Is there any way to reduce grain in a negative scan?

[snusmumriken]

Look at the examples I posted in #39 and tell me what part of the noise you're seeing is grain, and what part is digitally-originating noise.

Surely you are both right? What gets taken for grain in the scan is an artefact resulting from digitisation of the projected grainy image. Even if the negative was grainless the scan would show some noise. But the scan of a grainy negative will look ‘grainier’ than the scan of a fine-grain negative, typically more so than a darkroom print. I thought all that was well established?

 

[Hassasin]

What you do sounds sensible and is similar to my workflow, but strictly speaking, this doesn't reduce noise ('apparent grain') in the image. It does prevent artefacts introduced in the scanning process and those resulting from resizing an originally large image to a smaller format. I'd call that avoidance of part of the problem, and not reduction.

What I do agree with, is that avoidance in the sense you're practicing it is perfectly adequate/acceptable for my personal purposes, for the most part, although it depends on the scanner I'm using. I find that some degree of noise reduction is sometimes preferable on scans I make with the Flextight, but not so when using e.g. the flatbed scanner. See (again) the examples I posted earlier, which demonstrate how much of an impact choices in scanners have on the phenomenon, and thus, digital workflow decisions will need to keep these into account. Thus, I wouldn't state that one approach would be 'best' for any given situation.

In general, my workflow focuses, like yours, mostly on avoidance of excessive compound artefacts and overlaps with your choices:
(1) Disable any sharpening or detail 'enhancement' options during scanner, insofar as possible.
(2) Apply unsharp-masking only to the final image, after any other editing, in particular downsampling.
(3) Adjust unsharp-mask parameters to the structure of the image. Key parameters are indeed radius, amount and also threshold (which I often leave at 0, but not necessarily always).


I think it depends on the material you start out with and the end result you want to arrive at. So no, I don't think it's possible to suggest a single type of shoe that will fit any given foot on the planet.



Look at the examples I posted in #39 and tell me what part of the noise you're seeing is grain, and what part is digitally-originating noise.
Your Minolta scanner is a fine machine, but as I said in #42 this is not about whether a scanner is any good. The fundamental phenomenon that the noise you see in the image is the sum of all aspects of the preceding imaging chain. The statement that you could separate them out is akin to arguing you can mix two colors of paint and then sort them back out while viewing the painting.

I was purely referring to my own scans and what I see, none of the ones posted by you. I see film grain on top of any noise which I suppose it does not sore my eye to any extent, hence I never even thought of what you are showing.

I am not disagreeing with your findings at all. Perhaps this is about how we view end results and I don't slice and dice any technical sides, so long as the image looks good as a whole and is free of obvious mishaps in shooting or processing..

 

[albireo]

Surely you are both right? What gets taken for grain in the scan is an artefact resulting from digitisation of the projected grainy image. Even if the negative was grainless the scan would show some noise.

Can you show proof of this happening? How do you establish the negative is grainless? What is your baseline? A microscope photo of the grainless negative? Or a wet print?

An interesting way to prove the above claim (which is, as far as I know, unproven) is to setup a well-powered perceptual test. A few human testers, a few negatives. Each negative wet printed (with established and proven focusing methods) as well as scanned and digitally printed (using a well established methodology, e.g. scanner operating at native resolution, & fully capable of focusing on the film plane etc). Compare the prints. Analyse the data. Could the candidates reliably pinpoint the 'noisy' film scans? Does it happen frequently enough as to allow us to discard the null hypothesis? If not, this scanner grain noise doesn't matter, it's just another wet printer's neurosis.

Perhaps what's getting the wet printing purists' knickers in a twist in this thread is the relative magnitude of this real or imagined 'digital noise' vs the actual 'rendering of the scanned negative grain'.

We don't have a sample from OP to understand if what they want to minimise is the 'added digital noise' or 'just grain' in order to focus the thread. so we can only speculate.

I personally speculate that 'digital noise' might dominate and define the image and obfuscate the qualities of scanned grain only in a few cases. Here's a few I can think of.

• the negative is underexposed or too dense, and perhaps forces the scanner's CCD and/or the downstream ADC chain to operate beyond design parameters
• the scanner is performing internal upsampling to offer a 'promised' resolution which is beyond its actual capabilities (this is the case with the flatbed example shown by Koraks)
• the user (eg the lab) is inexperienced with post processing and is oversharpening the scans.

If none of the above happens, there is no evidence that grain from a scanning process is distinguishable from grain from a wet printing process from a perceptual standpoint. We can want, desire, hope the scanned grain to be an imperfect representation of the grain in the negative as much as we want, but until we prove it's different, it's not different.

Amateur audiophile communities have had a similar debate for ages. Many analogue purists (known as 'golden ears') claim 16/44.1K redbook CD standard music is full of 'artefacts' and 'it's completely possible to tell a CD apart from its fully analogue vinyl counterpart' but (as far as I know) given the same mastering of the recorded material, nobody has been able to detect a statistically significant difference.

So I wouldn't really fret about it too much. We have no samples from OP. We don't know what it is exactly that they want to minimize.

 

[RalphLambrecht]

I can whole-heartedly recommend 'TopazAI" for this job'. The customizing features make it perfect and once satisfaction is achieved,all settings can bestored as custtom defaults. Even batch-processing can be done.

 

[runswithsizzers]

So I wouldn't really fret about it too much. We have no samples from OP. We don't know what it is exactly that they want to minimize.

So true. Like so many other "help" threads, there seems to be a tendency for us to rush in and try to solve the problem before we really understand what the problem is. Not critisizing, just an observation about human nature. (Here I might define "human" as "men-of-a-certain-age." And if I am pointing fingers, I notice there are three fingers are pointing back at me ;-)

 

[albireo]

So true. Like so many other "help" threads, there seems to be a tendency for us to rush in and try to solve the problem before we really understand what the problem is. Not critisizing, just an observation about human nature. (Here I might define "human" as "men-of-a-certain-age." And if I am pointing fingers, I notice there are three fingers are pointing back at me ;-)

But that's an enjoyable activity too, I personally don't mind more open conversation especially when interesting new "food for thought" is presented and debated.

 

[snusmumriken]

Can you show proof of this happening? How do you establish the negative is grainless? What is your baseline? A microscope photo of the grainless negative? Or a wet print?

Fair questions, and no of course I can’t. It was a trivial part of my argument, namely that the grainer the negative, the ‘grainier’ the scan will look. (I assume I don’t need to evidence that?). So to that extent, @Hassasin was surely right.

But out of curiosity, what does a scan of an XP2 negative look like? I don’t use the stuff myself.

 

[Alan Johnson]

I haven't seen this question posted recently, so... In my scanning journey, I of course have noticed the film grain that comes along for the ride when you scan the negative. It occurs to me that there's a lot of software that will reduce image noise in digital images, but I don't recall having seen any that deals with analog grain, which is a different animal.

My scanner has a resolution of about 80 lppm and I scanned Adox CMS20 II ,~800lppm, and Adox HR50, ~287 lppm. It appears from these figures that each "pixel" of scan would likely cover many grains in the films.
What happened is that the scans of CMS20 are effectively grainless but at the rather small decrease in resolution with HR50 there was a sudden jump in scanned grainyness.

It may be that below a certain level of grainyness in the film the scanner does not pick up any of the many grains that one scanner "pixel" covers but above a certain level of film grainyness one scanner "pixel" does pick up the many film grains and presents them as one uniform "pixel " of scanned grain.

I am pretty sure there is no such sudden jump with wet printing from my 11x enlargements in a different setting (not these films).
From the above speculation, it may be possible to electronically reduce the size of one scanner "pixel".
However there is no way to get back to the many film grains represented in this one scanner "pixel"
I am not suggesting the above speculation is proven.

CD4-LC low contrast developer for scanning

Since my pile of silver gelatin prints got too big I have been storing my "prints" on Flickr , more recently on Flickr Pro which allows a maximum of 6144 pixels in the long direction. This gives about 85 lppm to the viewer. My film is 35mm CMS 20 II Pro scanned at 7200 [true~3600] setting on a...

www.photrio.com

 

[MattKing]

But out of curiosity, what does a scan of an XP2 negative look like? I don’t use the stuff myself.

I can't locate an example at the moment, but it looks like a scan of a C41 negative that doesn't have any colour in it.
One intriguing characteristic of the film though is that, unlike classic B&W film, over-exposure tends to decrease the appearance of grain.

 

[runswithsizzers]

@snusmumriken, I have a roll of 135 Ilford XP2 Super posted here: https://garywright.smugmug.com/Photography/Ilford-XP2-Super-Aug-Sept-2023
These were camera scanned - details in gallery caption. Be aware: I do post processing to taste, nothing extreme, but no heroic attempts to standardize my digitized film processing for controlled comparisons.
EDIT: I just noticed an oddity in the way my SmugMug website works. If I want see a photo at 100% magnification, I must first Right Click> Open in a new tab, and then click that to see the full enlargement.

 

[JerseyDoug]

Re grain aliasing:

On the Black Scale Lab web site:

"I found the camera to have less influence on the quality of the scans compared to a lens. The optimal resolution for scanning depends on the grain size of your film. As a rule of thumb, the optimal resolution range for BlackBox135 is about 12 - 24MPx, and for BlackBox120 optimal resolution range is 24-36MPx."

"Lower-resolution scans will not capture all details preserved on film, while higher-resolution scans could intensify grain in high-ISO film."

At least in the range of 12MPx to 24Mpx this is consistent with my experience.

 

[snusmumriken]

@snusmumriken, I have a roll of 135 Ilford XP2 Super posted here: https://garywright.smugmug.com/Photography/Ilford-XP2-Super-Aug-Sept-2023
These were camera scanned - details in gallery caption. Be aware: I do post processing to taste, nothing extreme, but no heroic attempts to standardize my digitized film processing for controlled comparisons.
EDIT: I just noticed an oddity in the way my SmugMug website works. If I want see a photo at 100% magnification, I must first Right Click> Open in a new tab, and then click that to see the full enlargement.

I really wasn’t expecting the blank areas to appear so textured in scans. Did it surprise you? Thanks for sharing.

 

[runswithsizzers]

I really wasn’t expecting the blank areas to appear so textured in scans. Did it surprise you? Thanks for sharing.

While processing the scans, I didn't really notice the "texture" at the time. The brighter blank areas (sky) seem pretty clean to me, but I do see the texture you mention in some of the darker areas. I did especially notice the grain in the tree fungus shot, but that frame was pretty thin due to being underexposed.

Now that you mention it, I do vaguely recall reading several online reviews of Ilford XP2 Super bragging up how the film is "exceptionally fine grained." And Ilford's data sheet promises "super-fine grain." So maybe I should have been suprised by the moderate grain seen in my results. BTW, I did just double check my post processing in Lightroom for this roll, and I mostly avoided the usual suspects that tend to exaggerate grain, like boosing Texture, Clarity or Sharpening.

Anyway, due to my very limited experience, I cannot say if my results with XP2 Super are typical or not. I like the film just fine, but no more so than many other b&w films. For me, it is less hassle to develop my own b&w film at home than it is to send it to someone for C41 processing, so I probably won't be shooting any more XP2.