Film Scanning - Optimum resolutions for different film types?

[peripatetic]

For the scanning gurus.

I mostly use the C41 BW films, Ilford XP2 and Kodak BW400CN.

Because:

400 ISO
Easy to get processed in a high-street lab.
Allegedly good for scanning.



I am (still) considering purchasing a MF outfit, and because I can't afford a Coolscan 9000 am wondering what resolution scans I should pay for. The labs seem to base their prices on the final output size (I guess they have to base it on something), but clearly there is no point in paying for a 100Mb scan when a 20Mb scan will do the same job with the film I happen to be using.

So here is my question: At what point does one reach severely diminishing returns with a given film type?

Is there any point in scanning XP2 at 4000dpi? Will everything beyond 2000dpi be just extra grain/noise? My intuition says that for ISO400 films there is probably little extra detail in the 4000dpi scan compared to a 2000dpi scan. With Velvia50 I would expect there is extra detail beyond even 4000dpi.

Does anyone have a handy matrix for optimum scan resolutions for different film types? [Or have I just got hold of the wrong end of the stick with this notion?]

It also has longer term purchasing implications, for example if I do 90% of my shooting with BW ISO 400 films (C41 or silver), it may be that there is no real point in getting a Coolscan9000 over a V750 for example. The V750 is reported as having a 2500dpi optical resolution, if the film only needs 2000dpi then I can spend the money I might have spent on some extra lenses, or a holiday, gift for the wife, etc.

Thanks in advance.
Craig

[rogue_designer]

with 35mm, I would maintain that a 4000dpi scan is still the benchmark, even with higher speed films. Your resolving the grain in more detail, true, but that only shows as "more grain" if you over enlarge.

But they use output size as the guide for a reason. The resolution you need is entirely dependent on the size you intend to print.

In my opinion.. If you won't print larger than 11x14/13x19, you don't need the Coolscan 9000 - the 750 will perform admirably. But it will start to suffer in comparison on larger prints.

[bmattock]

In practical terms, there are too many variables to make specific statements about which DPI will be good and which will be overkill for any particular emulsion. The lens, the scene, and the capabilities of the scanner in question all enter into it.

Generalized statements might be:

1) More DPI is better than less.

2) There is a point of diminishing returns, but one seldom knows until one explores it personally what that might be.

3) Medium format and larger film requires less intensive scanning.

4) Dedicated film scanners do a better job than flatbed scanners, everything else being equal, due to DMax and DMin, as well as speed.

That said, there are plenty of flatbed scanners which do a perfectly acceptable job. If you are going to be scanning a combination of 35mm and MF, it might be the best way to go in general, especially if you're on a budget. If you intend to scan LF, you may not have any real choice.

My personal choice has been a ScanDual IV (no longer made) and an Epson 4490. No complaints so far. I'd love to have 'better' quality scanning kit, but at the price point I paid, I believe I've done well.

[Pherdinand]

I've printed a 50x60 cm (i.e. 20x24") frame from acros 100 35mm, that i scanned on the epson v700 flatbed. Although sharpness might not be critical in that subject matter, there is absolutely no need for higher resolution in that image!
I scanned the same frame on the minolta dual IV and there was absolutely NO gain in sharpness.

With slides, it's a different story. Dark areas get noisy way too fast with the flatbed.

[Pherdinand]

i forgot to add: i scanned at 3600 dpi and upscaled it to the needed printing pixel size. Scanning at nominal 4800 or higher resolutions made no sense at all.
Moreover, at the touted 6400 dpi film-scan resolution, the epson produces some weird pixelation so it is actually BETTER to scan lower rez and enlarge.

[mfogiel]

There are 2 considerations to make:
- at which ppi are we extracting all the information from film
- what is the "optimum" size of your files

If you sift through internet, you will find that most agree even moving from 4000ppi to 6000ppi still shows more information, and others say only 8000ppi really covers it all.

Having said this, let's see it from the other end: how big quality prints do you need?
Some reckon printing at 720 dpi is better than printing at 400dpi. etc, but as far as I am concerned I have decided to stop at 360 dpi, and even 300dpi is perfect most of the time. The height of a 24x36mm neg is roughly equal to 1 inch , so if you need to print 11 inch high (without cropping), you need a magnification of 11 times, that gives you 360*11=3960, at 16 times (for a 16x20) you need 360*16=5760, so depending on the size of the print, if you need the maximum quality ( same sort of thing you expect from Leica gear, BTW) you have to decide.

From my experience, if you are shooting frequently, the cost of a CS 9000 pays itself back in a few months against using a lab capable of a properly done 4000ppi resolution scans.

[Svitantti]

Fast films have bigger grain, yes... But they also have small grain there. This is what actually gives film the dynamic range (at least one reason to it). Bigger grains will be activated first (least light), because moro photons hit them. Smaller grain (slower film) will take more light and this is why the highlights wont blow out so easily.

So I would say there is not much sense dropping the resolution if you want to capture what really is in the film, as well as possible.

[Pherdinand]

I'm sorry but let's avoid confusion here, shall we.

GRAIN in traditional black and white photography results from the small gaps that appear in the emulsion between the silver crystals.
Bigger grains mean bigger gaps. Although their size can be associated somewhat with the real size of the silver crystals, it is NOT the same. The silver crystals are MUCH smaller.
Grains don't get exposed by the light, therefore. Grains are holes that let the light through and are only present after development.

If you don't believe it, ask yourself why grain is black on a negative that was already inverted to positive.

Regards!

[Svitantti]

why grain is black on a negative that was already inverted to positive.

Hmm. What is a negative that is already inverted to positive? And is grain black or white or both?

The developer acts with exposed grains and makes them "black" on the negative. The fixer removes unoxposed part of the silverbromide. So the grain on the negative is black and on a print it is white.

Does this have something to do with this topic? I dont know... Still there are small and big grain mixed in any fast film and the biggest grains are missing in slow films emulsion.

[peripatetic]

Okay, so I think I have gone some way to discovering the answers to my questions.

Norman Koren's pages about MTF are very useful...

http://www.normankoren.com/Tutorials/MTF1A.html

And looking at the comparative MTF graphs from the Kodak website it seems clear that BW films, have superior sharpness to Velvia 100 slide film. This may not be a surprise to most of you, but it was to me.

BW400CN has a very similar sharpness to Velvia 100. T-Max 100 and 400 are much sharper than Velvia, and even T-Max 3200 is as sharp!

BW400CN tech specs...

http://www.kodak.com/global/en/professional/support/techPubs/f4036/f4036.pdf

TMax films...

http://www.kodak.com/global/en/professional/support/techPubs/f4016/f4016.pdf

This helps for comparative purposes because there are lots of resources about how much detail there is available in scans of Velvia.

In the Norman Koren page he makes it clear that there is extra detail to be gained from 8000dpi scans over 4000 and 2400.

http://www.normankoren.com/Tutorials/Scan8000.html (Scroll to the bottom of the page.)

So the basic answer to my question is that there is extra detail to be had from a negative all the way down to 8000dpi, so get the best scanner you can afford. Clearly from the Norman Koren article appropriate sharpening must be applied, as a well sharpened 2400 image is almost as sharp as an unsharpened 4000 image.

Addressing the practical issues however, there is a point where you have to say that you have enough resolution for your purposes. (And this has been the thrust of the excellent comments above!)

Consider 3 price points:
1. Epson V750 -> £400 (2400dpi optical)
2. Coolscan 9000 -> £2200 (4000dpi optical)
3. Hasselblad Flextight X1 -> £8000 (6400dpi 35mm, 3200dpi 60mm)
4. esoteric... and £stratospheric

So on 60mm film @ 2400dpi we get 5640 pixels.
5640/300 -> 19 inches
5640/240 -> 24 inches

In practical terms I think that is going to be big enough from a 6x6 or 6x7 negative for 99% of my prints. For the other 1% I can fork out for a drum scan.

For 35mm work I have my Coolscan V ED, which is working out fine up to 8x12, and starts to break up a bit at 12x18 depending on the image.

So moving up to 6x7 + V750 with current workflow should provide pretty good 24x28 images; probably rather better than my current 12x18 from 35mm. If I can find a way to afford it however, the Coolscan 9000 would seem to hit a pretty good price/performance point for 6x7.

[Pherdinand]

The developer acts with exposed grains and makes them "black" on the negative. The fixer removes unoxposed part of the silverbromide. So the grain on the negative is black and on a print it is white.

Does this have something to do with this topic? I dont know... Still there are small and big grain mixed in any fast film and the biggest grains are missing in slow films emulsion.

Sorry, you are simply wrong.
Grain is formed by the gaps appeaing between the silver crystals and, therefore, in the negative grains are holes i.e. white. Whne you invert it (or make a print), they are black.
This is not a is-the-zebra-black-stripes-on-white-or-white-strips-on-black thing. This is better defined.

What it had to do with the topic? Maybe nothing, but as somebody brought up the subject and gave a wrong idea about grain, i thought to correct it, eventually.

http://gallery.photo.net/photo/2953761-lg.jpg
Here it is, strongly magnified negative, and inverted. What is black and what is white?

[Svitantti]

Sorry, you are simply wrong.
Grain is formed by the gaps appeaing between the silver crystals and, therefore, in the negative grains are holes i.e. white. Whne you invert it (or make a print), they are black.

So are you saying there are no grains in the film emulsion before it is exposed and developed? What determines the size of the grain if there are no grains in the film at first? How do films have different ASA speed?

Even if this is as you say, why would grain be the "holes" between the grains and not the silver grains?

[Pherdinand]

Dear Svitantti,
"why would" is not a question i can answer:)
The "grain" we see and that gets bigger and more visible with increased film speed and agitation and and and, is = gaps between the much much smaller silver crystals.
Third time i write this.
There are plenty of stuff i don't know; this one i happen to have read, learned and seen with my own eyes on my own material.
I see there is no way to convince you so i point your attention to any book (a good read is "Schroeders Negativ Praxis", you can even find it tri-lingual: english,german,italian) or proper description that treats the photochemical process(ing) on this level - or simply to a microscope and a piece of film.
Perhaps you are not even so interested in it. That's also fine.
But please do not claim untrue things, for people might believe it.

Regards

[Svitantti]

This is not a is-the-zebra-black-stripes-on-white-or-white-strips-on-black thing.
This is exactly what it is...

In the first place I did not even claim anything that you started speaking of. And for second, that actual grain in the negative is black and is silver. The non-grain is clear filmbase which is not called grain. In a positive image formed from the negative, the white parts are the grains on the negative and vice versa, yes... And the black parts are the metallic silver grain of the paper, if the print is done by traditional wet printing.

The size of the metallic silver grains also grows when film speed is increased. It is true that the gaps in between them grow too, but this does not mean the gaps are the grain.

Fact is, that the size of the grain is determined when the emulsion is produced. The sensitivity of the film is determined by the size of the grains in the emulsion (or more accurately the distribution of course). Not by the size of the gaps, because the gaps are not sensitive to light. Sure you could make a film with very few grains and big gaps, but theres no sense in such film.

The image is formed by developing exposed grains. They are what appear in the negative as black (grains). Without those grains, there is no image (process the film without developer, you get no grains or as you would say, one very large grain :)).

If you want, you can say grain (in the final product, positive image) is the gap between the metallic silver grains. This is just like the zebra-case. But on the negative, the grain is the grain, not the gaps.

[russianRF]

Wikipedia says, "Film grain or granularity is the random optical texture of processed photographic film (http://en.wikipedia.org/wiki/Photographic_film) due to the presence of small grains of a metallic silver developed from silver halide (http://en.wikipedia.org/wiki/Silver_halide) that have received enough photons."

Wikipedia isn't necessarily always right, but if a fact has made it to Wikipedia, I think it's generally acceptable that many intelligent people believe it. The way the Wikipedia article is worded, I get the impression that the grain is indeed produced by the silver grains. I got that impression reading Ansel's "The Negative" as well...

I'm not saying you're wrong, Pherdinand (http://www.rangefinderforum.com/forums/member.php?u=526). Is there somewhere online that backs up your claim about film grain being "holes" between the silver molecules?

[Svitantti]

It is just a matter of how you think about it. Like with zebras :-). And in case of the final positive image, because...

In the negative, the grain logically is the metallic silver, and not "the space between the things (grains)".

[Pherdinand]

Dear Svitantti again,

The stuff i was saying is not "to say what I want to say". I am sorry but it IS the correct and only definition of that visible grain in the film. What is CALLED grain on a film in photography, and what is pointed out, referred to, as big or small grain, visible grain, smooth grain, beautiful grain, ugly grain and popcorn grain, golfball size grain, is the gaps between the clumped-together tiny little silver crystals. These gaps can be big enough to be visible by a magnification of less than 10 times, depending (yes) on film type AND depending a lot on the development procedure which you seem to forget above.

Indeed it would make more sense to say that grain is black in the film and it is the silver crystals...except that, it is NOT what we see. Silver crystals are extremely small, they clump together to form the dark part of the image in the emulsion, and the small gaps between them are the grains. The gaps can be smaller and more uniformly distributed, giving smaller visible grain on magnification, or can be coarser, bigger holes. The black silver crystals however NEVER show up separated, never form individually visible crystals, they are just a lot of them small *******s clumped together, in three dimensions (thus not only one single layer).Many many of these crystals lying ABOVE each other form a thick layer of opaque clumps and make a negative region very dense. This will give the highlights when inverted. If they are forming a less thick layer because received less light and most of them stayed as silver halide and were washed out from the emulsion, the negative will be "thinner" i.e. more transparent. Where there was no light at all, most of them do not reduce to metallic silver, almost all is cleared out and film gets highly transparent.
In the ideal case, the silver crystals are very small and very uniformly distributed in size and position and the layer they form is very uniformly getting optically thicker or thinner, leaving a grain-free look when magnified. However this can only happen when the silver crystals are extremely small and tight packed enough not to leave small gaps between them. Normally there ARE gaps between them which lets light through where it should not let light through. This gives the grain that appears under relatively low magnification and is white i.e. transparent. Projected + magnified on paper, it is black, no matter what the papers' resolution or granular structure is.

Think about it like this: Throw in a jar a bunch of small semitransparent glass balls. If they are very very small, they will fit very tightly and you get a uniform transparency of the "layer" if you look through it from the top. The level of opaqueness will be determined by how thick is the layer and how transparent are the balls. Now, if you increase the ball size, you will get less tight packing of them and eventually, the transparency of the layer will not be uniform, you will see gaps when you look through it. In the case of big big balls, i.e. big silver crystals, which are indeed determined by film type AND development, no matter how opaque the balls are there will always be gaps between them unless you stack a real thick layer, or you cleverly change the balls into something that fit better and leave no gaps, like, the famous t-max or Acros or Delta films have flat silver crystals that fit tighter than traditional 3D-shaped ones.
However you will never be able to discern the individual balls i.e. the individual silver crystals, these will always be clumped together and way too small individually - unless you have only a few in the jar and a really strong magnification, of course.
There is a clear definition of grain, as i said it's not the zebra thing.

Did you actually take a look on the image i posted above? The image is a digitally recorded magnified piece of film (negative) through a microscope, inverted digitally to positive. Has nothing to do with paper image grains.
The grains are black. Obviously, since the image is inverted, the grains in the piece of films were W H I T E.

I mentioned a professional book above. I quote from it:
"It is the gaps that result in grain showing in the print." Page 15. There are even some nice illustrations.
Wikipedia does not contradict what i say. It says, the grains are due to the presence of small grains of a metallic silver ,which is true - EVERYTHING in the imaging process is due to the silver crystals. Wiki does not say, however, that the grain visible in magnifications is these silver crystals, which they are not.

So, that's it, i can't say more obviously; you either believe it or look it up or forget it all and keep thinking that you can see the silver crystals growing in the new years' afternoon light :)

edit:here it is,further magnified, the nose of the dude (and inverted).

[Pherdinand]

I'm not saying you're wrong, Pherdinand (http://www.rangefinderforum.com/forums/member.php?u=526). Is there somewhere online that backs up your claim about film grain being "holes" between the silver molecules?

Oh and, intelligence doesn't have a lot to do with this, it's just relatively simple chemistry and optics. But if in doubt whether i'm just fooling around with you, i do happen to have a Ph.D. in physics (magneto-optics, nanoscience, materials science). Which does not make me a better photographer at all, alas.

Finally, it is not correct to call them "silver molecules". They are simple silver atoms with some inevitable junk around them, precipitated in a sub-micrometer crystalline form.

[Svitantti]

I am sorry but it IS the correct and only definition of that visible grain in the film. What is CALLED grain on a film in photography, is the gaps between the clumped-together tiny little silver crystals. These gaps can be big enough to be visible by a magnification of less than 10 times, depending (yes) on film type AND depending a lot on the development procedure which you seem to forget above.

Indeed it would make more sense to say that grain is black in the film and it is the silver crystals...except that, it is NOT what we see.


Are you saying it is not metallic silver (grouped together) that I see as black when I am in the darkroom using grain focusing aid? Or scanning film with a film scanner (watching negative image, not inverted)? They are not the grain, but the gaps? Then what is the "white" or transparent area?

I would guess that silver is not transparent, as your magic balls in the example are...

Anyway I think now the thing is, what we mean by the word "grain".

[Pherdinand]

When you see black material,it is the thick layer of silver crystals clumped together. When you see white dots, it is light going through the gaps. When you make a positive out of the negative, see my image above, you get GRAIN which is black. This is coming from the gaps between the silver clumps, which is consistent with its color.

This is all i said... don't know how to say it more clearly.

[russianRF]

Do you have any evidence besides your own claim that you can refer us to, Pherdinand?

[Pherdinand]

well, don't know. Exactly how many and what kind of references would you like me to quote, if the book i suggested was not enough? Or do you think a random internet site would be more believable?

OK here's wikipedia, what it says further:
"Granularity is a numerical quantification of film grain, equal to the root-mean-square (rms) fluctuations in optical density, measured with a...bla bla."
Fluctuations in OPTICAL DENSITY. Not diameter of silver crystals, or anything like that - simply the fluctuations of the optical density, i.e. the fluctuations of the "transparency" of the developed emulsion, fluctuations that result from the gaps between the non-uniformly distributed silver crystals.
Further, see Leslie D. Stroebel, John Compton, Ira Current, and Richard D. Zakia (2000). Basic Photographic Materials and Processes. Focal Press. ISBN 0240804058. this is the second reference from the wiki article on film grain. Check from page 258.
see here http://books.google.com/books?id=BRYa6Qpsw48C&pg=PA264&dq=microdensitometer+film+grain+noise&lr=&as_brr=3&ei=0MW8SNmsFomkiwHBnezzBw&sig=ACfU3U0Jvob8U9WVhCYJDT0oqx77xSehJg#PPA258,M1
Here you can also read about what actually happens when you look through such a negative (or make an enlargement), as well as about the fact that the silver crystals are only "more or less opaque". As i wrote, they are very small. Very small grains of metal are getting less and less opaque, and they have even a different opacity on different wavelengths (i.e. some colors pass easier through them).

You can also read through this book http://books.google.com/books?id=DJ9bLcB3eeMC&pg=PT107&dq=%22rms+granularity%22+1000&lr=&as_brr=3&ei=jd28SPCLCIbujgGo2LzzBw&sig=ACfU3U0G1mEb9UuFdUzkVBeUbITBlpMa7A#PPT106,M1 from page 90, gives a good idea about what happens with grain, why it depends on film speed and other parameters, and why it looks more apparent when the film is grossly over/underexposed. Further, what the difference is between traditional BW film grain and color negative film grain.

Finally, check the text on page 413 and further 415 in this book:
http://books.google.com/books?id=HHX4xB94vcMC&pg=PA417&dq=selwyn%27s-law&lr=&as_brr=3&ei=AeG8SKCiLI6UiAGBlbj0Bw&sig=ACfU3U1704AE7mqHCpGWLYIHkAKK8JoHcA#PPA414,M1

about grain, why grain can be seen at low magnification...lo and behold, the book says, the sub-micrometer silver crystals ("largest is 2 micrometers) would need 50x or more magnification to be visible, but you can see graininess due to two reasons:
1: the random distribution in 3 dimensions lead to an apparent clumping that leads to a random irregular patter on a much largeer scale;
2: they may actually be clumped together physically,i.e. they are in real physical contact forming the random larger-scale patterns.
Check also on page 416 under title Factors ffecting negative granularity. Point 3 says clearly: granularity is the result of variations in density over small areas.

The grains you see in the negative or in a positive enlargement of it, are NOT the silver halide crystals - they are the light going through the gaps (more transparent regions) between the silver crystal clumps (more opaque regions). A random pattern of variation in the opacity (optical density) of the emulsion.

Are four specialized books on the subject enough for reference? i think i did my homework,now can i watch some movie?:)

[mh2000]

with the Nikon you have not reached the limit of C41 b&w film scanning at 4000dpi and you can see a difference when you scan lower or down sample, with an Epson you cannot see the difference between scanning at full resolution, 1/2 or down sampling by a factor of two (except that it helps smooth out the noise). For MF and the Nikon scans you may scan lower to keep your file size down if you don't intend on printing large, but that will be your choice... with the Epson it just isn't there. As to grain, the Epson does not resolve any level of grain or grain clumps... the Nikon does... though in truth, you are not actually capturing the full grain of any film with either.

[ferider]

I have no idea how you guys get even close to C41 film resolution for 400 ASA and up. I am using a Nikon film scanner and am consistently limited by aliasing. Must be doing something wrong.

Roland.

[Svitantti]

When you see black material,it is the thick layer of silver crystals clumped together. When you see white dots, it is light going through the gaps. When you make a positive out of the negative, see my image above, you get GRAIN which is black. This is coming from the gaps between the silver clumps, which is consistent with its color.

This is all i said... don't know how to say it more clearly.

...And nobody is or was arguing with this at any point.

Except that the black in a wet print is also silver-based density, so the silver(-grain) is again black (not the gaps). Of course in a scanned image which is then inverted there is no silver and the white parts are a result of the silver.

Still even in this case "grain" can mean either the white or black dots or just generally the texture.

Also it is not only the gap size in the negative that changes when film speed is changed (of film is developed differently), but also the size of the silver clumps, called grains are different size and shape, which I guess is very logical... Bigger clumps, when the silver crystals (or whatever they are) means there are more parts where the crystals are missing because they travelled to a clump.

More film speed means bigger clumps and bigger gaps, but still there are also smaller clumps in between.

[russianRF]

I'm baffled, Pherdinand (http://www.rangefinderforum.com/forums/member.php?u=526). The very links you provided seem to argue against what you're advocating. From the very first link you provided:

"The exposed silver halide crystals are transformed during development into grains of silver that are more or less opaque, depending upon their size and structure (see Figure 11-1B). The developed silver grains seldom conform exactly to the shapes of the silver halide crystals. The size and shape of each silver grain depends upon the combination of exposure, developer type, and degree of development in addition to the size and shape of the original silver halide crystal. As the silver grains increase in size, the spaces between the grains through which light can pass freely become smaller. The overlap in the depth of individual silver grains results in a rather haphazard arrangement of silver grain clusters."

I cannot, for the life of me, get how you would read that passage as confirming your idea that grain itself is the space between the clusters of silver grain. The gaps, or holes, do indeed have to do with the optical density of the various parts of the emulsion, but the size, shape, and graininess seems to be determined by, well, the GRAIN of the silver.

I don't know how much further you want to argue this; you can't come up with anything to cite that directly supports your view; everything you have presented thus far, it seems, either does nothing to support you or actively refutes you.

[Pherdinand]

Oh my.
The silver crystals, call them silver grains if you like, are very small. Two micrometers at most, in photographic emulsions. It's in the text if oyu read further.
To see the silver crystals (call them silver grains if you prefer) you would need a rather strong magnification. This is also in the text if you read further.
In small magnification such as usual photographic enlargements, or scanning, the GRAINS or granularity of the image shows up mostly in uniform image areas like sky, skin tones, snow...etcetera. Is this true or not? I'd say it is.
Why? because the grains you see in photographic enlargements are NOT the image forming silver crystals, which are very very small, but the fluctuations in the optical density of the semi-opaque silver layer. The granularity of the image is coming from regions which are less opaque therefore light can pass easier i.e. gaps between the silver clumps, and these regions are much larger than individual silver crystal (call it grain if you prefer) sizes. A ten times enlargement shows already visible grain.

This is what is written there in all those books i quoted, and this is what is also logical if you look at a slightly enlarged negative as negative - grains are white - or, as positive - grains are BLACK.

It has NOTHING to do with the silver layer in paper enlargements - paper enlargements of course also have granularity but they are less visible than the granularity transferred from film, since the film is enlarged. Think about it: When you make a bigger enlargement, you get bigger grains, thus obviously it comes from the film.
Do you see WHITE grains in a photographic enlargement, a paper image? Or black grains? I personally always find the grains to be BLACK.

[russianRF]

Oh my.
The silver crystals, call them silver grains if you like, are very small. Two micrometers at most, in photographic emulsions. It's in the text if oyu read further.

It also says that the grains clump together. I realize what you're saying -- you're saying the grains are too small to see individually. Either way, this still doesn't prove your point about the "grain" we see actually being the holes. If the grains are microscopic, logically the holes between them are, too.

But grains clump together. So what?

To see the silver crystals (call them silver grains if you prefer) you would need a rather strong magnification. This is also in the text if you read further.

You're good at referencing texts. I don't see you actually connect these texts to your point, though.

In small magnification such as usual photographic enlargements, or scanning, the GRAINS or granularity of the image shows up mostly in uniform image areas like sky, skin tones, snow...etcetera. Is this true or not? I'd say it is.

No, it is NOT true. Grain shows up EVERYWHERE. If you look closely at the details, it's there too. You'll read further in the links you provide that grain is essentially noise, and is in EVERY part of the picture.

Why? because the grains you see in photographic enlargements are NOT the image forming silver crystals, which are very very small, but the fluctuations in the optical density of the semi-opaque silver layer.

And those fluxuations tend to be in clumps of silver atoms, hence, GRAIN.

The granularity of the image is coming from regions which are less opaque therefore light can pass easier i.e. gaps between the silver clumps, and these regions are much larger than individual silver crystal (call it grain if you prefer) sizes. A ten times enlargement shows already visible grain.

So you claim. So you CLAIM. Nothing you've provided backs up what you say. You have given links that describe grain, but nothing says within those links that the grain effect we see is in fact the holes between the grains.

This is what is written there in all those books i quoted,

And yet you can never manage to find a place where these books you quote actually say what you're saying. Funny, that...

You know what? Peace. Believe what you want to believe; you obviously are going to whatever anyone says or whatever anyone points out. Just, look at what I, and other people, have said here to you. What we're TRYING to tell you. Maybe you can at least understand how you've failed to convince us...

[40oz]

I believe all Pherdinand is saying is that in a positive made from a negative, the image is formed by the light passing between the "grains" of silver. A white speck on a positive is evidence of silver blocking light on the negative. So if you see a black speck in an otherwise light sky, it's not evidence of "grain," but rather the absence of such in that particular spot on the negative.

Now, IMHO, what most people refer to as "grain" on a print is the interplay of light and dark in a region of little detail, such as the sky, not just the dark specks. If you look at the print under a magnifying glass, you aren't seeing litle black specks of "grain," but rather little black specks of the spaces *between* the "grains."

A rather esoteric distinction, but an entirely valid point.

[Svitantti]

Yes 40oz and Pherdinand are right. The black parts of a positive image are the result of light passing through more opaque areas of a negative, where there are less silver atoms and more "empty space".

I just wonder why the only definition for "grain" would have to be the black parts of a positive image. As 40oz is pointing out, this is not the case in general talk and it is neither the only "real" definition of grain.

[mh2000]

yup, the black things on my print are what I think of as "grain" ... never thought of it, but if they are black on my print they are clear on my negative... not that it changes anything for me.

[Revolucion Artistico]

So, yeah despite what grain is or isn't i've been watching this thread because I just got my first scanner and really don't know what I'm doing and was hoping to get some useful tips or tutorials so any suggestions that may be of use would be most appreciated. Thanks in advance.

Ryan

[Revolucion Artistico]

So, yeah despite what grain is or isn't i've been watching this thread because I just got my first scanner and really don't know what I'm doing and was hoping to get some useful tips or tutorials so any suggestions that may be of use would be most appreciated. Thanks in advance.

Ryan

[Revolucion Artistico]

Whoa, sorry for the double post.

[Pherdinand]

Yea, i'm really sorry for heavily hijacking this thread :)

Look guys, i am not arguing for the sake of arguing, and there's no need to say "peace". This is at the very end of my "to get angry about" list :)

I think we more or less agree with Svitantti actually, except that I think there is a clear definition of what people call "grain" in a photographic positive image; it is the dark speckles/dots, which obviously can only come from gaps in the silver of the emulsion. He says, no, grain is both the black specks and the white regions in between.
Well, a matter of definition, i guess, although i don't like to re-define well established notions but the physics will not change by this.
As to your point, russianrf: I am truly sorry that you still cannot understand what i am talking about. As i said before and as made clear by the other later posters here, think of what you see in a real case: you see black specks in the print, these can only come from transparent regions in the negative. If you need Ansel Adams to convince you about what you see, then i am sorry, i have no book of him to quote. If you are more willing to read through those pages what i pointed out (i am sure you read most of it but maybe you missed the point?) please check especially under the subtitle "what influences grain" or something like that, as i said in the first post when i mentioned the online book preview links.
I understand your feeling. I also don't like to believe every babbler on the Internet, especially when he writes something that contradicts directly what I thought before about the subject. Trust me I was also surprised when i realized - first came the experiment, i enlarged a piece of negative and inverted it and started thinking what the heck, how can silver grains be BLACK on the POSITIVE??? then i bumbed into the Schroeder's Negativ Praxis which has the nicest and easiest to understand description of how a photographic emulsion is constructed and what photochemistry does to it, and all became clear.

About referencing: Yes it DOES say in the text i pointed you to, that silver crystals are under 2 microns and 50 times magnification starts to show them, not less. But you have to find it yourself. When you asked abotu references, not having an online version of the Negativ praxis, i just looked randomly at the links on the Wiki page, and there it was, within 30 minutes i found it. So, good luck.

But really, it is not so important. Image is image, grain is there no matter if black or white, silver or gaps, so if there were no boring rainy winter evenings, maybe it was not worth so many words. :)

Have a good light (and gap-free silver clumps).

[Svitantti]

Yeah I think we agree on most parts. But I think we have to remember, that we can also talk about the negative, where for sure, the grain is the black parts because the rest is (virtually) empty and transparent.

Talking about prints, I've never even thought about it, I mean if a grain means white or black thing. Anyway the black parts are the grain of the paper in a wet print, but the white parts are a "projection" of actual grain in the negative. I think most people would agree that a grain can be any single "dot", be it white on black (shadows in a print) or black on white (highlights)...