings except for adding two thirds of a stop of exposure. I did not use the clarity control.
In Photoshop CS4 I added annotations but did not do any sharpening.
After the tests, I realized I could have picked a better stamp. I wish I had used one with fine engraving: lots of thin lines and tiny text. Maybe I will if I want to do more pixel peeping in the future.
I was impressed with the 7D, especially at lower ISO's. I think it has the potential to take some very good pictures given good lenses and steady hands (especially with longer focal lengths). It has nearly twice the resolution of my other cropped sensor camera, the 40D. Does this mean that it magnifies every bit of camera shake and vibration by almost a factor of two? If it does then it means that it may require faster shutter speeds.
I took the stamp and scanned it at 600 dots per inch on a Canon 8800F scanner. It was obvious that the scanned version had much more detail than the 7D shots. The difference in effective resolution was significant but not enormous (468 pixels acros the stamp on the scanner and 383 for the camera).
I posted full size annotated test shots in a gallery on my website.
Was this a useful exercise? For the purist, with an optical bench in her basement, probably not. There are too many variables that were not controlled. For example, the stamp was illuminated by fluorescent lights that flicker; not every shot is going to have the same lighting.
Was this a useful exercise for me? Yes, it was. It is one of the things that has convinced me to keep the 7D. One of the other things that convinced me of that was my experience shooting a hockey game in a dark arena with the camera.
Was it fun to pixel peep? Not really. It made me realize how much work and how tedious it would be to do this in a serious fashion. I am glad that there are lots of people who are willing to put in the effort to pixel peep and test cameras for the common good. There are lots of great sites on the internet where I go to read about their results: DPReview, Rob Galbraith, Photography On The Net, etc.
Was this a useful exercise? For the purist, with an optical bench in her basement, probably not. There are too many variables that were not controlled. For example, the stamp was illuminated by fluorescent lights that flicker; not every shot is going to have the same lighting.
Was this a useful exercise for me? Yes, it was. It is one of the things that has convinced me to keep the 7D. One of the other things that convinced me of that was my experience shooting a hockey game in a dark arena with the camera.
Was it fun to pixel peep? Not really. It made me realize how much work and how tedious it would be to do this in a serious fashion. I am glad that there are lots of people who are willing to put in the effort to pixel peep and test cameras for the common good. There are lots of great sites on the internet where I go to read about their results: DPReview, Rob Galbraith, Photography On The Net, etc.
No comments:
Post a Comment