Neg scans (grain with FP4+)

[Prof_Pixel]

I would first ask the lab this simple question,,,, Did they use sharpening when scanning and wait for the response... for all good quality work the sharpening should be turned off ... I have a Imocan and input sharpening at the scan stage is the kiss of death...

... keeping in mind that scanning is a low pass (blurring) process and the image will require some sharpening at some point.

 

[Bob Carnie]

... keeping in mind that scanning is a low pass (blurring) process and the image will require some sharpening at some point.

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.

 

[Alan Edward Klein]

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.

Bob, what does that mean, low pass and luminosity channel?

 

[Prof_Pixel]

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.

Of course, but not every image goes through PS (or equivalent) - that's why the workflow being used is important. And, yes, the Luminosity channel since that's where the sharpness 'resides' . (The Green channel in an RGB image comes close).

 

[Bob Carnie]

Hi Alan.. I did not mention Low Pass, but I believe it is regards to PS sharpening protocal.

Luminosity channel or the L channel is the most important channel of all.. it is where one derives the underlying image structure and is the channel I use to increase or decrease contrast, It is the
channel that every person should be sharpening their files on... If you sharpen on any of the other channels, RGB or CMY the colour channels you are risking colour artifacts.

Dan Margulis has written two books and does workshops, he may have retired , I have all his books and I took his workshops 4 times.. not cheap at $2500 a pop. but his grasp of PS is/was unparalleled.

 

[Bob Carnie]

Of course, but not every image goes through PS (or equivalent) - that's why the workflow being used is important. And, yes, the Luminosity channel since that's where the sharpness 'resides' . (The Green channel in an RGB image comes close).

Not if its a picture of a red apple

 

[Ted Baker]

Not if its a picture of a red apple

But you guidance is PS centric, which of course is very relevant for PS users, and probably relevant to most scanning software, because of lack of development, or competition in that space.

Though I was lead to believe the Imacon software was pretty good, and not subject to these sorts of school boy errors?

 

[BMbikerider]

Hold on there. For those who think grain cannot be resolved by a scanner, well I will dispute that as long as you care to disagree. I use a Nikon Coolscan V sometimes but only very rarely for B&W because the grain can be very obvious. If it was the case that a scanner cannot resolve grain why did Nikon incorporate software to reduce the grain (at the expence of resolution)?
The colour corrected APO lens used in the Nikon Scanner will out perform any flatbed scanner where it may be difficult to resolve the silver grains.

 

[Bob Carnie]

Hold on there. For those who think grain cannot be resolved by a scanner, well I will dispute that as long as you care to disagree. I use a Nikon Coolscan V sometimes but only very rarely for B&W because the grain can be very obvious. If it was the case that a scanner cannot resolve grain why did Nikon incorporate software to reduce the grain (at the expence of resolution)?
The colour corrected APO lens used in the Nikon Scanner will out perform any flatbed scanner where it may be difficult to resolve the silver grains.

Have you seen a scan from a Creo Eversmart Supreme?? its a flatbed and pretty much equal to any drum scanner Aztec included.

Everything about scanning is controlled by a good operator.. in bad hands a Aztec is bad as a Epson.

 

[Ted Baker]

For those who think grain cannot be resolved by a scanner

No one said that. Grain and the silver particles are two different things. A scanner cannot even come close to resolving the particles. You can stand on the beach and see the texture in the sand very clearly, but it does mean you can resolve the grands of sand... A scanner can most certainly see the grain/texture patterns created, just like you see with your own eyes with a microscope or plain grain focuser. (none of those methods BTW will let you see the particles)

If it was the case that a scanner cannot resolve grain why did Nikon incorporate software to reduce the grain (at the expence of resolution)?

That's the heart of the issue, a scanner and for that matter an optical print will create different renditions of the silver particles on the film. A scanner/image processing can easily make it worse, and there are certainly ways to make it better.

 

[RalphLambrecht]

SInce you have problems uploading, and sending a crop of unknown size, I am offering to help to see if I can diagnose the problems, but never mind.

the uploading problem was solved and the crop is 800px on the long side as was suggested.Still, thanks for your offer.

 

[Richard Man]

the uploading problem was solved and the crop is 800px on the long side as was suggested.Still, thanks for your offer.

The suggestion was to RESIZE the longest side to 800 pixels, NOT crop 800 pixels out.

Sigh.

 

[Patrick Robert James]

Hold on there. For those who think grain cannot be resolved by a scanner, well I will dispute that as long as you care to disagree. I use a Nikon Coolscan V sometimes but only very rarely for B&W because the grain can be very obvious. If it was the case that a scanner cannot resolve grain why did Nikon incorporate software to reduce the grain (at the expence of resolution)?
The colour corrected APO lens used in the Nikon Scanner will out perform any flatbed scanner where it may be difficult to resolve the silver grains.

Well, I don't want to argue with you, just set you straight because you are basing your logic on assumptions. Nikon scanners do not resolve grain, you just think they do because you think you see grain.

There are flatbeds better than Nikon scanners. APO lenses don't mean much if the sensor doesn't resolve enough. Professional flatbeds like the Creo are better than a Nikon scanner. Also, the Minolta 5400 out resolves the Nikon by quite a bit, so the Nikon isn't even the best 35mm scanner.

I own a Nikon 4000 and for all around convenience I think the Nikon 4/5000 is the best 35mm scanner.

 

[Mick Fagan]

Ralph, this is an FP4+ 4x5" film developed reasonably densely for wet printing. I recently purchased a flat bed Epson V800 and this is about my 55th scan, so to put it mildly, I'm still crawling.

The image was taken with a 65mm lens with a centre filter, essentially, apart from a few dust spots and adjusting contrast I have this. The top empty sky has been cropped to make it more panoramic, plus it suits the subject.

There has been no unsharp masking done with this image.

Mick.

 

[BMbikerider]

No one said that. Grain and the silver particles are two different things. A scanner cannot even come close to resolving the particles. You can stand on the beach and see the texture in the sand very clearly, but it does mean you can resolve the grands of sand... A scanner can most certainly see the grain/texture patterns created, just like you see with your own eyes with a microscope or plain grain focuser. (none of those methods BTW will let you see the particles)



That's the heart of the issue, a scanner and for that matter an optical print will create different renditions of the silver particles on the film. A scanner/image processing can easily make it worse, and there are certainly ways to make it better.

Sorry I completely disagree. Silver grains, particles call them what you will are not the crux of the matter here. You don't scan unexposed film which contain silver in it's unexposed and undeveloped state so it is really a non-argument. I never mentioned and I don't think anyone else specifically mentioned silver halides (AKA particles) only yourself and being able to see them via a scanner.

The silver when developed clump together to form grain which can be resolved by quite basic flat bed scanners and most certainly by a high end film scanner.

 

[Ted Baker]

The silver when developed clump together to form grain which can be resolved by quite basic flat bed scanners and most certainly by a high end film scanner.

Well I think we talking about different things, and perhaps we are using terms slightly differently. I am using the term grain, to mean texture.

For example the ability to detect something is different to the ability to resolve fine detail, you could take an out of focus picture of a forest and as a result you may not be able resolve the trees properly, but depending on how out of focus it was you could still detect distinct texture or pattern, even pick out the trees...

Here are a few references from the folks at kodak... https://www.kodak.com/uploadedfiles...etters_filmEss_06_Characteristics_of_Film.pdf

Some excerpts:

When a photographic image is viewed with sufficient magnification, the viewer experiences the
visual sensation of graininess, a subjective impression of a random dot-like pattern.

Silver is developed or clouds of dye formed at the sites occupied by
the exposed silver halide. Contrary to widely held opinion, there is
little migration or physical joining of individual grains.

 

[BMbikerider]

I am talking about grain that we can see on reasonably large darkroom prints. I know it is grain which is formed by the silver halide particles clumping together following the development stage. The reason I say this when comparing a 12x16print in the darkroom the grain is evident but not obtrusive. The size of the grain varies depending on what film stock I use. With Ilford Pan F+, the grain is almost invisible at that size of enlargement, but change the film to say FP4+ or HP5+ the grain becomes more evident. When scanned on my Nikon Scanner the same variations happen I.E. Almost insignificant with Pan +and gradually more visible with FP4+ and HP5+. If I print some of my old Kodak High Speed Infra Red the grain is even larger - a characteristic of that film. So as far as I am concerned a good film scanner WILL resolve film grain from a developed negative.

Silver is developed or clouds of dye formed at the sites occupied by
the exposed silver halide. Contrary to widely held opinion, there is
little migration or physical joining of individual grains.

This I also take to task as well. I have a book which is quite old and was a publication by Ilford from 1966 (Ilford manual of Photography) where it explains why grain forms and it is the reason why it is more visible on 'fast' films than 'slow' versions. This is supported by the fact if a film is under developed/exposed the grains are much smaller than those when developed to the normal level. If a film is over developed/exposed the grain tends to be larger and far more pronounced.It also mentions the development also causes the invisible inert silver halide fragments to clump together during development to form the sections of the negative which are the light and shade representations.

I remain totally unconvinced by your argument.

 

[RalphLambrecht]

Ralph, this is an FP4+ 4x5" film developed reasonably densely for wet printing. I recently purchased a flat bed Epson V800 and this is about my 55th scan, so to put it mildly, I'm still crawling.

The image was taken with a 65mm lens with a centre filter, essentially, apart from a few dust spots and adjusting contrast I have this. The top empty sky has been cropped to make it more panoramic, plus it suits the subject.

There has been no unsharp masking done with this image.

Mick.

looks great

 

[Ted Baker]

This is I believe exposed and developed FP4, which is the under a microscope. (the OPs film)

The white scale is 20um, the blue line is roughly 6µm which is about 4000dpi (coolscan etc) and the blue is about 3µm, which is 8000dpi. I forgot to add a line for my epson flatbed which shows grain very well... it would be at least twice the size of the red line. As final bit of useless trivia the aperture size used by Kodak to measure graininess is 48µm across (2.5 times the length of the scale line)

The practical point of all this, that scanning like a pure optical enlargement is not perfect. And the various digital operations, the sampling, the re-sampling, and attempts to sharpen by identifying edges or using other techniques can make the apparent grain at least different to a traditional optical enlargement.

I am interested in techniques that minimise these effects in scanning. Often people will say things like 'a good scan is down to a good operator..."

This is the original thread with these images https://www.photrio.com/forum/threads/grainy-films-under-the-microscope.79400/

 

[alanrockwood]

Bob, what does that mean, low pass...?

Low pass means that an image is blurred. It comes from "low pass filter". In signal processing talk that means that high frequency components have been reduced in amplitude or blocked altogether. Since it is the high frequency components that are mostly responsible for capturing fine detail, reducing or eliminating the high frequency components causes a blurring of the signal.

"Frequency" in this context refers to space, not time.

 

[Adrian Bacon]

... keeping in mind that scanning is a low pass (blurring) process and the image will require some sharpening at some point.

Yes, as a last step after the image is at final output resolution and all other post processing has been done. At that point, sharpen to taste. If you do sharpening at any other point, all you’re doing is making the grain more obvious.

 

[Adrian Bacon]

Ralph, this is an FP4+ 4x5" film developed reasonably densely for wet printing. I recently purchased a flat bed Epson V800 and this is about my 55th scan, so to put it mildly, I'm still crawling.

The image was taken with a 65mm lens with a centre filter, essentially, apart from a few dust spots and adjusting contrast I have this. The top empty sky has been cropped to make it more panoramic, plus it suits the subject.

There has been no unsharp masking done with this image.

Mick.

FWIW, I’m a hybrid guy and that’s about what you should expect fp4 to look like with your set up, so you’re maybe doing a fast crawl.

 

[Mick Fagan]

FWIW, I’m a hybrid guy and that’s about what you should expect fp4 to look like with your set up, so you’re maybe doing a fast crawl.

Thanks, I really had little expectation, so far though, I have been reasonably pleased with what I have scanned from my 4x5" film archive

I did sharpen this image, but I didn't like what it did to the timber cracks in the king post of the fence; they sort of went slightly funny so I didn't keep it. Also, what you are seeing is a quite low resolution, the high resolution image on my computer sort of looks terrific.

I haven't changed my film developing regime one bit, old stuff and current stuff generally looks pretty good. I don't have anything flash computer wise, except my screen is 530mm wide by 297mm high, this being a pretty good ratio for landscape format, not that great for portrait, although an A4 portrait format is pretty nice.

Mick.

 

[flavio81]

I know little about scanning B&W negatives; that's why i Had my FP4+ negs scanned by a pro-shop but, the results where horrid(grainy). Is that normal?

Hallo Ralph. If the results are horrid, that's not a pro lab. I thought germans were obsessed with quality!!

Complain to the lab. We all know FP4+ gives great results, even scanned.

 

[Saganich]

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 grainclumps. I couldn't scan something that poorly if I tried...which suggests some problem with their process.