Regarding scanned film and digital

[ErikKaffehr]

Hi,

The top comparison with the Canyon shot is shown at 6096 PPI, while the lower one, castle facade with flowers is at resized to 3200 PPI. It's possible that a 3200 PPI drum scan would have less detail.

Best regards
Erik

Erik,

thank you for showing the samples from both scanners.
But I'd venture to say that the improved quality of the High-end scans is much more likely due to a higher quality of the drum scanner compared to the Minolta scanner, rather than to the actual scanning resolution.

If possible, it would be useful to add to your comparison a third scanning method - a 3200dpi drum scan (same resolution as the Minolta desktop scanner, but different technology).

Les

[Anders_HK]

Apart from that the red flowers in the flower pots are over saturated (missed highlight exposure???) for the film shots, I would concur of Erik's same as with what a poster stated to my tread from 2007; "The film scans have a real wonderful quality to them. While looking at these last 100% crops, I find that I'm really drawn to them and the ZD images that looked so great before look lifeless in comparison."  (reply #63) http://www.luminous-landscape.com/forum/index.php?topic=20970.msg153549#msg153549 If we overlook the red flowers, the green leaves look better in both film scans than the digital... and we are speaking of photographing a test target flower pot!!!!

While due to the pixels being sold many have interest in the resolution, there is much more. The 80MP backs (mine Leaf) are first what has impressed me compared to film in the finer gradation of colors, the more colors, the wider dynamic range enabling to process a shoulder similar to film and the more detail in shadow regions.

Film is still great, and in particular Velvia 50 which is magic for many landscapes... perhaps I am a later converter in going all digital, but I still nevertheless much respect those that stick with film. Film is great media! It has taken a long path for digital to reach similar levels in my view and experience...

Best regards,
Anders

[ErikKaffehr]

Hi Anders,

Just want to point out that the image here is from Ektar 100, a negative film, scanned on a high end drum scanner. The digital camera uses the same exposure. I would expect negative film to hold highlight detail much better than digital. So I don't think that over exposure would be the problem.

I don't really understand your statement about "test target flower pot", it's a daylight exposure at Nynäs Castle in Södermanland. The flower pot belongs to the castle.

I incidently chose this subject because I noticed a similar phenomenon on a Velvia shot a couple of years ago.

Best regards
Erik

Apart from that the red flowers in the flower pots are over saturated (missed highlight exposure???) for the film shots, I would concur of Erik's same as with what a poster stated to my tread from 2007; "The film scans have a real wonderful quality to them. While looking at these last 100% crops, I find that I'm really drawn to them and the ZD images that looked so great before look lifeless in comparison."  (reply #63) http://www.luminous-landscape.com/forum/index.php?topic=20970.msg153549#msg153549 If we overlook the red flowers, the green leaves look better in both film scans than the digital... and we are speaking of photographing a test target flower pot!!!!

While due to the pixels being sold many have interest in the resolution, there is much more. The 80MP backs (mine Leaf) are first what has impressed me compared to film in the finer gradation of colors, the more colors, the wider dynamic range enabling to process a shoulder similar to film and the more detail in shadow regions.

Film is still great, and in particular Velvia 50 which is magic for many landscapes... perhaps I am a later converter in going all digital, but I still nevertheless much respect those that stick with film. Film is great media! It has taken a long path for digital to reach similar levels in my view and experience...

Best regards,
Anders

[Anders_HK]

The digital camera uses the same exposure. I would negative film to hold highlight detail much better than digital. So I don't think that over exposure would be the problem.

Making same exposure does not necessarily result in correct. The red flowers are important since they attract the eye (and should be important for the test). Response at highlight transition or highest value with textural or tonal value may differ between the two medias (think right foot of histogram could be tuned to be more rightwards for digital than it is for film for the same exposure). Also the saturation response to red may differ and it could be that it saturated faster for the film.

Test target vs. flower pot, I would expect e.g. a landscape scene shot during best photographic conditions to show more clear the difference.

[ErikKaffehr]

Hi,

Is it your suggestion that exposure should be increased or decreased? Decreased exposure may loose shadow detail. The flower is not in shadow, it's rather highlight, that is one area where negative film is supposed to have an advantage. The reason that negative film holds highlight better is that it is compressed in the highlights. That reduces contrast in bright areas. Check this portion of the wall:

The detail in the 6096 PPI scan is much sharper but little of the structure of the wall is visible in the scan. Would you print this image at 240 PPI it would be about 1.5x1.8 m, on the screen it is shown at 3.6x4.5 m, due to screens normally having pixel pitch around 100 PPI. Try to look at the screen at 2m, which would match looking at the 240 PPI print at around 80 cm.

This image is scaled to 6096 PPI scan of the 6x7 negative. Alpha 900 top left, 6096 PPI drum scan top right, my own 3200 PPI scan bottom left.

Here is an image of the flowers at 6096 PPI actual pixels crop:


A screen dump of the full image is enclosed.

Just a side note. You are saying that exposure should be for the red flowers, but not necessarily at nominal ISO. Problem is that I need to develop film and scan to find out. Development of negative 120 films takes something like a week (sending by snail to a pro labb, namely Crimson in Stockholm). Than I need to scan to check. After a week or so the flowers may be gone ;-)


Best regards
Erik

Making same exposure does not necessarily result in correct. The red flowers are important since they attract the eye (and should be important for the test). Response at highlight transition or highest value with textural or tonal value may differ between the two medias (think right foot of histogram could be tuned to be more rightwards for digital than it is for film for the same exposure). Also the saturation response to red may differ and it could be that it saturated faster for the film.

Test target vs. flower pot, I would expect e.g. a landscape scene shot during best photographic conditions to show more clear the difference.

[Schwarzzeit]

Hi,

you have to keep in mind that the Kodak Ektar 100 film used for this test has a very saturated color response. It's excellent in bringing out subtle nuances of muted colors. But when the color of the subject itself is as saturated as the red flowers the Ektar lacks the fine differentiation of detail within the saturated tonal scale. Other less saturated color films might be better at this.

The processed scan as I sent it to Erik is a combination of the film's response to the scene, how the drum scanner read the film and finally my interpretation of the image data as a scanner operator. An optical print on RA-4 photo paper might lead to a different result. And the processed scan still has lots of room for tweaking out the desired image. You can e.g. use selective color correction and tame the saturation of the reds to bring out the detail in the flowers as you can see in the crop below which I left at eciRGBv2 color space (when I tried to convert it to sRGB some tonality in the saturated reds was lost, so you need a color managed browser to see the image as intended). My adjustments were +71% cyan, -13 magenta, +6 yelow and +12 black. Of course this affects all the reds in the image like the yellow walls. It also mutes the purple fringes of the leafs. So it could be useful to separate the flowers on another layer to fine tune these corrections.
The point is that what you see on screen is most often not the whole information of the file. A wide gamut monitor helps to look deep into the image. But it's still just one version.
The color separation of the Ektar is excellent and in my opinion at this IMAX sized negative much better than the 24 MP digital camera. The 6096 ppi crop is the pixel level of a 220 MP file.

-Dominique

[ErikKaffehr]

Hi,

The image is here for download: http://echophoto.dnsalias.net/ekr/images/PublishedTests/HESEktar01.tif

It's a 16 bit TIFF file, around 1.5 GByte.

Best regards
Erik

....
The processed scan as I sent it to Erik is a combination of the film's response to the scene, how the drum scanner read the film and finally my interpretation of the image data as a scanner operator. An optical print on RA-4 photo paper might lead to a different result. And the processed scan still has lots of room for tweaking out the desired image.

-Dominique