template_background_image
template_background_image
template_background_image
template_background_image
template_background_image
I have a beam of light to catch...
template_background_image
K.Spacey - K-PAX
Home Galleries Other arts Technique Contact Inspiration
template_background_image

Previous: Image quality< > Next: Conclusions

Index of this page

Actual size

Now looking at "actual pixels" level to an image is one thing, but depending on the size of the scan or digital file, it may actually be rather deceiving. At 4525 ppi, these scans represent a significant ENLARGEMENT compared to the original 60 x 90 cm test chart when viewed at "actual pixels" level. Assuming an average 90 ppi computer screen resolution (Photoshop by default assumes an even lower 72 ppi, my monitor runs at 92 ppi), you would need a computer screen of ca. 1.1 x 1.7 meters(!) to view the entire image at once! The scans on the previous page actually represent just a tiny crop of the entire image.

So how do the blown up images look at their "actual size" instead of "actual pixels"? Of course, the definition of what the "actual size" is, is completely arbitrary. A photographer can print his images at ANY size he wants, from a tiny post stamp size print, to an image covering a ten story high building block. However, as said before, we might arbitrarily define "actual size" to be the size of the original test chart, so a 1:1 reproduction. To show you this, I have resized each image to represent a 60 x 90 cm image size, the size of the original test chart. So in this case, you would need a computer screen of that size to view the entire image!

So here they are, the images at their "actual size":

Direct flatbed scan (Canon 9950F)

Kodak TMax 100

Resolution Resolution

Kodak Portra 160VC

Kodak Ektar 100

Resolution Resolution

Fuji Velvia 100

Sony Alpha 900

Resolution Resolution

Hey, that's looking quite different isn't it? You are probably highly surprised at how small these images are at their actual size, and you were probably expecting something bigger. As you can see from these images, even the worst images in terms of detail, the Kodak TMax 100 and Kodak Portra 160 VC shots, actually don't look that bad at the actual size of the photographed test chart. Still, Fuji Velvia 100 wins out on sheer detail, and the Sony Alpha 900 has the ultimate edge, it almost completely reproduces the original test chart's detail. Yet, Velvia 100 is getting very, very close to this objective too.

Also note the highly saturated and somewhat unnatural skin colors in the Portra and Ektar crops. Of course it would be simple to reduce color saturation in Photoshop and have them match the more natural colors of Velvia and the Alpha 900. However, these films were designed to give high saturation in especially the red part of the color spectrum, and hence reducing saturation would both defeat the purpose of these films, which lies more in landscape photography than portraiture, and cause an undesirable loss of saturation in other colors. For example, if you look back at the Color page, you will actually notice that both Portra and Ektar have a quite normal blue and yellow saturation. Thus, reducing the overall saturation of the image to combat the high red color saturation, will lead to an unacceptable loss in blue and yellow saturation.

Previous: Image quality< > Next: Conclusions

template_background_image