Re: Does MF color slides scanning in 24 bit still make sense
Hi Martin. You know t's a funny thing, but grain does also get smoothed out quite a bit by the scanning oil. The effect is obvious when you look at the image inside an air bell (usually caused by some tiny particle lifting up the film around it. That part, which has no fluid on it, is definitely grainier looking, even when it's color negative film, which has no light-scattering silver grains and thus no Callier effect. I concluded that the minute swellings of the emulsion surface where silver grains had been before being bleached out are sufficient to scatter a bit of light, much as scratches do. Another good reason to wet-mount the film.
Message: 1 Date: Mon, 18 Jan 2016 20:15:10 +0000 From: Martin Orpen <martin@idea-digital.com> To: "'colorsync-users?lists.apple.com' List" <colorsync-users@lists.apple.com> Subject: Re: Does MF color slides scanning in 24 bit still make sense Message-ID: <E30C17B4-19A8-4BAD-B9A9-CA41D2DE60D4@idea-digital.com> Content-Type: text/plain; charset=utf-8
On 18 Jan 2016, at 18:47, Paul Schilliger <pschilliger@sunrise.ch> wrote:
I did once some comparison scans when I worked with a drum scanner (a mid range one), wet and dry, and as far as super fine grain films are concerned, the gain was there when peeping at pixels, but at sight distance it was anecdotal. But the trouble of wet mounting is considerable. I thought it was better to scan dry, not to soak my precious slides into gels and solvents that might alter their life span. So, yes the process needs to be mastered and I am just supposing, but I'm expecting results that will please.
Oil mounting isn’t about reducing grain — you do that on a drum by opening the aperture manually and/or adjusting the focus point :)
You oil mount to stop light scattering when it hits the uneven (and scratched) surfaces of the film.
Reducing scatter as light passes through the film means you get more contrast and more detail.
It also gives you a very clean scan with no clean up required.
Also, oil and film cleaner don’t “soak” film — water is far more penetrating.
-- Martin Orpen Idea Digital Imaging Ltd
On 20 Jan 2016, at 01:11, Mike Strickler <info@mspgraphics.com> wrote:
Hi Martin. You know t's a funny thing, but grain does also get smoothed out quite a bit by the scanning oil. The effect is obvious when you look at the image inside an air bell (usually caused by some tiny particle lifting up the film around it. That part, which has no fluid on it, is definitely grainier looking, even when it's color negative film, which has no light-scattering silver grains and thus no Callier effect. I concluded that the minute swellings of the emulsion surface where silver grains had been before being bleached out are sufficient to scatter a bit of light, much as scratches do. Another good reason to wet-mount the film.
I’m not convinced about that theory Mike :) Repairing air bubbles is a regular task, especially around the bite marks in 5x4, but darkening the bubble area (or lightening on neg) and then removing the edge is normally sufficient for a seamless repair. I don’t recall any occasion where we’ve had to soften the image because it has been more grainy after the curve correction? Although this may be because 99% of our scanning is from colour originals? Not sure that the Callier Effect applies to the benefits of oil mounting anyhow? The drum light source is fixed. Oil mounting ensures that light isn’t scattered as it travels through the negative or transparency. Under *any* backlighting source, oil mounted negs have more depth and contrast than dry mounted negs. Regards -- Martin Orpen Idea Digital Imaging Ltd
B&W films usually have more matte emulsion surfaces than color films have. Wet mounting at least reduces the light scattering of matte emulsion surfaces. Grain in scans is aliased grain when the scan is made on CCD scanners, that noise can however be decreased in more ways, multi sampling for example helps to reduce noise of whatever kind. (Tim Vitale claims it is aliased on any scanner) What I find problematic in editing scanned B&W film is that the image contrast as seen depends on both pixel data and aliased grain formation. The sharpening and (micro) contrast tools need more subtle control than with original digital images or one has to take out the grain right away. Usually I try to keep some grain while editing. In my experience with MF and larger on film and desktop scanners the main advantage of wet mounting has been a more even focus all over the frame and a more even light transmission. In editing then the sharpening tools at least create equal changes over the image area, again with that "grain" as a second image component. With the Nikon Coolscan 8000 the carrier + wet mount film holder still showed differences in focus per scanned frame as the film holder shifts from one end to the other in the carrier and is then not kept horizontal enough to my taste. Tweaked the holder + carrier for a compromise. On the even light transmission, a warped film transmits light uneven along the scan path, I think this may show quite fast even with 35mm strips in a normal 6 frame holder I wonder whether the light source of a slide duplicator is aimed well enough at the RGB filtering of the digital camera sensor to get the best gamut etc. The sensitivity of the duplicating color film must have been aimed at the available media + light sources then, DSLRs not so far. Sony sensitivity A7R (etc?) 600 520 460 nanometer Fujichrome sensitive at 640 550 445 nanometer Fujichrome dye density at 660 550 445 nanometer Kodachrome sensitive at 650 560 450 nanometer Kodachrome dye density 640 540 445 nanometer The film numbers taken from the peaks in the documentation graphics. I also wonder whether RGB LED strips used by the guys that made the XY camera copy scanners do not make this worse if the RGB LED strips are of the narrow bandwidth type. I understand there are "RGB" LED strips that have three kinds of white LEDs aboard. For B&W however the use of green LEDs only near 520 nanometer should be ideal. Met vriendelijke groet, Ernst Dinkla Grafische Techniek Quad, piëzografie, giclée www.pigment-print.com On Wed, Jan 20, 2016 at 4:26 PM, Martin Orpen <martin@idea-digital.com> wrote:
On 20 Jan 2016, at 01:11, Mike Strickler <info@mspgraphics.com> wrote:
Hi Martin. You know t's a funny thing, but grain does also get smoothed out quite a bit by the scanning oil. The effect is obvious when you look at the image inside an air bell (usually caused by some tiny particle lifting up the film around it. That part, which has no fluid on it, is definitely grainier looking, even when it's color negative film, which has no light-scattering silver grains and thus no Callier effect. I concluded that the minute swellings of the emulsion surface where silver grains had been before being bleached out are sufficient to scatter a bit of light, much as scratches do. Another good reason to wet-mount the film.
I’m not convinced about that theory Mike :)
Repairing air bubbles is a regular task, especially around the bite marks in 5x4, but darkening the bubble area (or lightening on neg) and then removing the edge is normally sufficient for a seamless repair. I don’t recall any occasion where we’ve had to soften the image because it has been more grainy after the curve correction? Although this may be because 99% of our scanning is from colour originals?
Not sure that the Callier Effect applies to the benefits of oil mounting anyhow? The drum light source is fixed.
Oil mounting ensures that light isn’t scattered as it travels through the negative or transparency.
Under *any* backlighting source, oil mounted negs have more depth and contrast than dry mounted negs.
Regards
-- Martin Orpen Idea Digital Imaging Ltd _______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/ernst.dinkla%40gmail...
This email sent to ernst.dinkla@gmail.com
On 20 Jan 2016, at 17:04, Ernst Dinkla <info@pigment-print.com> wrote:
I wonder whether the light source of a slide duplicator is aimed well enough at the RGB filtering of the digital camera sensor to get the best gamut etc. The sensitivity of the duplicating color film must have been aimed at the available media + light sources then, DSLRs not so far.
The flash tubes in the old Bowens Illumitran have a spike at 470nm but the chart is then relatively flat all the way to 680nm. High quality Neutral Density filters flatten things out nicely — apart from the burst of heat starting at 680nm :) We’ve never had any problems with colour reproduction using Nikon sensors like the D800E & Illumitran. The reproduction issues are all about keeping the originals clean and flat :( -- Martin Orpen Idea Digital Imaging Ltd
On Jan 20, 2016, at 11:42 AM, Martin Orpen <martin@idea-digital.com> wrote:
We’ve never had any problems with colour reproduction using Nikon sensors like the D800E & Illumitran.
Indeed, color should be basically a non-issue with film input scanning, with some minor caveats. If the goal is a colorimetric reproduction of the original scene, so long as you’ve got a good profiling workflow, it should all “just work.” No combination of film chemistry and digital color capture is going to be a problem, though you could conceivably struggle with noise in some really really bad edge cases that don’t exist in the real world. As with any input profiling, the fidelity is going to depend a lot on how the (in this case original film) primaries behave and whether or not there’s a reliable way to map them to the standard observer, but the digital component isn’t going to be a factor; the colors are already being spectrally quantized by the film and that’s pretty much the end of the story. If the goal is to digitally reproduce the color of the final print (slide, whatever) that you would have gotten with traditional techniques, you’re going to need a different workflow...but, again, you shouldn’t face any significant fundamental limitations. From a simplified perspective, you just need CMY (no K) primaries matching those of the film with appropriate response curves and you’re done. Transformations to any other color space from that are straightforward. b&
I've always found that drum scans of color negative films show a lot more grain than a direct chromogenic print made from the same neg. I usually soften the grain through independent channel by channel noise reduction, but I'm curious to know how others deal with it. I"m not sure how a DSLR sees color negs, so I'm leaving the subject line as is for the moment. john castronovo
The visibility or accentuation of grain depends on the kind of lighting set-up used to make the capture. Think diffusion versus condenser enlargers of the film era and you'll have a good enough feel for the basics. Mark From: John Castronovo <jc@technicalphoto.com> To: Ernst Dinkla <info@pigment-print.com> Cc: 'colorsync-users?lists.apple.com' List <colorsync-users@lists.apple.com> Sent: Thursday, January 21, 2016 1:40 AM Subject: Re: Does MF color slides scanning in 24 bit still make sense I've always found that drum scans of color negative films show a lot more grain than a direct chromogenic print made from the same neg. I usually soften the grain through independent channel by channel noise reduction, but I'm curious to know how others deal with it. I"m not sure how a DSLR sees color negs, so I'm leaving the subject line as is for the moment. john castronovo _______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/mgsegal%40rogers.com This email sent to mgsegal@rogers.com
Yes, I’m sure that’s the case too. A drum scan uses a point light source while ‘c’ prints were made using diffusion enlargers. I want to know how others deal with this excess grain, which seems to only plague color negatives. Some purposely use larger apertures on the scanner and even put it slightly out of focus while others use noise filtering or blurring in Photoshop. Of course, those who photograph a negative on a lightbox using a DSLR on a copy stand wouldn’t have this problem. From: MARK SEGAL Sent: Thursday, January 21, 2016 9:07 AM To: John Castronovo ; Ernst Dinkla Cc: 'colorsync-users?lists.apple.com' List Subject: Re: Does MF color slides scanning in 24 bit still make sense The visibility or accentuation of grain depends on the kind of lighting set-up used to make the capture. Think diffusion versus condenser enlargers of the film era and you'll have a good enough feel for the basics. Mark -------------------------------------------------------------------------------- From: John Castronovo <jc@technicalphoto.com> To: Ernst Dinkla <info@pigment-print.com> Cc: 'colorsync-users?lists.apple.com' List <colorsync-users@lists.apple.com> Sent: Thursday, January 21, 2016 1:40 AM Subject: Re: Does MF color slides scanning in 24 bit still make sense I've always found that drum scans of color negative films show a lot more grain than a direct chromogenic print made from the same neg. I usually soften the grain through independent channel by channel noise reduction, but I'm curious to know how others deal with it. I"m not sure how a DSLR sees color negs, so I'm leaving the subject line as is for the moment.
You mean the Callier effect as mentioned earlier in one of the messages. I doubt that. Sure that effect could be introduced in scanner or slide copier illumination but true point light sources + condensors are not used in 99% of scanners but the drum scanners. The last can eliminate that effect again with aperture control which in practice is similar to CCD scanner multi sampling, the sensor resolution being below the scan sampling resolution. That is baked in hardware wise in the Epson and some Umax models but can be extended with Vuescan and Silverfast. For the Nikon scanners the multi sampling is a choice in the driver software. Even lighting in scanners is more or less impossible with true point light sources but (I imagine) a flying spot. In color negative film the elimination of the orange mask should not compromise the dye color capture, noise will appear one way or another. For color negatives discontinuous spectral lighting whether with LED or CCFL light sources can actually improve the dye capture if the spectral spikes are placed at the maximum dye densities and the sensor RGB sensitivity aimed at the combination of both spike and dye density. Intensity per channel controlled to get neutrality. Human observer has lost any meaning there. Color film should not create a Callier effect. The pepper grain effect in scans made on the Minolta scanners etc could be reduced with an extra diffusor in the light path but what actually caused it was not a Callier effect but this: https://luminous-landscape.com/fuji-pepper-grain-the-mystery-resolved/ The problem showed less with wet mount scans, as I wrote before film emulsion surfaces can play dirty tricks in scanning. Small bubbles in lenses have no influence, small bubbles in film do, I recall there was actually a commercial film that relied on it: Vesicular film. Not lost in obscurity I see: https://en.wikipedia.org/wiki/Vesicular_film In the seventees I had a Durst M610? enlarger. Normally equiped with a opal bulb tungsten lamp + 45 degr mirror + condensor lens. Fascinated by the German Veigel enlargers with their projector lamp lighting I tinkered a slide projection lighting in the Durst. Then adjust the overlap of the filament + mirrored filament correctly. Creating übergrain prints William Klein style. On cheap document paper. It had nothing to do with sharpness and all with contrast. Not usable for color either. Met vriendelijke groet, Ernst Dinkla Grafische Techniek Quad, piëzografie, giclée www.pigment-print.com On Thu, Jan 21, 2016 at 3:07 PM, MARK SEGAL <mgsegal@rogers.com> wrote:
The visibility or accentuation of grain depends on the kind of lighting set-up used to make the capture. Think diffusion versus condenser enlargers of the film era and you'll have a good enough feel for the basics.
Mark
------------------------------ *From:* John Castronovo <jc@technicalphoto.com> *To:* Ernst Dinkla <info@pigment-print.com> *Cc:* 'colorsync-users?lists.apple.com' List < colorsync-users@lists.apple.com> *Sent:* Thursday, January 21, 2016 1:40 AM *Subject:* Re: Does MF color slides scanning in 24 bit still make sense
I've always found that drum scans of color negative films show a lot more grain than a direct chromogenic print made from the same neg. I usually soften the grain through independent channel by channel noise reduction, but I'm curious to know how others deal with it. I"m not sure how a DSLR sees color negs, so I'm leaving the subject line as is for the moment.
john castronovo
_______________________________________________ Do not post admin requests to the list. They will be ignored. Colorsync-users mailing list (Colorsync-users@lists.apple.com) Help/Unsubscribe/Update your Subscription:
https://lists.apple.com/mailman/options/colorsync-users/mgsegal%40rogers.com
This email sent to mgsegal@rogers.com
participants (6)
-
Ben Goren
-
Ernst Dinkla
-
John Castronovo
-
MARK SEGAL
-
Martin Orpen
-
Mike Strickler