Wow - read this re: Film versus Digital debate!

[vijay_nebhrajani]

Rishi,

 

If your camera/tripod rings, typically the type of resonance you see in steel parking garages when cars go over a

bump (say at around 10-15 Hz), you'd see exactly the kind of pattern you see in your pictures. Linearly moving,

bright subjects look stepped and static subjects look "spread out" as if they were out of focus. Though possible,

it is difficult to imagine that a section of I-5 has bumps at exactly the same intervals causing such a regular

pattern. Different cars have different suspensions, and you would see different amplitudes of the bumpiness wave

depending on if it was a Lexus LS460 or a Dodge Ram truck. You don't see that in your picture, so Occam's razor

suggests it is camera vibration. With your digital picture, you are already at the sensor limit, and it is

impossible to distinguish the pixels from the car streak waviness.

 

I've taken pictures of fireworks from near the Golden Gate bridge in San Francisco, and I was standing on hard

ground but it was windy, and I saw exactly the same behavior - the firework streaks were stepped, and everything

else was slightly fuzzy, as if it was out of focus.

 

I'll admit that it is difficult to say with certainty that there were camera vibrations in your pictures, but it

seems likely.

 

I do agree that to do a meaningful comparison you'd have to create more controlled conditions - same lens, same

time of day etc. I'd also suggest avoiding long exposures, moving subjects etc; the best would be some complex

but immobile subject under studio flash but maybe that's a stretch.

[rishij]

Vijay,

 

I'm going through you argument for the theoretical resolution of film & proving Reichmann wrong. And here are a couple of glaring problems with your argument:

 

If you read Norman Koren's page here: http://www.normankoren.com/Tutorials/MTF1A.html

 

You will see that, according to manufacturer's specs themselves on film, the finest resolution film has a 50% contrast (response) at 40 cpmm, and only a 10% contrast (response) at 120 cpmm. So your claiming that somehow for 35mm film photography the limiting factor is lenses, and not the medium, seems wrong to me JUST based on manufacturer's specs (and I doubt they UNDERestimate their specs).

 

Secondly, you state:

 

"Now before somebody comes back and says that there may be some way for grains to be stacked along the third dimension (i.e., that of emulsion thickness) that can give the effect of having a smaller area for 48 grains (i.e., a film pixel), please realize that this can't be possible with binary grain since a single "black" grain will completely occlude all lighter grains, essentially resulting in a binary AND operation, i.e., creating another binary grain, third dimension or not."

 

I disagree.

 

For argument's sake, let's say grains *are* binary. Let's call the depth axis running through, say, 3 layers of film, the Z-axis. You're incorrectly assuming here that one black grain along the Z-axis results in pure black. Rather assume that 3 black grains stacked on top of one another along the Z-axis results in black. Two stacked on top of one another yields 66% (slightly darker than gray, which is at 50%) black, and one grain yields 33% black (slightly lighter than gray). It's not like each grain of silver is a 'black hole' of black. I'm sure that light can still travel through a grain of silver. Stack 10 layers together, and maybe it has a harder time traveling through 10 stacks of black silver grain. Of course, in reality, the random dispersion of silver grains means that it's unlikely that any two will be stacked perfectly on top of one another in adjacent layers; however, my scenario is purely for argument's sake. A single black will NOT occlude all lighter grains (& by 'lighter', if you're trying to disprove Reichmann's argument, I think you mean 'clear'), because you have to look at it as: a single black is lighter than 2 blacks grains stacked on top of one another, is lighter than 3 black grains stacked on top of one another.

 

More coming later...

Rishi

[rishij]

OK, Vijay, I see what you are saying about camera/tripod vibration. I need to ponder on it a little more but it does make sense. The only part of it that is hard to swallow is: since the aperture was so small, many light streaks have to add up to register a signal on the film/sensor, and all of these light streaks have to be from cars traveling at roughly the same speed, and then the sensor has to vibrate up/down at the same frequency (ok, believable if it's a resonance phenomenon) every time a car drives by on the bridge from where this picture was taken. If the cars were traveling at different speeds, we'd see different frequencies in the light streaks... but there really seems to be only one (dominant) frequency.

 

The one dominant frequency is what led me to believe that it is the road. Yes, different suspensions could give you different amplitudes. But different car speeds will give you different frequencies of the registered sine-like wave.

 

So which is it? LOL, this is hurting my head :)

 

More importantly, though, Vijay -- from an Imacon 848, do you expect to see a sharper scan from 35mm RVP 50? In my experience, I don't. I do have sharpening turned off on my Imacon though, because I don't trust it and I sharpen later in PS/LR.

 

Best,

Rishi

[vijay_nebhrajani]

Rishi,

 

Talking about black and white film for now, silver is opaque, so a black grain will not be 66% gray, it will be

opaque. If there are any clear grains along the z axis that it occludes, you won't get lighter shades of gray -

it is the case of an opaque object occluding a transparent one, resulting in an opaque object.

 

If you postulate that a grain is either clear or 66% gray (or any value other than black, opaque), what you are

saying is that silver is not opaque, which isn't true, right?

 

If you postulate that grain is completely black, then it must occlude, meaning that the third dimension is

meaningless etcetera.

[rishij]

I'm saying that an individual grain looks filamentous when developed, therefore it is not a <b>perfect</b> absorber (is that a word?) of light. Even if we do call grain binary, it's not that simple. It's not binary as in 'blocks 100% light being transmitted' or 'let's 100% light through'. That's impossible. Think of it this way: if you shine brighter & brighter light at it, some is bound to get through a black silver grain even if it is completely black. In such a case, having another black silver grain on top of it will allow *less* light to go through as opposed to just one black silver grain in the path of the light.

 

Therefore, two black grains stacked on top of one another will register darker, in a scanner say, than one black grain. Stack even more layers, and you've got even more tones representable by any given Z-axis value. Up to a certain point, of course. Let's just say that after 20 such grains perfectly stacked on top of one another, it becomes so dense that no light (from any practical scanner light source anyway) can get through. THEN, having 20, or 25, or 30, or 50 grains stacked on top of one another will not make any difference.

 

But I certainly don't think that you can say that one silver grain in and of itself is sufficient to completely block a path of light running right through it. There are clear areas within this filamentous region, as grain images show. Stack enough of these on top of one another though, and then you can reach pretty high densities.

 

And, of course, it's also possible to increase density by creating clumps of silver. What you use, more clumps, or more layers, will of course affect your resolution. The more clumps you use to represent darker tones (or lighter ones depending on positive/negative film), the lower your resolution. If you can expand the # of tones recorded by increasing the # of layers, that's probably preferable.

 

Again, this is all assuming that Reichmann's argument is correct.

 

And I still need to write more on that. Which I'll get to after I finish reading this Chem. Review paper by Theys & Sosnovsky (1997).

 

Rishi

[photo_smith]

"But I certainly don't think that you can say that one silver grain in and of itself is sufficient to completely block a path of light running

right through it. There are clear areas within this filamentous region, as grain images show. Stack enough of these on top of one another

though, and then you can reach pretty high densities.

 

And, of course, it's also possible to increase density by creating clumps of silver. What you use, more clumps, or more layers, will of

course affect your resolution. The more clumps you use to represent darker tones (or lighter ones depending on positive/negative film),

the lower your resolution. If you can expand the # of tones recorded by increasing the # of layers, that's probably preferable.

 

Again, this is all assuming that Reichmann's argument is correct."

 

Well it isn't correct.

 

When a single photon hits a positively charged AgBr+ molecule it separates the Ag and sends it on a journey towards the sensitivity

speck which is sulphur coated with gold (gold increases the chance of the bond and stops the Ag 'roaming') with that single photo a

filament of 3 atoms is produced.

The more photons that strike the grain the denser the filamentary structure becomes.

The image at this time is latent, developing the film amplifies the structure, developer time, strength, agitation all play a part on the

density of the filamentary structure.

Normally, the grains will reach a maximum density of 2.5-3.0 but this is very dependent on developer choice, but grains that reach such

densities are said to be developed to completion as more development time will not yield any more density.

 

Normally maximum resolution is achieved developing slower films (which have thinner and less layers) to a medium gamma say G0.65,

but tone suffers somewhat, so nowadays we normally develop to a lower gamma for increased tonal range with softer working (low

energy low p.h developers) normally metol (Kodak Elon) based.

One silver grain will not totally block light, but there are quite a few in the layers, test this yourself by exposing a film to the sun

developing to completion, and holding up to a light source it hold back light OK :)

 

Clumping is rare in emulsions these days as developers are now pretty low energy low p.h and is not a good way to increase density, in

fact I cant think of a single developer that promotes clumping, most include restrainers to avoid that and give better tone.

 

That is one of the problems with the LL article clumping is not desirable and doesn't give you tone, I find it ludicrous for Reichmann to

suggest that grains 'clump' to form tone like inkjets, that statement shows a gross lack of knowledge and is pretty much a 'made up'

fantasy, which is my problem with the whole article really- its plain dumb.

Mark

[vijay_nebhrajani]

I just had a thought - maybe people are confused about what a "grain" really is. Mark, I have taken the liberty of taking your picture and highlighting what a grain is. The grain is the shape that looks like a triangle with its corners chopped off, kind of like a medicinal tablet (tablet grain; or T-grain).

 

Grain is not those dark clusters that look like scotch-brite fragments. Those are the 'filaments' or 'lattice', but that is not the grain.<div></div>

[vijay_nebhrajani]

Rishi - you took those pictures from a bridge - a sure recipe for camera vibration. Bridges tend to oscillate

naturally at 20-50 Hz, especially metal ones, and there is constant vibration inducing traffic passing over the

combs where two spans join. There is also wind that causes the structure to vibrate, quite like a plucked guitar

string. The fact that you took this picture from a bridge greatly increases the likelihood that what you see is

camera vibration.

 

Also, after you've seen the above picture of a single grain, do you still concur with Reichmann?

[patrick j dempsey]

I didn't read this entire thread, it honestly looks like a bunch of blah blah blah about resolution. In my quip

about resolution, as far as I can tell from all the technical crap I've been reading in the last few years,

diffraction is the key limiting factor in resolution regardless of what media you shoot on, so all that the 100MP

sensor cameras of the future will be resolving over the 10MP sensor cameras or film emulsions of today is

diffraction patterns.

 

Since no-one has mentioned it, you might be surprised to hear what a friend told me about this year's Photokina.

Even with world economies suffering, every vendor at this year's Photokina reported better sales of film this

year than last. Apparently there were even four professional photographers speaking explaining all of the

technical reasons they still shot film either in-lieu or in-addition to digital. You can read into things what

you want, but the horses were at Photokina and the word from horses' mouths was that they were happy with their

sales and that they hope that film is here to stay.

[rishij]

Regarding your grain picture:

<p>

First of all, Vijay, that grain that you highlighted in yellow: is that an exposed, partially exposed, or unexposed grain? Because only the boundary is black.

<p>

Second, I can understand why a clump within the grain may be black, as silver ions are present inside of a grain or crystal. But what's up with the black clump on the upper left side of the grain? That's outside of the crystal. There shouldn't be any silver ions or anything there but the gelatin in between crystals. So what exactly am I looking at here?

<p>

OK, regarding Reichmann:

<p>

Ansel Adams <i>clearly</i> argues against Mark Smith & agrees with Reichmann (or, er, the other way around, chronologically speaking). So, I am going to quote a number of lines from Ansel Adams' book 'The Negative'... Ansel Adams was a pretty smart & knowledgeable guy, so I assume he knew what he was talking about. However, in the face of new evidence and reconsideration, sure, I'm willing to rethink even what Ansel Adams says. Kinda like how I'd rethink the Bible or any religious text in the face of (and by that I mean lack of) any believable evidence. But I digress...

<p>

"I remind you that the 'grain' we see in the print is not the grain itself but the effect of light passing through the spaces between the grain clumps. As with the dots of the printing-press plates (halftones) the grains are themselves <b>of the same density</b>, but the number and size of grains in a given area define the effective density." <i>Page 182</i>

<p>

"Examination of a photographic negative with a magnifier reveals that it is not made up of a continous range of white-to-black values, but that such values are simulated using a controlled deposit of individual black specks. These specks are the <i>grain</i> of the emulsion, the reduced metallic silver deposited when a halide crystal responded to light and 'developed'." <i>Page 19</i>

<p>

OK I think that's enough to convince you that Ansel Adams is (was, R.I.P.) of the Reichmann camp.

<p>

Now, let's turn to Wikipedia (hey, no judging):

<p>

Under 'Silver Halide', we find:<br>

"When a silver halide crystal is exposed to light, a sensitivity speck on the surface of the crystal is turned into a small speck of metallic silver (these comprise the invisible or latent image). If the speck of silver contains approximately four or more atoms, it is rendered developable - meaning that it can undergo development which turns the <b>entire</b> crystal into metallic silver."

<p>

Now, let's turn to this website: http://www.cheresources.com/photochem.shtml

<p>

It says: "In general, as the grain size in the emulsion increases, the effective light sensitivity of the film increases - up to a point. An optimum value of grain size for a given sensitivity is found to exist because <b>the same number of silver atoms are needed to initiate reduction of the entire grain by the developer </b>despite the grain size, so that producing larger grains reaches a point of diminishing returns and no further benefit is obtained."

<p>

Look also at the image there, it shows the entire crystal turning into metallic silver.

<p>

Now, let's turn to this article: http://findarticles.com/p/articles/mi_m1511/is_8_21/ai_63583781/print?tag=artBody;col1

<p>

"If a silver halide crystal already has a light-blackened duster containing a critical minimum of silver atoms--four, typically--that cluster's electric field draws electrons to the silver ions around it, and <b>the whole billion-ion crystal is quickly blackened</b>. All the other crystals, unexposed or barely exposed, the developer leaves alone; the crystals react individually because the gelatin isolates them from one another."

<p>

OK, I could go on and on. But you see my point: Mark Smith & co., you are literally challenging all the literature out there in saying that each grain can exhibit a continuous tone from white-to-black.

<p>

Not that there's anything wrong with that. It's just that you're going to have come at the prior literature with more evidence than I've seen presented in this thread. And I actually WANT to believe you :)

<p>

Rishi

[vijay_nebhrajani]

Answer me this:

 

Suppose I am photographing a source of white light, say sunlight through a ground glass (very bright, uniformly

so). Let's also assume that I don't focus on this backlit ground glass. Then you agree that light through the

lens will hit every point on the film equally - because we simply simulate a diffuse, bright source. Let us

assume I simply use the meter reading and take a picture.

 

Q: What do I get on the negative?

 

A: 18% grey (or 13% or whatever the meter is calibrated to).

 

Light struck every grain equally - and surely there was enough light to ionize at least four Ag atoms in every

grain. This should have resulted, after development, in a black negative because "the whole billion-ion crystal

is quickly blackened".

 

Unfortunately we don't get a black negative but a grey one.

 

How do you explain this anomaly?

[rishij]

More evidence against Mark Smith and for Reichmann:

 

Consult the journal article (well, review): Chemical Reviews, 1997, Vol. 97, No. 1 (actually I could post a link to a PDF of it, but is that legal?). This article gets into the hardcore chemistry of film (& I'm a chemist, well, in a chemistry graduate program anyway).

 

It says: "Absorption of more photons by the same crystal results in electron migration to the atom of silver at the surface [of the grain]. In this way an invisible speck of afew silver atoms adheres to the surface of the silver halide crystal. This speck, composed of more than three atoms of silver, is a latent image site and serves as a conductor through which electrons are transferred from the developing agent to the entire silver halide crystal. Imaged silver halide grains thereby become black metallic silver which has the microsopic appearance of black filaments." (Page 86).

 

OK, evidence aside, let's look at this from a theoretical perspective:

 

In color film, these silver grains must be sensitized to respond to certain colors. Therefore, they have spectral sensitizers adsorbed to the crystals. These are excited by light, then donate an electron to the crystal, which can be picked up by a silver ion to make silver metal. So, if you're saying that the latent image actually determines how black the crystal will become, you have to have a rather extensive range of silver ions that become silver metal... the total number of such silver ions that become silver metal then determine the 'dynamic range' of ONE crystal. What is that dynamic range, Mark Smith? 0-1000? 0-10,000? Let's say 0-1000, for argument's sake. Then, there also have to be at least 1000 molecules of spectral sensitizer (ring-containing molecules that are much bigger than a silver ion) adsorbed to the crystal, all ready to donate an electron to a silver ion. Unlikely. UNLESS you have a source of electrons that can regenerate the spectral sensitizer after it has donated an electron upon being excited by a photon.

<p>

Next, the number of silver ions reduced to silver metal (mostly at the boundaries of the crystal, since those are the areas most accessible by photons), has to determine the rate of the development (or reduction) reaction. So, based on whether 4-1000 silver ions were reduced to silver metal per grain, 1-10 billion silver ions would be reduced to silver metal within each crystal. Does the chemical reaction rate really depend on exactly how many silver ions were reduced to silver metal to form the latent image? I don't know... some hotshot chemist needs to answer this question. But it seems pretty unlikely to me ESPECIALLY given that all these technical papers suggest that the reduced silver metal molecules migrate to the electron rich site (silver or gold sulfides) on each crystal. I.E. the crystals are doped with silver or gold sulfides; but, only ONE such site exists per crystal. Upon exposure, the reduced silver molecules cluster at this site. Are you telling me that the number of reduced silver molecules clustered at this site, per grain, determines how black/gray the crystal becomes, and that a whole range of tones can be represented just by how many reduced silver atoms cluster at this site? What is the maximum number of reduced silver atoms that can cluster at this site? What are the maximum number of electrons that can be donated to silver ions to form silver atoms upon simple light exposure? And then can this simple signal be amplified to create all the tones between white and black? Given that the reaction proceeds because of the increase in conductance of silver metal vs. silver ion, can you really generate all the tones between white and black just based on how many such conducting silver atoms are present at the ONE site on each crystal?

 

THESE are the questions that need to be answered to prove Mark Smith's hypothesis correct. All I've seen so far is a bunch of handwaving.

 

I admit I need to read more the come to a definite conclusion myself, because both arguments still make some sense to me, but I'm still leaning toward the Reichmann camp. Mark, please convince me. I want to be convinced :)

 

Rishi

[rishij]

Vijay,

 

You yourself wrote: "Grain size is variously stated as being 1-10 microns (micrometers, 10^-6 m). 200 cpmm [the resolving power of a lens] means that each individual line is 1mm/200 or 5 microns across. if you want this to be properly resolved, you need grain size less than 5 microns."

 

If the smallest feature a lens can resolve is 5 microns, and the finest grain is considerably smaller than 5 microns, then each grain will NOT be hit by a photon in your scenario above. Therefore, you will not get a black negative. Instead, you will only get certain grains falling with this 5 micron region exposed and 'blackened'. Let's also not forget that film does not consist of just 1 layer. Color film itself has at least 3 layers, so, for gray light, the number of photons exciting any given spectral sensitizing molecule is cut in 3.

 

These are all great arguments though. Let's figure this out!

Rishi

[peter clark]

I will always use film, for my work which is panoramic landscape, digital will never be able to replicate the quality achieved

by using film, and I could never print the sizes I do with film if I used digital. Well I could but the image would fall apart.

 

Use what works for you and the work you are doing.

[vijay_nebhrajani]

Rishi, surely you aren't trying to imply that the image from the unfocused lens has spots of light areas of 5 micron diameter and the rest is dark. That isn't how light works, but OK, take that hypothetical lens off, and point the camera at the sun. Will that be uniform enough light distribution for you?

 

And forget color film for now - lets just concentrate on B&W for the sake of simplicity.

 

Also please answer me this: why does increase in development time increase negative density? After all, for a binary grain, amplification of four Ag ions should result in a black grain regardless of development time, agitation, temperature etc., right?

 

To everyone else: I know that this is getting pretty arcane, and most photographers don't want to be bothered by details about physics and chemistry, but this is turning out to be quite the in-depth discussion about these issues. I haven't seen much of such debate elsewhere, so I personally feel that it adds value to photo.net. If this has started bothering you, please ignore this thread, or maybe the mods could consider moving it to another area (although when this is done, the focus and continuity of the thread are generally lost).

 

I am acutely aware that the art of photography is what is supposedly sublime, and these technical topics are third-rate wasters of time and energy, but please, please let us continue. Some of us enjoy understanding what we are doing. Patrick Dempsey, if you had read this thread in some more detail, you would have seen me celebrating that the digital-film war is about to be over because both media are no longer the limiting factors, so finally we can go about photographing with either without feeling inferior.

[photo_smith]

Rishi

 

The argument you posted backs up mine and Kodaks claims that grain is filamentary:

"It says: "Absorption of more photons by the same crystal results in electron migration to the atom of silver at the

surface [of the grain]. In this way an invisible speck of afew silver atoms adheres to the surface of the silver halide

crystal. This speck, composed of more than three atoms of silver, is a latent image site and serves as a conductor

through which electrons are transferred from the developing agent to the entire silver halide crystal. Imaged silver halide

grains thereby become black metallic silver which has the microsopic appearance of black filaments." (Page 86)".

 

Look what it states:

That a grain is made up of microscopic black filaments, Reichman suggests that the whole grain is TOTALLY black

either on or off, binary, it has two states.

Here is my explanation.

http://photo-utopia.blogspot.com/2007/10/chumps-and-clumps.html#links

 

We know from the way film works that when a single photon strikes an AgBr+ (the structure is positively charged) that

the photon energizes the Ag atom and separates it from the Bromide atom. The Ag atom then moves towards the

sensitivity speck to form a latent structure which is later amplified in development.

The filamentary structure can develop from 3+ atoms up to the entire grain and all those states between.

A normal negative has states between 0.10--3.0 density.

 

You reading of the paper puzzles me am I right in that you are assuming that there is only 1 sensitivity point?

In reality each grain has many hundreds of thousands of these specks, the silver is not black metallic until after

development, I think you are slightly confused about the stucture of the individual graind before development.

 

If Reichmann is right (which he isn't) grain can be not be transmissive it must totally block light (black binary 1) with the

clear parts transmissive (clear binary 0)

But as the grains are shown to be filamentary they block varying amounts of light depending on how many photons strike

the grain (more photons denser structure)

Grains are basically photon counters, if they were binary 1 photon would have the same effect on a grain as a million.

 

So far from your PDF agreeing with Reichmanns position it actually undermines it.

If you need more proof borrow a copy of "The theory of the photographic process' by Mees and James from your library,

or alternatively go to APUG and ask you questions on the emulsion forum, the person there called Photo Engineer (Ron

Mowery) is the man I put my article to for proof reading, he is pretty much the film expert.

[photo_smith]

Rishi

 

Just an aside when you quoted all those sources as flying in the face of 'my argument' you are forgetting one very important fact, I'm not saying the grain is not totally made up of black metallic silver, obviously it is.

To most people they are just black specks, but upon extreme magnification you will see it is the silver ATOMS not the

actual grain that is SOLID black, if film can be proved to be filamentary it passes light in varying amounts.

 

You asked about my picture:

 

First of all, Vijay, that grain that you highlighted in yellow: is that an exposed, partially exposed, or unexposed grain?

Because only the boundary is black.

 

So ask yourself if that grain is only black around the edge which of the binary states is it?

If grain is either black or clear it can be binary but as soon as a third state exists the binary argument is invalid.

 

That picture shows a Kodak T grain emulsion. T grain emulsions have the sensitivity centers placed around the edge,

these are called "ears" If you go to the post on my blog you'll see at the top of my post a photo of Ilfords Delta structures which are hexagonal, each point of the hexagon having a 'sensitivity ear'

 

If you can understand that grain can have many states the yellow highlighted grain shows a partially exposed grain one

that has only developed around the sensitivity areas- if grain is binary that can't happen.

[rishij]

Vijay,

 

You write: "Also please answer me this: why does increase in development time increase negative density? After all, for a binary grain, amplification of four Ag ions should result in a black grain regardless of development time, agitation, temperature etc., right?"

 

Simple, because increasing development time allows the developer to start reducing silver ions within crystals that were unexposed. Chemistry is all about reaction rates. The reduced silver atoms in exposed crystals serve as catalysts, allowing the reaction to initiate and thus proceed rather quickly. Crystals without reduced silver atoms will still be reduced, but the the probability of this reaction starting within any given finite time span is less if there are no conducting silver atoms near the crystal surface to begin with.

 

That takes care of that part of your argument.

 

Now, you say: "Rishi, surely you aren't trying to imply that the image from the unfocused lens has spots of light areas of 5 micron diameter and the rest is dark."

 

You're right, I'm not trying to imply that at all. I'm saying that if the resolving power of the lens is only 5 microns, then sometimes this lens will direct photons towards one portion of this 5 micron area, and sometimes to another portion of this 5 micron area. If 5 microns really is the limit, then it can't direct photons, in any coherent fashion, to any one portion *within* this 5 microns; rather, it directs photons randomly within this 5 micron region. Thus the exposure of every grain within this 5 micron region will be a stochastic process dependent upon *time* of exposure. Since the time of the exposure will be adjusted to as to result in 18% gray for this particular feature being recorded, not all grains within this 5 micron region will be exposed within the limited time frame of the shutter speed.

 

Rishi

[vijay_nebhrajani]

<i>Simple, because increasing development time allows the developer to start reducing silver ions within crystals

that were unexposed.</i>

<p>

Oops. Now the developer invents the image does it? It can't develop <i>unexposed</i> grains. If it could, there

would be no conceivable relationship between the optical image formed by a lens and what the "film" recorded.

<p>

<i>Since the time of the exposure will be adjusted to as to result in 18% gray for this particular feature being

recorded, not all grains within this 5 micron region will be exposed within the limited time frame of the shutter

speed.</i>

<p>

If the light is continuous, i.e., there are the same number of photons striking a unit area, and if the grain

distribution is random, how do some grains get hit by photons and others not? I mean for film to behave in the

way you describe, it has to be <i>sentient</i>. Besides, forget the lens - just worry about exposing the film to

constant luminance.

<p>

This is what I mean by reductio-ad-absurdum. Assuming grain is binary results in absurdities, film resolution

only being one of them.

[rishij]

Mark,

<p>

You write: "You reading of the paper puzzles me am I right in that you are assuming that there is only 1 sensitivity point? In reality each grain has many hundreds of thousands of these specks, the silver is not black metallic until after development, I think you are slightly confused about the stucture of the individual graind before development.

<p>

By 'sensitivity point', I mean that silver or gold sulfide which serves as an electron trap. That is, a photon strikes the crystal, liberates an electron from a halide ion, and this electron is trapped by this silver or gold sulfide electron trap. Eventually, a silver ion travels near this electron trap (the site of the gold or silver sulfide; only one such site exists per crystal), and then picks up that electron from the silver or gold sulfide, reducing it to silver metal.

<p>

Mark, I appreciate the clarification. But you need to decouple:

<p>

<ul>

<li>Reduction of silver ions due to photon excitation</li>

<li>Reduction of silver ions due to a combination of developer and alread reduced silver atoms near the silver crystal surface</li>

</ul>

<p>

Specifically, how many silver ions can be reduced to silver atom, per crystal, by simple photon excitation? 100? 1000? 1 million? Because this'll determine the overall dynamic range of each crystal in your argument. Your argument also assumes that the number of silver ions reduced to silver atoms at each crystal site determines the total number of silver ions reduced within the crystal upon addition of developer. But if all the silver ions reduced to silver atoms upon exposure, per crystal, cluster at one site within each crystal (where the silver/gold sulfide resides), then how can you correlate the rate of reaction (reduction) with the number of alread reduced silver atoms at this site? I need some chemical mechanism rationale here...

<p>

Rishi

[rishij]

"Oops. Now the developer invents the image does it? It can't develop unexposed grains. If it could, there would be no conceivable relationship between the optical image formed by a lens and what the "film" recorded."

 

Um... yes it can. It doesn't happen because it's a function of reduction potential of the developer and/or time. Typically both are adjusted such that this does not happen. If you had a really strong reducer, sure, it'd reduce the silver atoms in the unexposed grains.

[rishij]

"If the light is continuous, i.e., there are the same number of photons striking a unit area, and if the grain distribution is random, how do some grains get hit by photons and others not?"

 

Because the lens is trying to focus photons on a given area. For argument's sake, let's reduce this 'area' to an absurdly small area. That is, the area that you are sampling to see if a photon hit is absurdly small. Since the lens can't perfectly focus photons onto any area smaller than 5 microns, do you think it'll always focus said photons to a tiny patch, say, 1nM in diameter? No, because it might focus a photon onto a 1nM in diameter area somewhere else within this 5 micron area.

 

On top of that, this'll be a time dependent process. Over enough time, yes, probably all of the areas within this 5 micron area will be hit by a photon or two. But not within a finite time span.

[photo_smith]

Rishi

 

By 'sensitivity point', I mean that silver or gold sulfide which serves as an electron trap. That is, a photon strikes the

crystal, liberates an electron from a halide ion, and this electron is trapped by this silver or gold sulfide electron trap.

Eventually, a silver ion travels near this electron trap (the site of the gold or silver sulfide; only one such site exists per

crystal), and then picks up that electron from the silver or gold sulfide, reducing it to silver metal.

 

You speak as if there is only one 'sensitivity point' per grain there are many, just looking at a T grain you can see these

points "called ears' round the edges each ear can have many sulpher/gold.

 

Specifically, how many silver ions can be reduced to silver atom, per crystal, by simple photon excitation? 100? 1000? 1

million?

1 photon will give a structure of 3 atoms of black silver upon development.

 

The silver atoms are converted to metallic silver by the developer, it magnifies the latent image.

All this isn't MY theory it is from Kodak engineers, Mees and James etc It's reichmanns theory that flies in the face of

established knowledge.

 

I'll ask You a question.

If film is binary, and tone is formed by 'dithering' how and what causes these wholly black grains to move together to

form tone, after all they must do this upon exposure, the grains themselves would have to decide where to position

themselves to form that tone.

Reichman states that it is like an inkjet dither- well a Rip does that (a very complex piece of software) how do the grains

know where to position themselves.

But the binary argument is a binary one soon as you find one grain that is neither black or clear it is invalid, I'm still

waiting for your explanation of how a binary system can have a third state of partially exposed.

 

Before accepting Reichmans 'new theory' you need to prove how these states occur.

[vijay_nebhrajani]

<i>Um... yes it can. </i><p>

Not in the sense of photographic development, it can't. If it could I could process a roll of unexposed film and

get back perfect portraits of my children.

<p>

As for your second point, I've repeatedly said, get rid of the lens, just point the camera sans lens at the sun

and shoot. No femtometric focusing issues then, are there?

<p>

Mark raises the same issue as I - "how do the grains know where to position themselves?" - if they did, film

would be <i>sentient</i> - like I said before.

<p>

Once again, don't confuse resolution with tonality. A grain can have many shades of gray, but it can't represent

more than a unit of resolution information, a pixel, if you will.

<p>

At the risk of repeating myself, please look at Mark's picture - the tablets are the grains, not the black

clusters. Those tablets are supposedly why Kodak named their technology "T-grain".

[danielleetaylor]

I'm not going to try quoting/responding to all the text written while I've been gone. I will merely add some

comments based on what I've skimmed...

 

* Rishi - thanks for replying and posting the Ansel Adams quotes. I'm a bit perplexed as to why Vijay and Mark

ignored those. They are not arguing against Michael Reichmann, but against Ansel Adams. I'm dying to see a quote

from someone as respected as Adams stating that he is wrong and silver comes in shades of gray.

 

* Silver halide crystals vary in size, shape, and orientation within a single emulsion. That means they also vary

greatly in sensitivity. In the diffused light, middle gray exposure thought experiment Vijay describes, not all

of the crystals would be exposed sufficiently. The illumination of the frame may be uniform, but the crystals are

not by any means uniform in their sensitivity to that light. (Sorry Rishi, but I have to disagree with your

explanation. In Vijay's thought experiment, the exposure should be uniform. It's the crystals which are not uniform.)

 

* Along those same lines, I seem to recall that the degree of exposure of a single crystal does influence how

completely AgBr is converted to Ag and, therefore, the resulting density. That doesn't matter much in Vijay's

thought experiment, but matters in normal scenarios.

 

* There is no AgBr left after fixing. I'm not entirely sure Vijay realizes this given his comment about a grain

having billions of "possible states" based on the number of atoms/molecules in the grain, and his comment that

there are "partially exposed" grains.

 

* Vijay - yes, developer will reduce unexposed grains to silver given excessive development time. This is taught

in any basic photography class that deals with B&W materials. Why else do you think development time is so

critical? Please note the following comment and source:

 

"Prolonged development would, of course, increase overall density through the development of unexposed grains (fog)."

http://www.cheresources.com/photochem.shtml

 

If you let unexposed film sit in developer longer than the recommended development time it will continue to fog

until, at some point, you have a black frame. (I don't know specifically how long it takes to reach that point as

I've never tried it. I try to avoid fogging my B&W film.)

 

* Grains do not "move together" to from tones. Dithered printing systems use software to precisely position the

dots of ink because those dots are large relative to the medium as contrasted with, say, grains in film, or

grains on B&W paper. By controlling the patterns more tones can be produced from those dots, and with fewer

perceptible patterns or defects.

 

* Again, there is no such thing as a "partially exposed" grain after fixing. There is only silver.

 

* Again, Reichmann's theory is not new. It was Adam's theory.

 

* Just thinking about the language of film should reveal the truth here. We measure and discuss the density of

silver in film. We do not measure or discuss the "tone" of silver, as if silver came in a range of tones.

 

No offense, but I think perhaps some people in this thread need to read The Negative and spend some time in a

darkroom.