The third picture is the "real thing". Nikon F3 with Nikkor 50mm f/1.8 and Tri-X Pan (TX400) push processed in Xtol to ISO 800, and scanned. I found platinum (warm) toning suited this image, so I used it consistently during processing for all versions.I shot on Tri-X for many years and loved the texture it's grain could produce.
(http://www.luminous-landscape.com/forum/index.php?action=dlattach;topic=76486.0;attach=77606;image)
Without wishing to undermine your thesis (with which I agree wholeheartedly), the Tri-X image appears to my eyes to have strange sharpening artefacts in the grain. It doesn't resemble real film grain to my eyes.
I thought scanners really don't produce grain structure as they really are in the negative. That may account for this. COuld this have to do with the scanners resolution ability is not as high as the grain is small?For example my Imacon 949 scanner reproduce film grain in a very accurate way . Flatbed scanners are not capable to do this.
What do you think of shooting an evenly lit monochrome target on various film types and then having high resolution scans made of those samples to create your own grain library? You'd then use it as a layer with varying opacity and use the darken, multiply, lighten, or screen modes.
Unless I'm sorely mistaken, what I'm seeing in these crops is film grain. As you would expect, the Across is much finer than the Tri-X. Both images were sharpened of course, however the Across image was actually sharpened a tad more when I re-viewed the data. This is the film grain I'm seeing isn't it?
Both images were scanned on a Epson V700 flatbed scanner, at 2400ppi as I recall.
Hi John,
No, you are not looking at grain. At best you are looking at a 2-dimensional shadow projection of a 3-dimensional aggregate of many grains that overlap. The aggregates are not uniform, that's why it looks grainy.
To resolve true grain, a much higher resolution (approx. 8000 PPI should do a much better job) scan is required, and preferably of a thin/mono-crystal layer emulsion. What you are looking at is influenced by grain-aliasing (http://www.photoscientia.co.uk/Grain.htm) and the characteristics of a more or less diffuse light-source.
To really mimick the characteristics of film granularity, one is required to also consider that the graininess looks different in transparent versus opaque areas of the film, which will also look different and produces different (micro-)contrast depending on the type of lightsource (diffuse / parallel / collimated) with which the film is trans-illuminated. That effect is called the Callier effect (http://www.idigitalphoto.com/dictionary/Callier_effect).
Cheers,
Bart
This thread got me interested, not so much as to whether film grain can be reproduced digitally (I think it can), but rather whether scanned images will accurately show film grain. I looked through my archives and found an image of a waterfall I took with two different kinds of film (Fuji Neopan Acros 100 and Kodak Tri-X 400). Below are two 100% crops from the scans of these images. The images were taken on the same set-up with moving, re-focusing, etc. The crops look a tad different because I had not fully processed the Tri-X image and it needs some straightening.You can't see real grain from a V750 (or similar scanners) But you can see it(as i have mentioned before) from Hasselblad -Imacon scanners or even from a Nikon Super Coolscan 9000 ED scanner. And you do not need to apply any sharpen .This is possible because of the excellent optical systems and sensors that use those scanners . Of course you can see the grain ,even better maybe , from "real drum" scanners. To make an accurate comparison of the grain (digital vs analogue) one needs to compare digital prints from the scanned negative and analogue prints from that negative at high magnifications(40x50 cm for example).I am sorry for my possible mistakes in English. :)
Fuji Across 100
(http://www.cothronphotography.com/photos/i-zg9MRjc/0/X2/i-zg9MRjc-X2.jpg)
Kodak Tri-X 400
(http://www.cothronphotography.com/photos/i-9QVSx9c/0/X2/i-9QVSx9c-X2.jpg)
Unless I'm sorely mistaken, what I'm seeing in these crops is film grain. As you would expect, the Across is much finer than the Tri-X. Both images were sharpened of course, however the Across image was actually sharpened a tad more when I re-viewed the data. This is the film grain I'm seeing isn't it?
Both images were scanned on a Epson V700 flatbed scanner, at 2400ppi as I recall.
If you're interested, the Acros full image can be seen here:
http://www.flickr.com/photos/cothronphoto/6943533643/
John
You can't see real grain from a V750 (or similar scanners) But you can see it(as i have mentioned before) from Hasselblad -Imacon scanners or even from a Nikon Super Coolscan 9000 ED scanner. And you do not need to apply any sharpen .This is possible because of the excellent optical systems and sensors that use those scanners . Of course you can see the grain ,even better maybe , from "real drum" scanners. To make an accurate comparison of the grain (digital vs analogue) one needs to compare digital prints from the scanned negative and analogue prints from that negative at high magnifications(40x50 cm for example).I am sorry for my possible mistakes in English. :)
Yes, I understand that now. I have seen film grain under magnification before, and other than being much more defined I didn't see much difference in that and what I see in my scans. That being said, I'm looking at a shadow in both cases as Bart explained...albeit a much better resolved and focused shadow using a microscope. ;DI will post one for you when I have time.
If you have a 100% crop from a scan you did with your Imacon I'd love to see it, just for curiousity purposes :)
Yes, I understand that now. I have seen film grain under magnification before, and other than being much more defined I didn't see much difference in that and what I see in my scans. That being said, I'm looking at a shadow in both cases as Bart explained...albeit a much better resolved and focused shadow using a microscope. ;DThis is a portion of a 38x56 cm enlargement at 100%.Scanner Imacon 949
If you have a 100% crop from a scan you did with your Imacon I'd love to see it, just for curiousity purposes :)
Isn't the TrueGrain app available to reproduce real (not simulated) film grains? http://grubbasoftware.com/
How does this fit into your needs and "picture"?
This is a portion of a 38x56 cm enlargement at 100%.Scanner Imacon 949
Film Tri-X 400 , 135 mm developed in D-76 Dilution 1+1 .
No sharpen at all.About normal contrast.