Neg scans (grain with FP4+)

[Eric Rose]

I found Nikon scanners to be almost unusable for B&W negs. With the exception of XP2 of course. I can get scans from my V750 Pro that beat the Nikon scanner hands down. Grain, grit, whatever you want to call it was always the issue when using the Nikon scanners. Seemed to really accentuate it for some reason.

 

[markjwyatt]

At higher scanning resolutions like > 3200 for 35mm TriX I find grain is more pronounced (depending on the image, large sky areas vs cluttered close focus). If that's combined with heavy sharpening in the scanning process its worse. Judging from the post looks like heavy sharpening of the grain clumps. I couldn't scan something that poorly if I tried...which suggests some problem with their process.

That is what I am thinking, and with Ted Baker's observations this makes sense. I just shot a couple of rolls of FP4 at ISO 125, had them developed and scanned to 4000 dpi by The Darkroom. I am very happy with some of the images, but was a little shocked by some of the grain. I still have not gotten the negs back yet, so maybe I under exposed some. Here are some samples I have uploaded. https://www.flickr.com/search/?user_id=157638541@N07&view_all=1&text=Ilford FP4

I did very little post processing, but I did play with the dynamic contrast (In ON1 2018). Basically, I dropped the lowest scale quite a bit to de-emphasize the grain structure (but perhaps also killed some of my micro-contrast in Zeiss and other lenses), then bumped up the medium and large scale contrast for the overall image. I am considering standardizing on Ilford films given that Acros is practically gone (I have a few rolls of 35mm still), but may also look at Kodak Tmax.

The shot "Tree on a hill" really shows the grain. https://www.flickr.com/photos/157638541@N07/46038088421/in/dateposted-family/
"hillscape" does also, but here the grain is nice and works well with the scene. https://www.flickr.com/photos/157638541@N07/32167865318/in/dateposted-family/

 

[Mainecoonmaniac]

yes agreed but if sharpening is on at time of scan real bad things can happen... Sharpening is best done in PS in the Luminosity channel.

So you work in LAB instead of RGB?

 

[GLS]

I can't say I've noticed excessive grain when digitising FP4+, but then I use a DSLR rather than a scanner. The grain is there, but quite subtle; certainly nothing like as bad as some of the examples posted in this thread. I can post some examples if people wish.

 

[markjwyatt]

I can't say I've noticed excessive grain when digitising FP4+, but then I use a DSLR rather than a scanner. The grain is there, but quite subtle; certainly nothing like as bad as some of the examples posted in this thread. I can post some examples if people wish.

I am hoping that will help. I am setting up to do that with my mirrorless (Fuji XT-2), but so far am still playing with color slides. Negatives should be next. I need to figure out the easiest quality way to do the inversion without having to go to Photoshop.

I would like to see a few examples if you do not mind.

 

[GLS]

I would like to see a few examples if you do not mind.

Sure. Here a few from one roll, shot at box speed and developed in DD-X. There are the full frames and then 100% crop portions from each. The 100% crops only have minimal sharpening applied (as they were digitised using RAW, so require it).

 

[markjwyatt]

Sure. Here a few from one roll, shot at box speed and developed in DD-X. There are the full frames and then 100% crop portions from each. The 100% crops only have minimal sharpening applied (as they were digitised using RAW, so require it).

...

The grains structure is much more subtle. Thanks. Are these medium format? Mine are 35mm.

 

[GLS]

Yes, sorry forgot to mention. Shot with a Hasselblad.

 

[markjwyatt]

Yes, sorry forgot to mention. Shot with a Hasselblad.

Definitely smaller relative grain size (same actual grain size). May be consistent with what I see in that case, not sure.

 

[GLS]

Your scans are done by a lab, correct? If so, do you know how they apply sharpening, and how much? If an inappropriate and/or overly aggressive sharpening algorithm is used it can artificially exaggerate the grain.

 

[markjwyatt]

Your scans are done by a lab, correct? If so, do you know how they apply sharpening, and how much? If an inappropriate and/or overly aggressive sharpening algorithm is used it can artificially exaggerate the grain.

Thanks- I should ask them that. When I ask questions they tend to be pretty responsive.

 

[RalphLambrecht]

I can't say I've noticed excessive grain when digitising FP4+, but then I use a DSLR rather than a scanner. The grain is there, but quite subtle; certainly nothing like as bad as some of the examples posted in this thread. I can post some examples if people wish.

I really would like to see them and inspect one up close.Can you share the details of your process, please?

 

[markjwyatt]

I did ask The Darkroom if the use sharpening and they say:

"...the only sharpening that occurs is pre-programmed by the manufacturer Noritsu and is within industry standards (we do not add any additional sharpening). If shots are even a hair under or over I do know that will bring out the grain more"

 

[Stephen Prunier]

I can't say I've noticed excessive grain when digitising FP4+, but then I use a DSLR rather than a scanner. The grain is there, but quite subtle; certainly nothing like as bad as some of the examples posted in this thread. I can post some examples if people wish.

GLS,
Can you say what DSLR you use, and are you shooting the negative ontop of a light box? I'm thinking about going this route. I have a FF DSLR, but a crappy scanner HP Scanjet 4050.

Thanks

 

[jtk]

That is what I am thinking, and with Ted Baker's observations this makes sense. I just shot a couple of rolls of FP4 at ISO 125, had them developed and scanned to 4000 dpi by The Darkroom. I am very happy with some of the images, but was a little shocked by some of the grain. I still have not gotten the negs back yet, so maybe I under exposed some. Here are some samples I have uploaded. https://www.flickr.com/search/?user_id=157638541@N07&view_all=1&text=Ilford FP4

I did very little post processing, but I did play with the dynamic contrast (In ON1 2018). Basically, I dropped the lowest scale quite a bit to de-emphasize the grain structure (but perhaps also killed some of my micro-contrast in Zeiss and other lenses), then bumped up the medium and large scale contrast for the overall image. I am considering standardizing on Ilford films given that Acros is practically gone (I have a few rolls of 35mm still), but may also look at Kodak Tmax.

The shot "Tree on a hill" really shows the grain. https://www.flickr.com/photos/157638541@N07/46038088421/in/dateposted-family/
"hillscape" does also, but here the grain is nice and works well with the scene. https://www.flickr.com/photos/157638541@N07/32167865318/in/dateposted-family/

I use FP4 @ 200, stand process for an hr with Rodinal 1+100, Nikonscan and see zero grain with 11X17 Canon Pro 10 inkjet prints (good inkjet is higher resolution than wet darkroom due to enlarging lenses). I know roots of Darkroom from yesteryear and am certain they aren't your problem...tho you seem to be asking too much from a Noritsu, which is only a convenience machine.

What is "lowest scale" and why would you sacrifice "grain structure" (if you see it in prints) to the look of Tmax?

And, of course, why are you using a lab for B&W film processing?

 

[markjwyatt]

What is "lowest scale" and why would you sacrifice "grain structure" (if you see it in prints) to the look of Tmax?

Dynamic Contrast adjusts the contrast at different size scales ("small" medium" and "large"). "small" tends to bring out fine details like textures. I lowered contrast on the "small" scales to de-emphasize the clumpiness of some of the scans. In the case of the negatives I used as examples, I saw that the one I really thought was grainy was a little dense (over exposed)- "Tree on a hill". The other one where i liked the grain structure "hillscape" did not look over exposed. This is probably part of it. I still used the dynamic contrast to de-emphasize the grain structure a little (especially in the sky which was still dense- I did use a medium yellow filter).

I also noted that the grain structure gets exaggerated by the ON1 viewer depending on how the image is scaled. It is usually not as bad in the final appropriately scaled (for screen size) jpeg.

Also, The Darkroom returns high resolution scans (6774 x 4792 for 35mm) as 24 bit (8 bit x 3 color channels) jpegs. I do not think they would do 16 bit TIFFs. This might help.

The shot "Tree on a hill" really shows the grain. https://www.flickr.com/photos/157638541@N07/46038088421/in/dateposted-family/
"hillscape" does also, but here the grain is nice and works well with the scene. https://www.flickr.com/photos/157638541@N07/32167865318/in/dateposted-family/

 

[GLS]

I really would like to see them and inspect one up close.Can you share the details of your process, please?

GLS,
Can you say what DSLR you use, and are you shooting the negative ontop of a light box? I'm thinking about going this route. I have a FF DSLR, but a crappy scanner HP Scanjet 4050.

Thanks

No problem, I'm happy to share the method. It surprises me there aren't more people doing it, as the results from what I've seen are clearly superior to flatbed scanning (and much faster once your setup is sorted). I guess people become set in their ways and don't want to change, and the DSLR method doesn't seem to get discussed much online. Anyway.

Here's a rundown of the method I use:

- DSLR + a good macro lens capable of 1:1 (I use a D810 and Tokina 100mm macro);
- A sturdy copy stand, which the above is mounted to (I use the Kaiser RS2XA);
- A diffused LED lightpad as light source, on the bed of the copy stand (I use a MiniSun LightPad, but any LED which provides even, daylight balanced light will do);
- An appropriately-sized glassless negative carrier from an enlarger, which holds the film flat whilst "scanning" (this sits on top of the lightpad)
- Rocket blower to remove dust from film
- Cable release

So, DSLR/macro on copy stand, pointing straight down at the film mounted in a glassless negative carrier on top of an LED lightpad, which provides even illumination. I did initially play around with other methods of holding the film (card frames weighted down, ANR glass sandwiches etc) but none gave as good/sharp results - or as rapidly - as using the negative carrier. I got a 6x7 carrier off ebay, which works for both my 6x7 and 6x6 formats. I set the height of the DSLR on the stand such that when focused on the film it is almost filling the digital frame on the shortest dimension; for my medium format work I just take a single shot like this, then crop, rather than multiple 1:1 shots + stitching as I find with the resolution of the D810 it really isn't necessary and just adds a ton more work. Large format film would however benefit from this approach I'm sure (as would DSLRs with smaller resolutions), and obviously if digitising 35mm film you would need the 1:1 capability if you didn't want to crop. You could use a non-macro lens in a pinch, but it won't give you as good results as the lens won't be optimised for close focus, nor be as flat field as a macro. You must also ensure that the camera sensor and film are on the same plane (I use a spirit level to check this), or you will get uneven sharpness acros the frame. I use a cable release to autofocus on the film grain (the cable release is important, as touching the camera in any way to engage focus moves the camera enough to throw the focus off once you let go of the camera again!). Once focus is thus set, I turn off the shuter release focus coupling in the camera menu, so using the cable release won't engage the focus again. Finally, I use mirror lock up to take the shots . In terms of exposure I use RAW in manual mode, with ISO 64 and f11 and a shutter speed which ensures I get an exposure as far to the right as possible without clipping any of the colour channels; if necessary you can always bring down the exposure later in RAW, but this way you get the most tonal information from the film. Negatives can then be inverted in the software of your choice. Slides can be a bit trickier to get them looking like they do on the lightbox; I've found that I have to crank the shadow recovery quite a lot from the RAWs (especially with Velvia, having so dense a base). Again, the D810 excels in this area, so other DSLRs having a lower dynamic range may struggle here (for e.g. requiring two blended exposures).

I hope this helps.

 

[markjwyatt]

GLS- I am working on similar workflows, initially for slides, but eventually for negatives. I am using a Fuji XT-2 + Pentax bellows + various adapters and I have a Pentax bellows 100mm lens, which is ok for the slides once I got a 1/2" rod machined to extend the Pentax slide adapter out. I was about to tackle lighting when I got a Durst ChromaPro slide copier. I am switching to that and it looks really good. It looks like a 75 mm lens will work best. I picked up a 50mm Nikkor and a 75mm Schneider (Companor- which may be on the lower end of quality) enlarging lenses. I have not tried the enlarging lenses yet (I tested with camera lenses to get the focal lengths correct). The illumination from the Durst is great. I think if I knock a small locating pin out, I can feed negatives through also. Based on the size of the light box and plate on the Durst, it looks like it will mainly be good for 35mm and 4x4 cm- may stretch to 6x6 cm, need to try for uniformity and need to produce a new top plate.

Have you tried enlarging lenses? I see them recommended because they tend to be flat field in nature. The 75mm Schneider Companor appears to be a budget lens with a square aperture, but I will give it a try. The 50mm Nikkor is great, but is more difficult with the crop sensor without cropping the slide. I am also considering copying with film, then having the film scanned (E6 for slides, then negative film for negatives- so I end up with scanned positives). I know it is one more intermediary, but for archival purposes, it may actually be better. Some of the images I want to save go back to at least the 1940s (I have prints that go back further). For now I will try direct to digital.

 

[Stephen Prunier]

No problem, I'm happy to share the method. It surprises me there aren't more people doing it, as the results from what I've seen are clearly superior to flatbed scanning (and much faster once your setup is sorted). I guess people become set in their ways and don't want to change, and the DSLR method doesn't seem to get discussed much online. Anyway.

Here's a rundown of the method I use:

- DSLR + a good macro lens capable of 1:1 (I use a D810 and Tokina 100mm macro);
- A sturdy copy stand, which the above is mounted to (I use the Kaiser RS2XA);
- A diffused LED lightpad as light source, on the bed of the copy stand (I use a MiniSun LightPad, but any LED which provides even, daylight balanced light will do);
- An appropriately-sized glassless negative carrier from an enlarger, which holds the film flat whilst "scanning" (this sits on top of the lightpad)
- Rocket blower to remove dust from film
- Cable release



So, DSLR/macro on copy stand, pointing straight down at the film mounted in a glassless negative carrier on top of an LED lightpad, which provides even illumination. I did initially play around with other methods of holding the film (card frames weighted down, ANR glass sandwiches etc) but none gave as good/sharp results - or as rapidly - as using the negative carrier. I got a 6x7 carrier off ebay, which works for both my 6x7 and 6x6 formats. I set the height of the DSLR on the stand such that when focused on the film it is almost filling the digital frame on the shortest dimension; for my medium format work I just take a single shot like this, then crop, rather than multiple 1:1 shots + stitching as I find with the resolution of the D810 it really isn't necessary and just adds a ton more work. Large format film would however benefit from this approach I'm sure (as would DSLRs with smaller resolutions), and obviously if digitising 35mm film you would need the 1:1 capability if you didn't want to crop. You could use a non-macro lens in a pinch, but it won't give you as good results as the lens won't be optimised for close focus, nor be as flat field as a macro. You must also ensure that the camera sensor and film are on the same plane (I use a spirit level to check this), or you will get uneven sharpness acros the frame. I use a cable release to autofocus on the film grain (the cable release is important, as touching the camera in any way to engage focus moves the camera enough to throw the focus off once you let go of the camera again!). Once focus is thus set, I turn off the shuter release focus coupling in the camera menu, so using the cable release won't engage the focus again. Finally, I use mirror lock up to take the shots . In terms of exposure I use RAW in manual mode, with ISO 64 and f11 and a shutter speed which ensures I get an exposure as far to the right as possible without clipping any of the colour channels; if necessary you can always bring down the exposure later in RAW, but this way you get the most tonal information from the film. Negatives can then be inverted in the software of your choice. Slides can be a bit trickier to get them looking like they do on the lightbox; I've found that I have to crank the shadow recovery quite a lot from the RAWs (especially with Velvia, having so dense a base). Again, the D810 excels in this area, so other DSLRs having a lower dynamic range may struggle here (for e.g. requiring two blended exposures).

I hope this helps.

Thanks

 

[GLS]

Have you tried enlarging lenses? I see them recommended because they tend to be flat field in nature

No, but it may work OK. I just went straight to the macro (in fact I bought it for this purpose) as it was the obvious choice.

 

[markjwyatt]

No, but it may work OK. I just went straight to the macro (in fact I bought it for this purpose) as it was the obvious choice.

That is why I bought the bellows 100. It works on the belllows well (with the extension rail). It did not work on the Durst (cannot extend enough to focus). The Durst has machined (ground) and pinned reference surfaces for the mounting which I would have to defeat to extend so it is easier to buy a new lens.

 

[markjwyatt]

...also- if it were desired to copy slides to slides, film choice is minimal today. Realistically it would be Provia or Ektachrome for color. Neither is probably ideal. Velvia would be too punchy I suspect. Any other ideas? Negative to positive processing of some type?

edit: orange mask kills the direct reversal potential; Rollei's ScanFilm 400CN Pro does not seem to be available, at least not for 35mm.

 

[jtk]

A passing comment: its easy to reduce contrast with Ektachrome if your camera allows multiple precisely registered exposures. Canon F1 always allowed that, Nikons needed expensive modification. It's easy to establish correct level of flash exposures by adjustment of flash unit (e.g. Vivitar 285). That calls for some bracketing. Pin registered masks work beautifully... darkroom wizardry....same as masking for Cibachrome.

 

[jtk]

I found Nikon scanners to be almost unusable for B&W negs. With the exception of XP2 of course. I can get scans from my V750 Pro that beat the Nikon scanner hands down. Grain, grit, whatever you want to call it was always the issue when using the Nikon scanners. Seemed to really accentuate it for some reason.

I doubt many Nikon users would agree with any of that unless they've got a very good slide-copying setup.. I've gotten good results with a Pentax APS-C camera (i.e. half of full frame) and an excellent Pentax macro lens on a cannibalized Durst enlarger stand.

Nikon has announced a new slide-copying device for their mirrorless cameras...that will undoubtedly be excellent.

Minolta scanners do diffuse grain and "dye clouds" ("whatever you want to call that"...is well known) with the 3rd party diffuser, but that extends Minolta's already far-too-long scan times.

 

[jtk]

yes agreed but if sharpening is on at time of scan real bad things can happen... Sharpening is best done in PS in the Luminosity channel.

A qualified YES: Be sure you're sharpening a duplicate/copy and not the original...AND you may be better off sharpening with NIK because it's non-destructive until you confirm it. I usually avoid any sharpening at all with 35mm/Nikon.