Wow - read this re: Film versus Digital debate!

[vijay_nebhrajani]

Bernie: "Who's this Baines character anyway? And why should I defer to him?"

 

The guy from whose book those scans and images came, and whom I quoted - "Baines, H: The Science of Photography." Guy got the images from Kodak Research Labs, Harrow. Presumably Kodak knows what they are talking about.

[vijay_nebhrajani]

"Bernie: Sorry, 'Minimum' wasn't the right word. That should have been 'maximum', or what I really meant - 'smallest' resolvable feature."

 

Beat you to it - I got what you were trying to say and answered too - just above your post.

[bernardwest]

<i>The guy from whose book those scans and images came, and whom I quoted </i><p>

 

I didn't mean that literally. I know how he relates to the thread, but not why I should respect the dude. You mention Kodak. Don't forget the initial 400x grain image comes from Kodak as well.

[vijay_nebhrajani]

"Don't forget the initial 400x grain image comes from Kodak as well."

 

Sure, knowledge changes. Baines does say that there was the belief earlier in the binary theory, and that has

changed.

 

Also, the initial picture, like I pointed out, was a magnification of the catchlight in the models eye (actually

on the pupil) - very high contrast to begin with - so the grains would be nice and thick and black. Therefore

that image is at best, inconclusive. Now an image of gray grains - I can't see how that is inconclusive. If you

can see how, I'd like to know too - and I mean this seriously.

[theplayboyandthepoet]

hi phil,

 

there is one advantage to a film camera that digital will probably never have: no one will bother you, or

interrupt you, or come up to you, in the middle of a shoot -- or even worse, in the middle of a shot -- to ask to

see the preview pane. :-)

 

clients can be really impatient, and i feel the relationship between photographer and subject, and the nature of

shoots, has changed dramatically with the introduction of digital photography, and the preview pane.

 

with a film camera, people just lose interest in the camera, or back away scared. friends don't want to play

with it, because they don't know how to use it, and it's no fun if you can't immediately see what you've just

shot. analog has become that alien to most people.

[arjun_mehra]

It's amazing how many responses this infernal topic has received...

[vijay_nebhrajani]

"It's amazing how many responses this infernal topic has received..."

 

What amazes me is how many people seem to be following this thread. Evidently this thread has a lot of, uh, information content.

[bernardwest]

<i>Now an image of gray grains - I can't see how that is inconclusive. If you can see how, I'd like to know too - and I mean this seriously.</i><p>

 

I'm not saying that this new image is not necessarily right, it's just that it is so small and unclear. There's no doubt there are grey shapes in those images, but we know that shallow depth of field can cause these as well. Alls I'm sayin' is that we need a nice clear, say 1000x1000 minimum pixel image that we can all see nice and big on our monitors, and then make a final determination.

[msitaraman]

Gosh, Vijay, your responses to this thread are almost as numerous as all of your other posts to photo.net since 2001...

 

Anyway, welcome back to photo.net!

[djcphoto]

Film rocks, digital sucks. Simple really.

[vijay_nebhrajani]

Hey Mani - thanks! I've always been around, but used to post only occasionally. But it sure is great to run into

you again. Still using your Rolleis? About this thread - I don't know - just a fun topic, I guess.

[vijay_nebhrajani]

<i>Bernie: "I'm not saying that this new image is not necessarily right, it's just that it is so small and unclear. There's no doubt there are grey shapes in those images, but we know that shallow depth of field can cause these as well. Alls I'm sayin' is that we need a nice clear, say 1000x1000 minimum pixel image that we can all see nice and big on our monitors, and then make a final determination."</i>

<p>

What you say is a valid doubt - but remember the context - Baines is trying to show "gray grains" (or partially transmissive silver specks) - so if that were a false image, he would be lying and Kodak would be endorsing it. Kinda doubtful, but hey, another controlled experiment can't hurt.

[vijay_nebhrajani]

So Bernie, to recapitulate - assuming for now that the Kodak "gray grain" image is valid, what is your current position? Which "camp" are you in now?

[bernardwest]

<i>recapitulate</i><p>

 

Interesting choice of word there VJ...<p>

 

If the image is valid and repeatable, then i'm in that camp. I'll remind you though, that I've never really been in any

camp, not being conversive with the minutiae of the argument. I've based my whole position on the original 400x

image and anecdotal evidence of Daniel and Adams.

[photo_smith]

Bernie

Who's this Baines character anyway? And why should I defer to him?

 

Well, he was Kodak Chief Chemist and Deputy Director of Kodak Research in Harrow UK. He had a first class Hons

degree in Photographic imaging and was instrumental in many emulsion advances.

 

Bernie

I've based my whole position on the original 400x image and anecdotal evidence of Daniel and Adams.

 

Well DLT culled his image from Tim Vitales pdf which in turn was borrowed from Kodak Tech doc H1. The image has

been used slightly out of context. Unfortunately H1 is not current, I've seen it but no longer have a copy but have been

promised one by Kodak Research. when I spoke to them on Wednesday.

 

Rather than base your whole belief on one image, what you should do is use those images for a point of research, the

original texts that go with that image are crucial. I have tried to lay out the mechanism by which film works and my

understanding of that process comes from H. Baines, C.E.K Mees and my conversations with Kodak engineers.

Over the last 18 months I've read several books including the ones mentioned as well as ones suggested by my sources

at Kodak. This evidence is NOT anecdotal but taken from literature used at places like RIT when teaching.

If you need extra info get a copy of C.E.K Mees works.

This is from In his book 'The Fundamentals of Photography' C.E.K Mees states:

"Any silver deposit in the negative will let through a certain proportion of the light which falls upon it. A very light deposit

may let through half the light, a dense deposit one-tenth, a very dense deposit one-hundredth or even

only one- thousandth".

Similar to Baines? You bet could be they are both wrong, the established rules of emulsion chemistry and physics taught

by Kodak are wrong and Reichman and DLT are correct based on what they 'see' in one image-you decide.

http://photo-utopia.blogspot.com/

[marshall]

Not that this is relevant, but an average full-length novel tends to run 110,000-120,000 words. This thread exceeds 160,000.

[bernardwest]

Wow, that's impressive. I'm now thinkin' "War and Peace" territory. We could call our thread "Black and(or) White(clear)"...

[razzledog]

35mm film has never been so good. I use whatever medium is most suited to the job, but a DSLR is nowhere near as much fun as shooting 24x72 transparencies with this baby........<div></div>

[bernardwest]

<i>Rather than base your whole belief on one image, what you should do is use those images for a point of research,

the original texts that go with that image are crucial.</i><p>

 

Mark.. I've been doing some reading over the last few days, and have been having a look at at least one of the books

you have been quoting (Mees). All the stuff I have read have been serious scientific treatises on the subject. NOw I

haven't been able to get into them in any great depth, but have had what I would call a 'detailed browsing'. The three

I've got with me now are: Mees (1954) - revised edition; Kowaliski (1972) "Applied Photographic Theory"; and

Marchetti and Eachus (1992) "The Photochemistry and Photophysics of the Silver Halides", in the journal "Advances

in Photochemistry" vol 17. These last two authors where with Eastman Kodak Labs. And as far as I can tell,

Kowaliski was with Kodak in Paris. ALL of them (including ALL the other ones I looked at as well, but haven't

quoted) have a statement to this effect: "Once initiated, development of an exposed individual grain continues until

<b>ALL</b> the silver halide is converted into metallic silver". This meshes with the literature that Rishi quoted

earlier as well. My interpretation of this (and other info I have read) is as follows: The development reaction intiation

is a random process. Once a grain that has been exposed begins to develop it develops to completion VERY

quickly. Infact other stuff I have read states that the rate of reaction is exponential, ie. it speeds up. This is because

it is self-catalysing. If you stopped development early, the state of play would be this: Due to the random initiation of

the reaction, a random sample of grains would have started developing, and a random sample wouldn't have started

developing. The vast majority of those that had started developing would have developed to completion due to the

very fast rate of reaction. A small number would have stopped at some intermediate stage. These would be

your 'gray grains'. But there would be a VERY small number of them. The vast majority of grains would be either

Black or clear.<p>

 

There are a couple of other points to make as well. Firstly, ALL these books, including the 1992 reference, make it

clear that this field still contains a lot of unknowns (as at 1992 anyway). THey ALL talk in terms of "hypotheses".

Clearly somethings are well known, but the further back you go in the literature, the more they where hypothesising

about a lot of the stuff. Mees (1954), a book Mark lauds, has this statement in it: "In view of the continuity between

the formation of the latent image and the production of a visible image, there seems to be every reason to believe that

the latent image produced by exposure to light consists of aggregates of atoms of metallic silver". More: "In

conclusion, it is not proved that [sensitivity] centers exist before exposure although there is evidence in favor of the

assumption". That is, the state of play for the field when this text was written, wasn't even sure of how latent images

form. It goes on to state a number of other competing theories. The point is, much of this work at the time was

hypothesis. We need more up to date references to hinge our own hypotheses on. <p>

 

The second point I wanted to make was the chemistry of development, as talked about in this thread, is a very

simplified version of what goes on. There are other compounds in the gelatin and the developer which have effects on

the rate of initiation and development of the developer.

[photo_smith]

Bernie I'd like that reference because this statement is incorrect.

Once initiated, development of an exposed individual grain continues until ALL the silver halide is converted into metallic

silver.

That statement would only be true if you develop the film to completion, as they used to do in the 1940's. since the

increased popularity of roll films most development is 'gamma based' i.e different films have different developer times it

is rare to develop to completion.

In most cases people over expose and under develop, this increases tone.

You also state:

"A small number would have stopped at some intermediate stage. These would be your 'gray grains'. But there would be

a VERY small number of them. The vast majority of grains would be either Black or clear".

Which is also true, but not under normal developing conditions.

Most developers are low alkaline, slow acting (normally metol based) so it is rare that you will get developed to

completion in mono films.

 

Another statement:

The development reaction intiation is a random process. Once a grain that has been exposed begins to develop it

develops to completion VERY quickly. Infact other stuff I have read states that the rate of reaction is exponential, ie. it

speeds up.

 

Although the development process is self catalytic it isn't that fast in fact it's pretty controllable, modern films also

include bromides which slow down development, it is possible to have very slow development over hours, sometime referred to as 'stand development'.

If the development is too quick tonality suffers, and development inconsistencies occur.

Keep reading Bernie. BTW my book is the later Mees and James edition (1977)

Mark

[rodeo_joe1]

I'd like to throw another spanner into the works of the "binary" film grain argument. The very fact that an electron

microscope has to be used to see those silver filaments should tell us something very pertinent, and that is that

visible light just isn't capable of resolving the filaments of silver from the spaces in between them. Blue light has a

wavelength of around 400nm - just under half a micron - and therefore cannot pass through a "hole" much smaller

than that.

 

Sorry, but the wire-wool pad analogy just doesn't work at the scale of film grain, because visible light just can't

squeeze through the sub-micron sized gaps between silver filaments. The optical "school" microscope is a far more

realistic way of viewing film grain, since it's the interaction of developed silver with LIGHT that we should be

considering, not its interaction with accelerated electrons.

 

In the real world, black and white (and they don't call it that for nothing) film grain is opaque. The only thing that gives

the illusion of continuous tone is the random distribution and varying size and sensitivity of the halide crystals in the

emulsion.

 

OTOH, there seems to be a comparison going on between monochrome silver-based negatives and RGB colour

pixels. Hang on guys! You just can't do that. There are only 256 shades of tone available in an 8 bit greyscale, not

16.7 million. If we're going to compare like with like, then we should be comparing film grain with 255 (8 bit) or 65535

(16 bit) shades of grey. Either that or compare the 16.7 million hues obtainable from RGB pixels with the clusters of

cyan, yellow and magenta coloured blobs of dye generated in a colour film. And BTW, film manufacturers go to a lot

of trouble to ensure a reasonably consistent size (around 5 microns), density and distribution of those blobs of dye,

which kind of undermines the continuous tone argument for film.

 

I can't help but agree with those contributers who have turned their backs on these purely theoretical arguments. It's

results that count, and I for one can clearly see that prints from my full-frame 35mm sized digital cameras easily

compete with anything I can get from medium format film. And I don't have any nightmares about the lab messing up

the processing, or having to take immaculate care of film originals in case they get scratched, dusty, mouldy or

faded.

 

The commercial reality is that film is dying, and rather than cling to a sinking ship, I just want to get on with taking

pictures for as long as I have strength to press a shutter release. Now excuse me while I take a few handheld shots

in near darkness with my new D700, which I could never have got with film.

[photo_smith]

Joe:

Sorry, but the wire-wool pad analogy just doesn't work at the scale of film grain, because visible light just can't squeeze

through the sub-micron sized gaps between silver filaments. The optical "school" microscope is a far more realistic way

of viewing film grain, since it's the interaction of developed silver with LIGHT that we should be considering, not its

interaction with accelerated electrons.

 

From "The science of photography"

"It has been mentioned that if development of an emulsion grain is observed under a microscope it is seen to start at

one or more points on the surface, and to spread therefrom throughout the grain (p.112). Under these conditions, once

development has started, it proceeds with rapidity, and the whole grain is almost immediately converted to metallic

silver. It was, therefore, concluded that a partially developed emulsion must consist largely of completely developed and

completely undeveloped grains, with very few partially developed grains. This conclusion was erroneous and based on

the assumption that the conditions on a microscope slide are similar to those in the body of an emulsion layer. A partly

developed film or plate does in fact contain a high proportion of partly developed grains --- if it did not do so, there would

be no such thing as fine grain development."

 

BTW grain is not opaque.

In his book 'The Fundamentals of Photography' C.E.K Mees states:

"Any silver deposit in the negative will let through a certain proportion of the light which falls upon it. A very light deposit

may let through half the light, a dense deposit one-tenth, a very dense deposit one-hundredth or even

only one- thousandth".

[bernardwest]

Mark, my totally non-expert opinion is that you may be confusing the development process. From what I've read, it

is essentially a two-stage thing. "Initiation" and "Development" or something to that effect. The Initiation phase is

what takes up the most time. And the initiation time is randomly applied. That is some grains will begin developing

straight away and others will take a long time, all randomly distributed. From memory, this random initiation is

independant on amount of exposure, but is somewhat related to the size of the grain, and by implication, the number

of sensitivity centres it has). So, within grain-size classes, the initiation of development is random. Once the

development begins, it goes to completion very fast (i think the reference I found stating this is one of the ones I

didn't borrow from the library - I could only carry so much :)).<p>

 

The Mees(1954) text I have is titled "The theory of the photographic process". I accept that you weren't referring to

this one, and that you have a more up to date version. The quotes about development are this: "Once initiated,

development of an individual [exposed] grain continues until <b>all</b> the silver halide is converted into metallic

silver." - Mees(1954) page 101; "...plates coated with a thin, dilute emulsion layer in wihich the grains are well

separated and lie essentially in a single plane. When such a layer is exposed to light and developed, some of the

grains, apparently a random selection, become more or less <b>completely</b> reduced to silver while the rest

remain unaffected until removed by fisation. This is true even if attention is confined to grains of some definite size

and shape." = Mees (1954) pg. 169 (nb. I should point out that Mees is the editor of this text, and the two quotes are

from two different contributing authors); "The purpose of development is to reduce the exposed crystals

<b>entirely</b> to metallic silver in as short a time as possible, without modifying the unexposed ones." - Kowaliski

(1972) - pg 413; This example ends with the quote "....Similarly, as long as the electronic affinity remains above the

Fermi level the trapping of an electron is highly improbable, but as soon as it decreases below it the centre traps an

electron of the developer and the reduction [of silver] then continues at a <b>steadily increasing rate</b>." -

Kowaliski (1972) pg 392. The susbstance of what came before this quote (too long to type in here) was how the

latent image centre in an exposed crystal grows with development. In this particlular discussion there is also this

quote "...which allows an increase of the latent image centre by the next photoelectron to three silver atoms; these

again attract another silver ion and the process continues until the <b>complete reduction of the

crystal</b>."; "However, at present, no microscopic theory detailing the formation of the latent image, the basic

photochemical product responsible for image recording, has receieved complete acceptance." - Marchetti & Eachus

(1992) - pg 147. Keep in mind though this was 1992, so things could be different now.; "...These small silver clusters

[latent images] catalyze the development (conversion to metallic silver) of the <b>whole</b> silver halide grain (about

10^10 silver atoms) whent the coating is immersed for a few minutes in a suitable reducing agent - the developer." -

Marchetti & Eachus (1992) - pg 152.

[bernardwest]

<i>The very fact that an electron microscope has to be used to see those silver filaments should tell us something very pertinent, and that is that visible light just isn't capable of resolving the filaments of silver from the spaces in between them. </i><p>

 

Good point. Not that I'd expect you to find it in the above novella, but we have mentioned this numerous times. Thankfully the occurance of EM images has declined since then.

[photo_smith]

Bernie

What you are saying is very different from what is in my text:

A partly developed film or plate does in fact contain a high proportion of partly developed grains --- if it did not do so,

there would be no such thing as fine grain development."

 

when you see 'the complete reduction of the crystal' are you confusing the whole crystal with the development in

proportion to exposure?

I have had several meeting and e-mails with Kodak engineers (which I forwarded to Rishi and Vijay) and they have all

stated that grains do not develop to completion, normally they develop structures that are proportional to the exposure

they receive. I think it would be unusual for a grain that receives little exposure to develop the whole crystal in the same

manner as one that has more exposure.

Bernie I'd read on if I were you, I've spent 18 months reading this stuff and I'm convinced that grain is not opaque or

'black and clear' ALL the Kodak guys I've spoken to confirm that.

meanwhile its 1-30 am here

sleep