Wow - read this re: Film versus Digital debate!

[vijay_nebhrajani]

<i>Vijay: "Do you see opaque silver specks on a gray background because of the ND filter? No."

<p>

Bernie: "Actually, that's exactly what I would expect to see. How could it be white? Our eyes don't act like an

auto levels command in photoshop."

</i>

<p>

Wrong.

<p>

Try this: darken a room, and open this page on your computer. These words are black on a white background. Easy

enough to observe, right?

<p>

Then turn down the brightness of your computer screen. Wait for your eyes to adjust. What do you see? Black words

on a gray background?

<p>

Please try this and post the result of your experiment.

[rishij]

Bernie:<i>"DLT, Rishi, at least we have had a small win as we finally have Vijay talking about RESOLUTION. It's

taken a while, but we are making headway. He has agreed that it takes far more than 1 grain to respresent VISIBLE

tone, so already we have him backtracking a little."</i>

<p>

No I don't think he has. He's saying that 2 grains can leave behind silver specks representing 1 line pair. So I

don't think he's backtracked one bit. I also have a hard time arguing with him now, because he may be right about

that. I mean, some films can resolve near 100 lpmm but with a very low contrast of response (like below 10%

response), right? So I guess that may mean that two areas of film equivalent, each, to the size of one grain,

side by side, could vary ENOUGH (& by that I mean *very little*) in macroscopic tone

to discern a line from absence of line... but the tone difference would be very small, right, because one grain

(or, more correctly, the area of film from where one grain used to reside)

by itself can't make that many tones. Which is why a larger area, the size of many more grains (like 500 grains),

can create many more tones and therefore have a 100% response (or a huge difference in density - of silver

filamentous growths - between the area of the film representing the black line and the area of the film

representing the absence of a line)... Oh! Actually that makes a lot of sense. I think I just figured it all out!

<p>

This explains why the film resolution may be 35MP for black and white lines (high contrast), but more like 12MP

for 'real-world' contrast.

<p>

OK guys, let's ponder on this theory a bit.

<p>

I do think Vijay has brought up some very valuable and believable points of late, after I scolded him for not

responding to some of my burning questions :)

<p>

Um, seriously, let's take Vijay's last few posts very seriously. I think he's making some very good points. I

need to ponder a bit more, but DLT & Bernie... let's take a step back and think about this as both a 'Vijay's

Analog' (contributes much less, I believe) AND half-tone process.

<p>

Rishi

[rishij]

<i>"Try this: darken a room, and open this page on your computer. These words are black on a white background. Easy enough to observe, right?

<p>

Then turn down the brightness of your computer screen. Wait for your eyes to adjust. What do you see? Black words on a gray background?

<p>

Please try this and post the result of your experiment."</i>

<p>

Vijay's right. Our eyes adjust to make the brightest specular object 'white'... in fact, I believe that's also how we white balance.

<p>

Rishi

[bernardwest]

<i>Vijay Nebhrajani , Nov 16, 2008; 08:57 p.m.<p>

 

Vijay: "Do you see opaque silver specks on a gray background because of the ND filter? No." <p>

 

Bernie: "Actually, that's exactly what I would expect to see. How could it be white? Our eyes don't act like an auto levels command in photoshop." <p>

 

Wrong.</i><p>

 

You are kidding, right? Let's play a little game called <i>reduction ad absurdum</i>. What happens when you put a ND filter on which has a light transmission of say 10%? You say white. So if 90% light transmission still equates to a 'white' image, then clearly so should a 80,60,40,20,10,5,2,1,0.5%..... Please explain this little bit of magic?

[bernardwest]

Careful Rishi.... Ascend out of Vijay's microscopic realm, and come join me in the real world. Think of a real world example. In fact, I experienced a good one yesterday afternoon during a big storm. I was out driving when this storm hit. The brightest object in my vision most certainly wasn't white. It was the <b>very</b> dull grey/green sky and cloud that had descended all around me. Infact, here's another one. Go stand out in your backyard at night. Find a really dark section of your yard that isn't illuminated. Do you see any white here?

[rishij]

Bernie,

 

Our eye-brain system sees relative differences in tone/luminosity within the field of view.

 

Rishi

[rishij]

Bernie,

 

That's why I said 'specular', which is rather different from the diffuse light you saw from the clouds.

 

Additionally, this adjustment our eye-brain system performs (or interpretation is probably a better word) only works within certain limits.

 

Which is why when you step into a darkroom, red doesn't become white.

 

But within the field of view of the microscope, Vijay might have a point.

 

Additionally, I still maintain that the number of tones possible by an area of film the size of 1 grain is very small. Not the thousands of tones that Vijay claims. If that were the case, then the 'response' of the film wouldn't be so low at 100 lpmm. Right Vijay? How do you account for that?

 

Rishi

[vijay_nebhrajani]

<i>Bernie: Do you really take us for being this stupid? It borders on offensive. Perhaps your intent isn't to win

us over, but to win over casual observers of this thread who might not be following along with all the lines of

reasoning. If you can't see something, that certainly doesn't mean it does not exist. None of us are even coming

close to saying that. What it means is this: If you can't see it, it can't contribute to the VISUAL phenomenon

that is resolution. Please stop the condescending debating tactics. It does nothing to add to whatever

credibility you have left.</i>

<p>

I called that argument stupid. Why do you choose to identify yourself as such? Sorry for being offensive - I'm

quite exasperated at this moment that any rational argument I may provide is met with an observational data

point, which could be mistaken or limited in scope. Instead of verifying that the observation has no flaws or

limitations, you choose to take that observation as the final truth and smack down any rational argument that

says it may not be so.

<p>

You think I want to win you over or win over other observers of this thread? I don't want to. Wait a minute - I

don't see anyone else supporting me even now, and for all my arguments, nobody has been won over yet. This is not

about taking sides. If you want to participate in a process to find the truth, cool, if not that's OK. You can

continue on your political tirade like "we are winning" or "we have been saying this". You think that is not

simply a case of taking sides? What is <i>your</i> stand? What is <i>your</i> theory? Put forth your arguments

and points clearly so I can either agree with you or disagree with you. Put them forth in bullet points please.

<p>

If you can't see it, it can't contribute to the VISUAL phenomenon that is resolution? What exactly do you mean by

this? Your terminology is quite vague at this point. Can you see 50 line pairs per millimeter on 35mm film? Your

naked eyes can't resolve that much but it exists alright. At a contrast ratio of 1.6:1 no less. Can you see

atoms? Bacteria? Viruses? Bits and bytes?

Electric current? Yet they contribute to all manner of physical phenomena. So what do you mean? Is a visual

phenomenon only that thing that you can see with your naked eyes? Is it that which you can see only with a 400x

microscope? Is it that which you can see with an electron microscope? At what point do you say something is not a

"visual phenomenon"?

<p>

Bernie, please get your story and your terminology straight - I don't mean to offend you personally - I'm sure

you're a fine guy and like Daniel said, I'd gladly buy you a beer. I don't mean to sound condescending either. If

I sound exasperated, it is because I am, but I don't mean to be condescending or insulting. Sorry about that again.

[rich_evans]

Bernie: Absolutely not, but, that's for another thread.

 

OK - I overstated it, I should have said "...more pixels per unit area...", and while that may be fodder for another thread, it is obtusely related to this nano-resolution thread.

 

And I've still not seen anyone address resolution as a function of diffraction at this scale. I guess I'll have to dredge up some old papers and see if I can postulate something that may help explain - or not - gray vs white vs black as it relates to diffraction effects. ;-) --Rich

[ndelamerced]

I've been following this thread and I would like to give my 2 c worth. I believe that at atomic or even molecular level, and atom

or molecule is binary. The Ag is either an atom or molecule. In film, high ISO means that there are fewer layers of light

sensitive chemicals in the emulsion and more layers and more light sensitive chemicals in low ISO. Any unreacted chemical,

meaning light did not touch it is washed away in BW film. In short, each light sensitive molecule is a sensor and if the size of a

digital sensor is 5 microns, how many molecules do you think will fit in that space? Can Vijay or Mike give an answer to this?

My apologies if my thinking is too simplistic. Thanks.

[vijay_nebhrajani]

<i>Bernie: "You are kidding, right? Let's play a little game called reduction ad absurdum. What happens when you

put a ND filter on which has a light transmission of say 10%? You say white. So if 90% light transmission still

equates to a 'white' image, then clearly so should a 80,60,40,20,10,5,2,1,0.5%..... Please explain this little

bit of magic?"

</i><p>

Did you try the experiment I requested you to? You chided me for "hypotheticals" (right?), so don't bring one up

yourself. Here is a simple, practical experiment - just try it for yourself. Its not like you need specialized

equipment or anything or more than a couple minutes of your time.

[rishij]

Vijay:<i>"Wait a minute - I don't see anyone else supporting me even now"</i>

<p>

Hmmm... read my last 2 posts, buddy.

<p>

Rishi

[bernardwest]

<i>This is not about taking sides</i><p>

 

Unfortunately, this is what it has become. One the rational side we have Rishi, Daniel and Myself, who have had an interesting discussion, and I can confidently say I have learnt quite a bit from these two. On the otherside, we have <strike>Mark's</strike> uhh... Devil's adovocate, whose only purpose seems to be throw mud at our argument and hope it sticks. I'll put forth my argument (which I have elucidated many times anyway, and in form that's readily interpreted, unlike your convoluted psuedo-logical arguments) in bullet points if you do likewise. You go first, eh, because everyone else knows where I am arguing from. But hey, why not try yet another bolded capitlized summary of my argument: <b>TONAL RESOLUTION</B>. If you can't work out what my argument is from that and all i've written, then there is not much hope for you.<p>

 

<i>I'm quite exasperated at this moment that any rational argument I may provide is met with an observational data point, which could be mistaken or limited in scope.</i><p>

 

The problem is that many (if not most) of your premises in the raa's that I can deconvolute AREN'T rational.

<p>

<i>If you can't see it, it can't contribute to the VISUAL phenomenon that is resolution? What exactly do you mean by this? Your terminology is quite vague at this point. </i><p>

 

I can't possibly think how I could make this any clearer. I'll put it in simple sentences. Resolution (as it relates to photographic prints) is a VISUAL phenomenon. It is thus, because resolution (as it relates to photographic prints) is concerned with LIGHT waves. Are you seeing a connection here? Visual...Light? If we were talking about Electron Microscopy resolution, then we aren't talking about visible light, and therefore not a VISUAL phenomenon.<p>

 

<i>Can you see atoms? Bacteria? Viruses? Bits and bytes? Electric current? Yet they contribute to all manner of physical phenomena. </i><p>

 

I can see atoms, when they clump together to form VISUALLY perceptible objects. Hang on... that sound vaguely familiar to what I've been arguing all along... when developed silver clumps together is becomes VISUALLY perceptible, and hence, at that point, can contribute to the VISUAL phenomenon of resolution. As for you bits and bytes and electric current, no I can't see them (except for say lightning), but once again your point is lost on me. A bit and a byte is a CONCEPTUAL phenomenon. Electricity can never be a VISUAL phenomenon (unless you talk about lighting and sparking). These two examples don't even fit in with your reasoning (or lack thereof).

[bernardwest]

<i>Did you try the experiment I requested you to? You chided me for "hypotheticals" (right?), so don't bring one up yourself. Here is a simple, practical experiment - just try it for yourself. Its not like you need specialized equipment or anything or more than a couple minutes of your time.</i><p>

 

You are becoming more and more farcical by the minute. Why is it ok for you to patently ignore objective evidence and continue to pseudo-logicise, but we have to play by a different set of rules. Let's see you take some of your own medicine. Like you say for reductio ad absurdum, logic is all that matters. Let's here what you think the flaw is in my logic. Let's see if you really can play this game.

[vijay_nebhrajani]

Yeah, Rishi - saw them after I posted. Thanks, and thanks for agreeing that at least some of what I say makes

sense, but most of all, thanks for recognizing that I hadn't backed up at all.

 

I will put forth some other explanations in a bit - actually spent the whole day at work dealing with an

engineering crisis. Will probably need to be here way past 3 AM. Sigh. Warmest November weather here in Silicon

Valley in more than a hundred years, and I'm working. Sad.

[bernardwest]

<i>Rishi Sanyal , Nov 16, 2008; 09:23 p.m.

 

Bernie,

 

That's why I said 'specular', which is rather different from the diffuse light you saw from the clouds.</i><p>

 

But at what point does a 'specular highlight' change to a 'grey tone' and vice versa? It's a nonsensical argument in the scope of this thread.

[rishij]

You know what I realized? When Vijay attempted to be noble and say that 'in the worst case scenario', film grain would be 5 microns in size (as opposed to being smaller), and then calculated 100 lpmm resolution, he was actually doing himself a favor. Because if an area of film the size of one grain could take on 0 to thousands of tones, and grains were smaller than 5 microns (e.g. 2 microns), then film should be able to resolve much more than 100 lpmm. Which is not the case. Nice one... tricky tricky, Vijay!

 

Actually, I think the crux of my argument now is that the amount of light that an area of film the size of ONE grain can block is very small... therefore, if an area of film the size of one grain is completely developed, and the one next to it is completely undeveloped, the macroscopic difference in tone between these two areas will be very small; hence the really low response (contrast) at 100 lpmm. This is why you NEED a halftone process, with lots of silver clumps, to effectively block light and create tones other than clear.

[rishij]

I still think that most filamentous growths from one sensitivity center would be largely opaque, because no one's showed me an EM image of a filamentous growth with large gaps. Correct me if I'm wrong.

 

I think in the end this is gonna get down to an argument of which effect matters more? Filamentous growths having tones, or the halftone process involving filamentous growths clumping into larger opaque particles that can block light effectively as opposed to the clear areas of film (i.e. opaque filamentous growths/clear film base ratio has more effect on tone than opaque filamentous growths/semi-opaque filamentous growths). That last point is really subtle.

 

What do you think Vijay? Bernie, take a deep breath, and tell me what you think of that? DLT?

 

And where Reichmann at??

 

Rishi

[bernardwest]

<i>Vijay Nebhrajani , Nov 12, 2008; 11:51 a.m.<p>

 

As for resolution, if a grain can have continuous values</i>[which Vijay asserts]<i>, then why - <b>the</b> largest of

these grains would determine the smallest resolvable feature size</i><p>

 

First it was one grain, now it is at least two. Who knows where he will finish up Rishi. This was made to point out

that VIjay has been migrating towards our argument that multiple grains are needed to create visual tone.

[vijay_nebhrajani]

<i>Rishi: You know what I realized? When Vijay attempted to be noble and say that 'in the worst case scenario',

film grain would be 5 microns in size (as opposed to being smaller), and then calculated 100 lpmm resolution, he

was actually doing himself a favor. Because if an area of film the size of one grain could take on 0 to thousands

of tones, and grains were smaller than 5 microns (e.g. 2 microns), then film should be able to resolve much more

than 100 lpmm. Which is not the case. Nice one... tricky tricky, Vijay!

</i><p>

I actually inversely (and quite correctly) correlated grain size and resolution - by pointing out that the

largest crystal size limits resolution, even in the presence of smaller crystals. Perhaps the smaller crystals

could give you better resolution, but not if there were even a small percentage of larger crystals. Silver halide

crystals do what is called information compression. Even if 10 cpmm worth of information fell on a crystal, the

crystal wouldn't record all the information. It would integrate (as in the mathematical function of integral

calculus) the amount of light over time and compress that information and store it as a proportional number of

ions, waiting for later

amplification by the development process.

<p>

Because of this information compression and later amplification, the original information of light distribution

on a single crystal is lost. Hence the largest crystal, which integrates a larger amount of optical information

over time, performs the most information compression, i.e., loses the most information, i.e, contributing the

most to limit resolution.

<p>

How is this "tricky, tricky"?

<p>

By the way, I still have a little explanation left in me - especially about partially transmissive silver specks

- but it'll have to wait for a bit. Still busy at work.

[vijay_nebhrajani]

Oh, I realized - I seemed to imply that the one single largest crystal on one corner of a film area the size of a football field would be enough to limit resolution on the farthest corner of that field.

 

Sorry. In our language (science, math) when we say something like that it generally means that there is enough frequency of occurrence of these large crystals - and that they are evenly distributed.

[danielleetaylor]

<P>Vijay,</P>

<P></P>

<P><i>Finally, finally, someone gets my argument.</i></P>

<P></P>

<P>Your argument has always been understood. Your problem is that your theory predicts evidence <b>which does not

exist, much less exist in the quantity needed to form the basis of tone in B&W photography.</b></P>

<P></P>

<P><i>Go put a neutral density filter in your microscope between the film and the objective. That neutral density

filter has tone, right? See the grain through the microscope. Do you see opaque silver specks on a gray

background because of the ND filter? No.</i></P>

<P></P>

<P>Yes. Heck, the background without a ND filter doesn't look "white" because of the effect of the film

base/gelatin. With one that background would be even more gray. But you are right that the silver specks will

remain opaque.</P>

<P></P>

<P><i>Silver specks, too small to see under an optical microscope - attenuate light.</i></P>

<P></P>

<P>Not by themselves they don't. Do you understand why an ultra thin coating of solid matter will appear

transparent? It's because there aren't enough atoms grouped together to attenuate the light to any observable

degree. Again, you can't image a feature smaller than the wavelength of radiation you're using to perform the

imaging.</P>

<P></P>

<P>When solid matter appears solid, it's because the group of atoms <i>as a group</i> is large enough to

attenuate light to an <i>observable degree.</i> You seem to think there's this whole, rich realm of intermediate

gray tones as a silver halide crystal is developing out. There's not. As the cluster of silver atoms crosses that

threshold between unobservable (i.e. smaller or thinner than the wavelength of light) and observable, it pretty

much crosses over into opaque because that's the nature of the atoms/natural atomic arrangement we're dealing

with. Silver is, by nature, opaque.</P>

<P></P>

<P>Man may be able to direct the arrangement of silver atoms to produce a translucent coating, but this very

precise and specific arrangement is not observed to occur in the random, natural silver structures in B&W film.</P>

<P></P>

<P>We can go round and round about the particle/wave nature of light, the imaging of matter near the threshold of

light's wavelengths, the structure of silver filaments in film, etc, etc for another thousand posts. None of it

will matter because we will ultimately still arrive at your reductio-ad-absurdum: that grains are both observable

as gray and not observable as gray.</P>

<P></P>

<P><i>If they attenuate light they act like tiny ND filters, attenuating light and contributing to tone in the

damn final print.</i></P>

<P></P>

<P>Anything observable in the final print will be more easily observed under magnification. But we do not observe

any gray specks of silver. They cannot both be observable and not be observable.</P>

<P></P>

<P><i>"I can't see something so it can't exist and can't possibly affect anything else that I can perceive

either."</i></P>

<P></P>

<P>Nobody ever said sub-wavelength deposits of silver did not or could not exist. <i>But by definition if I

cannot see something, then that something cannot be affecting my visual perception. To affect my visual

perception is to be seen!</i></P>

<P></P>

<P><i>I can't see gray moons so the color of the surface of the moon must be white.</i></P>

<P></P>

<P>I don't know about you, but when I look at the moon I can see gray textures on the surface without the aid of

a telescope. If I were to spend 20 minutes in a pitch black room then walk outside and stare at the full moon at

night, for a moment it might appear white, but only for a moment until my eye adjusted to the proper exposure.</P>

<P></P>

<P><i>But so do those non-opaque ones, invisible under an optical microscope</i></P>

<P></P>

<P>It is absurd to suggest that something can be seen at 10x but not seen at 60x or 400x.</P>

<P></P>

<P><i>I'm quite exasperated at this moment that any rational argument I may provide is met with an observational

data point, which could be mistaken or limited in scope.</i></P>

<P></P>

<P>I'm quite confident the observation is fine since I can confirm the halftone nature of B&W film using a

microscope, a grain focuser on a projection from an enlarger, and a really large print from small format film.</P>

<P></P>

<P><i>Instead of verifying that the observation has no flaws or limitations, you choose to take that observation

as the final truth and smack down any rational argument that says it may not be so. </i></P>

<P></P>

<P>In science rational arguments are smacked down by observation. Is anything about quantum mechanics "rational"

to us? Heck, if reasoning trumped observation I could discredit all of quantum physics right now. Burn all the

textbooks and papers and go back to teaching classical physics because the quantum realm is not "rational" to the

human mind.</P>

<P></P>

<P><i>If you can't see it, it can't contribute to the VISUAL phenomenon that is resolution? What exactly do you

mean by this? Your terminology is quite vague at this point. Can you see 50 line pairs per millimeter on 35mm

film? Your naked eyes can't resolve that much but it exists alright.</i></P>

<P></P>

<P>Let's play with that example. A viewer is asked to observe 35mm film from 5 miles. The viewer will ask "what

film"? The film is moved to 10 feet. The viewer will see the frame of film, but no detail on the film. Move the

film to 6 inches from the eye. Now the viewer can see the image on the film.</P>

<P></P>

<P>The opposite will never happen. A viewer will never perceive the frame at 5 miles but be unable to see it at 6

inches. That's not how vision works.</P>

<P></P>

<P>Now do you see (no pun intended) why we are telling you that you cannot have gray grains at 10x which

disappear at 60x and 400x? Or gray grains in an 8x10 print which are gone in a 60 inch print? Or unobservable

gray grains which none the less are observed to be gray tones in a print?</P>

<P></P>

<P>I know I've said this before. Practically this argument has been done for a long time. Even if a gray grain

could exist, with just the right thickness and atomic arrangement to be translucent and not opaque,

it certainly does not occur with sufficient frequency to be the basis of any tone in a final print. If gray

grains formed gray tones, we would see a massive number of them under a microscope.</P>

 

[bernardwest]

Vijay... you haven't answered my post from 9.14pm.

[benny_spinoza]

Don't you guys need to ever work??

[jonathan_reynolds]

The answer to your question is highlighted by the fact that this thread is so very long. The differences are not overwhelming. I can make a 20x16 BW print from 35 mm film using darkroom equipment I bought and paid for in the 1980s, and for quality it holds its own with what I see made from digital equipment of equivalent format.

 

I cannot afford the equivalent digital gear and materials. On the other hand, if I wanted to start afresh with film today, I could easily afford to because the equipment has little value now, and the materials are still reasonabl priced.

 

Additionally, darkroom chemistry at a macro-level is fun. You don't actually need to know how many angels fit on the head of a pin.

 

So for me, film is still very viable.