The image on the left was shot using a Better Light scan back and a 55mm lens. The image on the right was shot with a Nikon D1x with a 17mm-35mm Nikon lens at 17mm. Both were shot under the same conditions. The images shown here are only intended to illustrate the conditions under which each was shot, the
actual shot comparison is made using the raw, uncompressed files.
The Better Light image was a panoramic capture at 75% resolution using a standard tone-curve; the D1x image was captured in raw format at 100%, and opened with the Camera Raw converter in Photoshop CS2 (default settings), both images were saved as 16-bit, RGB TIFFs at 300 ppi.
I copied the chair from the D1x file and pasted it into a layer within the Better Light file. I moved the D1x chair layer (at 50% opacity) within the Better Light image until I found a chair at the conference table that closely matched the size and angle of the D1x chair. In this way, both device samples would be of a similar subject, and rendered by each device with nearly the same number of raw pixels (i.e. without resizing either image). Through this procedure, a comparison of the output between the two devices could be made that would eliminate pixel count as a variable.
The Better Light image was a panoramic capture at 75% resolution using a standard tone-curve; the D1x image was captured in raw format at 100%, and opened with the Camera Raw converter in Photoshop CS2 (default settings), both images were saved as 16-bit, RGB TIFFs at 300 ppi.
I copied the chair from the D1x file and pasted it into a layer within the Better Light file. I moved the D1x chair layer (at 50% opacity) within the Better Light image until I found a chair at the conference table that closely matched the size and angle of the D1x chair. In this way, both device samples would be of a similar subject, and rendered by each device with nearly the same number of raw pixels (i.e. without resizing either image). Through this procedure, a comparison of the output between the two devices could be made that would eliminate pixel count as a variable.
Back
Bayer Array Sensor
Continue
Tri-Linear Sensor
