15 Exposure 120 Film Back for Mamiya 645 AFD

q.g._de_bakker · April 18, 2011

Rishi, I have done my empirical testing continuously for more decades than i care to remind myself of. So have millions with me. I do believe your results. We can see what we can see. What should make you think a bit is your "So I don't even understand how changing the # of exposures to 15 fixes the problem. Maybe if they changed the # of exposures to 10 or 11..." Why do people who believe there is a problem say it can be fixed that way? According to your test and calculations, it wouldn't. If i would rephrase that as "it would not make a difference", we're already at where we should be at. ;-)

rishij · April 18, 2011

we're already at where we should be at. ;-)

</blockquote>

NO, we're absolutely not where we should be at. Advance each frame an extra 21mm, and then we'll be at where we should be at.

Also, with your experience, surely you can quote me the mm distance in between 120 frames for 16 exposures/frame. Then I can redo my calculation. Maybe it does come out to 15.

Please please please if you have some 120 film lying around uncut, measure the distance between frames & let me know so I can redo the calculation. Else we'll have to wait days for my own test!

Thanks,

Rishi

rishij · April 18, 2011

Here's the bend in the film (RVP 50) after sitting for only 1 hour on the rollers:

I mean... really... I wish this were a joke. It's not!

Mamiya? Phase One? Hello? Ansel Adams is rolling over in his grave...

-Rishi

rishij · April 18, 2011

And here's Frame #4, which spent just 5 minutes on the rollers:

5 minutes, folks. Which sometimes amounts to the time I need to meter a complex scene (ok, a bit of hyperbole there, but you get the idea).

-Rishi

q.g._de_bakker · April 18, 2011

Rishi, Have to run, so only a short reply. The distance between frames on 120 film is not set to that degree of precision. It all depends on the transport mechanism, how loose the film is on the roll, the thickness of the paper, the thickness of the film, and such. So it can and will vary from film type to film type, from back to back, from brand to brand (of back, and of film) from roll to roll, from the first frame to the last. I will suggest you run another series of tests though. And in that test you should actually exposing the film in the regular fashion and examine the resulting images. Reflections are great to detect minute deviations from the plane (you can measure the curvature/circumference of the earth with rather astonishing ease and accuracy using light reflecting off a bath of mercury). But would you not want to know how bad this would affect the images, rather than how bad it can look when looking at the film?

rishij · April 18, 2011

"And in that test you should actually exposing the film in the regular fashion and examine the resulting images."

</blockquote>

Q.G., that is on the books. It's exactly what I intend to do. But even without doing those tests, let me put things in perspective again. The film bend jumps millimeters off the pressure plate. Let me now quote an article that addresses the need for micro-adjustment for autofocus & talks about the levels of tolerance needed to ensure sharp focus from a lens onto a high resolution sensor:

How small of a variation? One of his sources said as little as 20 microns (0.02mm or 0.0008 inches) is sufficient to cause side to side variation. You can only detect that amount of variation with laboratory grade laser equipment, I’m told. Medical-grade machine parts (used in arthroscopes, etc.) are expected to have tolerances of about 50 microns, (<a href="http://www.devicelink.com/mpmn/archive/06/06/020.html">3</a>) and it would seem unlikely that a camera lens mount would be made more than twice as accurately as a medical arthroscope. In other words, with a top quality wide-angle lens on a high resolution sensor, we can perceive a 20 micron difference, but the manufacturer probably can’t make the part more accurately than +/- 50 microns at a reasonable cost.What concerns me is the authority with which you spoke without ever indicating that you performed empirical tests yourself.

-"This Lens is Soft" & Other Facts, original article <a href="http://www.lensrentals.com/blog/2010/03/this-lens-is-soft-and-other-facts">here</a>.

</blockquote>

Microns. So we're talking about the film bend introducing a misalignment of focus (a couple millimeters) that is 1-2 orders of magnitude more than the levels of misalignment between lens & digital sensor that result in perceivable loss of sharpness.

So, I'll do the test... but do I even have to? Furthermore, let me quote some personal communication with Bill from www.getDPI.com forums:

"I fully understand your problem. Although I haven't shot film recently, the problem you document was quite bad, particularly with the Hasselblad and Mamiya which both bend the film around the rollers at more than 90 degrees. The only solution was to shoot at f11! I moved to the Rollei 6008 system where the magazine is quite a lot longer because the film does not get bent at all. It just comes straight off the roll into the film gate with no bending. It was a big improvement."

-Bill Caulfeild-Browne

</blockquote>

What concerns me is the authority with which you spoke earlier, dismissing these 'rumors' that needed to be 'debunked'. Made me comfortable enough to buy this entire system, thinking it may serve me better than a 5D Mark II. I'm not blaming you at all; I'm a grown man who has to make his own decisions... & I can't say this was entirely not-worthwhile... I think we all learned something here.

But, I have to ask: have you performed the simple test I performed above? Have you shot high frequency material & scanned at high-resolution & verified yourself there's no loss in resolution due to these hideous bends in the film?

-Rishi

q.g._de_bakker · April 18, 2011

Rishi, What still concerns me is how you summarily dismiss many hundreds of thousands of people using MF for just about 100 years without noticing anything bad, until Zeiss said to look at our films and how they bulge. So (as you did concerning that 35 mm vacuum back thingy) you swallow that hook line and sinker, and look and see your film bulge, and show great concern about that, without (!) even testing what all those many hundreds of thousands could have told you. And yes, you are far better served using MF film (even rather crappy MF film cameras) than that Canon. We don't even have to consider film flatness to know that. Now, could i interest you in the millions of excellent large format photos that have been made since photography was invented, and then tell you to have a look at how sheet film sags in the holder? I think i know what would capture your attention - the beautiful pictures or the sagging film - then too ;-)

rishij · April 18, 2011

Again, that test is on the books & I'll post my film scans hopefully within the week. So please be patient.

In the meantime, I'm arguing that there's no way it can't make a difference because the numbers (mm vs microns) speak for themselves. Unless you want to argue that Velvia film isn't high enough resolution next to a modern DSLR for this to matter... which I certainly hope you don't think.

Once again, one carefully done, repeatable, reproducible negative result is all that is needed to 'debunk' this problem. Sure many people may not have seen it because:

<ul>

<li>They didn't look for it/didn't scan at high enough resolutions</li>

<li>Shot the roll in rapid succession</li>

<li>Shot at high apertures, like the f/11 that Bill mentions</li>

<li>Shot shallow DOF portraits where it's hard to tell what's supposed to be in focus vs. not</li>

</ul>

I could quote you many reasons why many people don't see it. Doesn't mean the problem doesn't exist. A guy on these very forums shot newspapers on a wall and saw the same problem (here's his <a href="001GYA?start=8">post</a>): a section in the middle of the frame was completely out of focus across the frame (left to right). That's exactly what the predicted outcome is for my findings. Bill confirms the problem in his findings. That's enough evidence for me to shy the heck away from this format, but, like I said, I'll confirm with actual shots.

But regardless, there's enough evidence to show that the design flaw exists. And worse, it's easily rectifiable... but not by me or any end-user. And now impossible, it would seem, given the drop of film support by these companies. What a tragedy!

And what will convince you that it is, in fact, a real world problem? A clearly visible loss in sharpness could indicate a 2x drop in resolution, or more. What would people say if Canon released the 5D Mark III & followed it up with a statement "The middle 10% of your image may drop from 28MP to 14MP resolution or less every now and then if you shoot at wide apertures". Imagine that!

Again, I'll attempt the test, though it's rather hard to quantify the exact loss in resolution since you need to have a chart that's already at the limiting resolution of the film at the center of the frame. Or a chart with increasing density of lines from left to right... but the resolution captured might then depend on lens aberrations. I'll have to think a little harder about this...

-Rishi

P.S. I would be willing to bet you that the film sag with LF can cause a loss in resolution, & people either shoot at small apertures or otherwise justify that 'it's ok because the enlargement factor from large format is so much less'. Which defeats the point that if there's a sag you're not doing the film's resolution justice. Whether or not that matters to you is a different story.

rishij · April 18, 2011

Let me quote Andrew Schank from his photo.net post <a href="003kkW?start=8">here</a>:

Many people have experienced a lack of sharpness in some of their 120 images because of this and don't correctly attribute it to film curl. It often just looks like you slightly missed the focus. On a resolutiom chart with my Rollei,I can take several images without changing anything on the camera except advancing the film, and there can be large differences in the lines per mm resolved. It is less noticeable at f11 & 16 than at f4 and f5.6.

</blockquote>

There you go. How many people do these sorts of tests with resolution test charts? It's not exactly easy nor the most fun thing to do. If you're just going to say 'it doesn't matter in the real world' I will still point out to you that regardless you would be losing resolving capability of your film at these heavily curled spots, & that places a severe limitation on your imaging format especially in the face of advancements in digital sensor technology. For example, sometimes I will shoot a faraway landscape at f/4 with my 70-200 f/4 L IS so that I can shoot handheld in available light. It would really not be cool if I had to accept the fact that I will experience softness due to the f/4 aperture... that would be a severe limitation.

-Rishi

q.g._de_bakker · April 18, 2011

Rishi, Run the test yourself. You're jumping from reasons why people (you do know the number of people you are talking about?) might have missed it to reports of that they haven't, from there to that they have because noone shoots test charts to who knows what's next. This entire thing is like people looking for a nice small boat to have some fun on the water with, reading a report that you can't actually sail a boat on water, because water is soft, and the boat will sink in it, and if it doesn't, the water is always moving and rippling (at best) and that's a huge problem. They will go and look at water, see how easy it is to submerge their hands when they put then in the water, proving that there is indeed very little resistance and anything you pt on it will indeed sink. And lo and behold, still water is indeed very, very rare. So there! It's proven you can't sail a boat! Next someone tells them people have been doing so for thousands of years, and their reply is to ask whether that someone tried it himself by doing the same thing they have, and how anyone could arrive at a different conclusion. And besides, there are reports of boats that sunk. So... ;-) It is not (!) a problem, except in the minds of people who indeed believe that that airplane they see in the sky can't possibly fly, because everyone knows that [etc. etc. etc.].

rishij · April 18, 2011

Oh c'mon Q.G. That metaphorical comparison is just silly & you know it.

I provided you with evidence that shifts of 20-50 microns between the lens & sensor can cause softness on modern high-end DSLRs (otherwise why would AF microadjustment help?). A 5D Mark II can, at best, resolve ~156 lines/mm with its sensor given its pixel count (simple math). Mauro Franic & I showed that Velvia 50 + lens can resolve 150 lines/mm for a 1:20 contrast chart. Fujifilm themselves measured their film & claim that it resolves 160 lines/mm for 1:1.6 contrast (without lens). So we were right in the ballpark for film + lens combination. Hence the resolving power for the Velvia is at least as good as, if not better, than that of the 5D Mark II.

Therefore, you cannot argue that the same 20-50 microns (or more) shift will cause softness on a 5D Mark II but not on Velvia. And I showed you film popping millimeters out from the pressure plate.

For your ship/water analogy, the guy claiming that water is soft would have to show you a boat that sinks that is indeed constructed similar to a boat that floats, at which point you, in disbelief, would try to find a flaw in the sinking boat. In your analogy, looking for that flaw in that sinking boat is equivalent to finding a reason why Velvia film that pops that much does not cause a loss in resolution (I had to spend a while thinking about your analogy). Well, the only way that is possible is if Velvia resolves less than a 5D Mark II sensor. Which is just not true according to Fujifilm.

QED.

-Rishi

P.S. It'd be pretty funny if I proved myself wrong... but as a scientist, I have to always be open to the possibility :) You'll hear back from me hopefully by the end of the week with real world tests & scans.

rishij · April 18, 2011

And for anyone who thinks that leaving the frame of film that was previously on the rollers on the pressure plate for a while straightens out the film, think again...

Here's frame #2 which sat on the rollers for 12 hours, and then sat on the pressure plate for 1 hour (after advancement to shooting position):

Hardly any change, in my opinion. Not surprising, since the pressure plate hardly exerts any real pressure in my estimation...

Rishi

bill_caulfeild_browne · April 18, 2011

As I've been (accurately) quoted by Rishi, I'll pitch in. I have used 4 by 5 inch cameras and yes, film flatness is an issue, though sheet film never was bent. It still bulged a bit in the middle of the holder. The saving grace is that the 4 by 5 lenses are not very fast and in any case, especially for landscape shooters, are used considerably stopped down.

With Hasselblads (in the 1980s anyway) I found the problem quite acute. So acute in fact that I could see no point in using F2.8 or worse, f2. (I had a 2000 FC ). But as a landscaper, I generally shot for max dof so it wasn't a major issue for me.

And Hassy's had the problem of backs mating properly with the camera body, too, which didn't help. I moved to the Rollei 6008 partly because its magazines did not require any bend in the film and I was able to shoot at larger apertures with confidence.

I still have hundreds of Kodachrome 120 shots from that era. That film had very high resolution and was very unforgiving of any focus errors, but rarely did I have a problem with the Rollei shots. Bill

q.g._de_bakker · April 19, 2011

Rishi, As a scientist, you should get things straight, not mix things up. The depth of film and that of a digital sensor differ considerably. So the focussing accuracy required for either does not transfer to the other. It's more difficult to get an image in focus on digital sensors. So though i didn't even want to, i certainly can argue that "the same 20-50 microns (or more) shift will cause softness on a 5D Mark II but not on Velvia", and be absolutely right too. But that's irrelevant. Here's (another) question for you: do you think you (or an AF system) can set the lens to film distance (i.e. focus) within microns of what it is supposed to be? And have you measured the "milimeters" the film bulges? The simile is a bit silly, yes. Of course it is. But really not inaccurate. The guys who told us that film bulging is a terrible problem should (like you) not just pontificate about the fact that they can show that film bulges, but at least also show that that is a problem. They haven't, just as you still haven't tried to find the problem. Bill at least has reported results. I don't share his experience (and yes, i also use the f/2 lens, at f/2), and i am sure very many with me don't. But that at least is a report not just of a 'cause' for possible problems, but of a result of a cause. So again, stop worrying about causes, and start looking for effects. I will again remind you of the 35 mm vacuum back episode: you are sure you have never seen a problem, but a single mention of a 'solution' and you're convinced there must be?

rishij · April 19, 2011

I will again remind you of the 35 mm vacuum back episode: you are sure you have never seen a problem, but a single mention of a 'solution' and you're convinced there must be?

</blockquote>

Who said I was convinced of a problem with 35mm film? I'm absolutely not convinced, because 35mm doesn't make any sharp angles around any rollers!

And have you measured the "milimeters" the film bulges?

</blockquote>

Yes. Sorry next time I'll place the ruler in the image when I take shot of the film bulge.

Here's (another) question for you: do you think you (or an AF system) can set the lens to film distance (i.e. focus) within microns of what it is supposed to be?

</blockquote>

Instead of asking so many questions, why don't you answer one?

It's more difficult to get an image in focus on digital sensors.

</blockquote>

Why? Light that falls on the sensor plane is focused by microlenses, making the effective focal plane pretty thin. Film also has a very thin focal plane, though technically a different one for each color emulsion. I have no idea how the two compare, but can you give me some rationale for this claim?

-Rishi

P.S. I'm tired of repeating myself, but, I have a job. It's been a day since I identified this problem, so give me a break. The results are coming. Meanwhile, the theoretical side of this discussion is interesting to me. But if you're going to make bold claims, it'd really help the rest of the community if you backed them (e.g. 'It's more difficult to get an image in focus on digital sensors').

rishij · April 19, 2011

Here's (another) question for you: do you think you (or an AF system) can set the lens to film distance (i.e. focus) within microns of what it is supposed to be?

</blockquote>

To answer your question: yes... within a tolerance of 20-50 microns according to that article. Or presumably at least elements can be moved within the lens (these aren't single element designs) such that the effective simple lens to film distance is changed on the order of hundreds to tens of microns.

Let me put it this way: when I rotate the manual focus ring on my 50mm lens, I can move the barrel in sub-millimeter steps. My rough estimate holding a ruler close to the front barrel: I can move the ring about 10 discrete times or more as the barrel extends from 0mm to 1mm. Those are 100 micron steps. Any one of those steps can completely throw off the focus of an object 10ft.

A very unscientific test, but, it'd indicate that the answer to your question is yes.

Since you asked the question in a rather authoritative tone, I assume you have the answer?

-Rishi

rishij · April 19, 2011

Also, FYI, the AF motor in the Rebel T2i can well outresolve the movements I can make to the manual focus ring with my hand, thereby nailing focus even better than my hands can (although I can do just as well with my bare hands with a lot of trial & error).

Gotta love technology. I think this is one of the reasons we see defects more easily today is b/c digital technology has allowed us to make critical evaluations with relative ease.

-Rishi

zack_zoll · April 19, 2011

Rishi- Not to get involved in the film flatness debate (never had an issue on my Hassy or Yashicamat, but did on cheaper 120 cameras, so I can see both sides here), but I will comment on the 'harder to focus with digital' bit.

Film is three-dimensional, even if it is perfectly flat. It is just barely three-dimensional, but it is. Colour film especially has three or four layers of emulsion, each sensitive to a different colour. This means if your focus is off by a few microns, or if you have a small amount of chromatic aberration from a cheap UV filter or a poorly coated lens (or a close-up filter on a lens that doesn't like them), often each layer of emulsion will see what it wants to see, and disregard the 'wrong' bits. This may lead to one or two layers of emulsion that are out of focus, but another that isn't, which makes the image look sharper.

Digital sensors, even those like the Foveon with layers, are really still two-dimensional. The first layer is covered in microlenses, which shunts the image off to where it has to go. That means that focus (and colour alignment) needs to be perfect on that first layer, as the microlenses effectively act as a single layer of emulsion.

This is a VERY miniscule difference mind you, and in practice does not make a huge difference. I only see a difference when using 1.4 lenses, and even that can be chalked up to the worse viewfinder on most digital cameras. It does result in a quantifiable difference in CA though.

q.g._de_bakker · April 19, 2011

Rishi, This is getting off topic, but still a fun debate: it's not about whether a lens can move to any position (it of course can, from stop to stop it moves through any position in between), but whether you are able to hit the desired position within that accuracy of 50 microns either way. That involves a bit more than just being able to move a lens. First, for instance, you have to be able to determine what that desired position would be. Anyway, here is some more 'disconcerting' information to ponder. ;-) You have a Pentax, my data are relevant to Hasselblads, but i doubt that there is much of a difference. The body length of these cameras is adjusted within a tolerance of plus or minus 30 microns. The film position within the magazine within a tolerance of plus or minus 50 microns. There are similar tolerances on focussing screen position and mirror angle. And yet these are held in rather high esteem for being precision machines (and not undeservedly so), capable of producing images of excellent quality. How could that be? ;-) And further: i never said you were conviced there is a problem with 35 mm film. I have pointed out to you that you were absolutely convinced there wasn't one, and that the mere mention of someone marketing a 'solution' was enough to sway you and make you doubt what was your knowledge and conviction. I did that (repeatedly) not to suggest that you think there is a problem, but to demonstrate how these reports produced by Strategic Marketing Departments produce the desired effect with the greatest of ease. People will believe absolutely anything if told in the right way. Beware of that. ;-)

rishij · April 19, 2011

Q.G.:

I have pointed out to you that you were absolutely convinced there wasn't one, and that the mere mention of someone marketing a 'solution' was enough to sway you and make you doubt what was your knowledge and conviction.

</blockquote>

No, what made me 'doubt' was my scientific mindset which is trained to always question.

And yet these are held in rather high esteem for being precision machines (and not undeservedly so), capable of producing images of excellent quality.

</blockquote>

Again, maybe because it wasn't as easy to observe as it is now with digital sensors (that provide quicker feedback & therefore allow tests to be carried out with less hassle) & technology such as Live View with 10x magnification. Now it is extremely easy to critically judge focus/sharpness errors by comparing images side-by-side on a monitor or layering them in Photoshop & switching between layers. This sort of critical evaluation just wasn't available to the average consumer decades ago. You may argue at this point that such critical evaluation is only for academic purposes, and I might agree with that. But since I have a scientific mindset & wish to evaluate a system's potential before I invest in it (especially in the face of options like MF digital or 24MP full-frame dSLRs), I do not find this sort of debate or critical evaluation unreasonable.

That involves a bit more than just being able to move a lens. First, for instance, you have to be able to determine what that desired position would be.

</blockquote>

FYI when I reported that I could move the lens element hundreds of microns (maybe less) in my earlier post, and even nail focus but not as easily as the camera nails focus, I was assaying this by Live View at 10x magnification. That's how I determined what was in perfect focus or not.

Also, I have the Mamiya 645AFD III, not the Pentax.

rishij · April 19, 2011

Zack, thanks for your answer! That makes sense... I'd started to put two and two together last night when I figured that the plane of focus is really thin for digital SLRs b/c of their microlenses, while the plane of focus on film varies due to the emulsion layers... but I didn't figure it would make much of a difference. For film, anyway, wouldn't you want to nail the green-sensitive layer, given that that's what humans are most sensitive to?

I can perhaps buy that the film may be more forgiving, but I do know that with one of my telephoto lenses (L-series, tack sharp for most of the field) that renders the left ~15% of the image slightly soft at f/4 does so on both my 5D & my 35mm film shot in my EOS-3. In other words, though I haven't done extensive side-by-side testing to quantify the difference, in real world shots I can see the falloff of sharpness on the horizon with shots taken on 35mm Velvia as well as a 5D with that same lens. Neither seems more forgiving than the other. I assume this is due to a misaligned element that leads to that part of the image missing the focal plane, b/c the glass on this lens is otherwise perfect & it's supposed to literally be Canon's sharpest lens.

-Rishi

q.g._de_bakker · April 19, 2011

Rishi, The point that still doesn't come ac ross, it appears, is that film technology has been tried and tested over and over again, by countless people, both sloppy and uncritical, and the analy precise, over an exceedingly (compared to how long it should take for anything bad to show itself) long time. If anything was up with film, we would know it. Not that film is perfect. But if there was that problem Zeiss wanted us to buy that Contax and vacuum back for, we would have known. We really would. You're "easy to observe" point is something you should think about for a while. Remember that we're dealing with a visual medium. If something marring image quality hides itself particularly well, it, by doing so, lessens itself as a problem. If noone ever notices, it's not there. The fact that problems are so much more easy to detect in digital capture says a lot about digital capture. And that's it. If you really believe that now, people compare images more than ever, you will have to think again. If you think that computer screens (of all things!) makes comparing easier or even more accurate, you are on the wrong track entirely. So critical evaluation was more available back when digital technology did not exist. Less so now. What has changed is that nowadays anyone can post scans and digitally captured images on the web, and claim that we see something looking at them that may or not be there. (And don't get me started about the nonsensical "comparisons" that indeed abound on the web. ;-) ) I too do not find critical evaluation unreasonable. What i do find unreasonable is unreasonable "evaluation". Things like ignoring the already mentioned ongoing test, by countless of people, and the results of that, asking if i (or anyone) has tested film yet... (What do you think? ;-) ) Anyhow, let's wait and see what your test results show. A few miscellaneous notes: Micro-lens arrays on sensors are not the problem. Sensors are shallow. Softness in images can be caused by quite a few things besides film flatness and focussing error. If a lens' performance drops at infinity when set to a closer distance, it will be mostly the effect of focussing, i.e. the lens is not focussed to infinity so the horizon will be soft (what else can it be? Something people who use hyperfocal distances to 'focus' should know, but - proving your point about not looking closely enough - given the ongoing popularity of the 'technique' many apparently don't. Proving also my point that many people rather believe what they read somewhere than what they (should) know and understand ;-) ). It can also be just a matter of lens design, and the resulting behaviour, worsening the effect of disfocussing. Why do you think it would be a misaligned element, i.e. a manufacturing/quality control error? P.S. My aplogies for mixing up the Pentax with the Mamiya. The point however remains the same. What do you make of it?

rishij · April 19, 2011

If anything was up with film, we would know it.

</blockquote>

You continue to deny the 'countless' (if I may use that word) people who do claim that they see real-world softness due to film flatness problems. Just search photo.net. Seriously, arguing with you is pointless if you pick & choose which results to believe. You're trying to prove something doesn't exist for which there is evidence that it does exist.

If you really believe that now, people compare images more than ever, you will have to think again. If you think that computer screens (of all things!) makes comparing easier or even more accurate, you are on the wrong track entirely. So critical evaluation was more available back when digital technology did not exist. Less so now.

</blockquote>

We're gonna just have to agree to disagree. Yes it is much easier now to test the sharpness of lenses with a digital camera using live-view to ensure focus & then by doing side-by-side or layered comparisons in Photoshop than it was to use a huge magnifying loupe or a light microscope to judge sharpness of one film frame vs. the next (that you have to switch between) that you may or may not have properly focused/exposed, etc. Even comparing prints side-by-side is not as easy and is more prone to human error than switching between layered images in Photoshop. To suggest anything else is ludicrous. You may argue that practically speaking, we should just look at the print. But for academic purposes in evaluating a system, the accurate and objective method is the more preferred.

Now, judging of film resolution is still hard in today's age, b/c scanners add another layer/variable. Probably a light microscope is still the best way. But even that is not as objective as a stellar/optimal scan on a high-resolution scanner that is then run through mathematical analysis software such as Imatest.

Micro-lens arrays on sensors are not the problem. Sensors are shallow.

</blockquote>

Sensors are shallow compared to the depth of film? Again, if you make statements like this, please back them up with resources stating actual numbers. What Zack was saying was that the micro-lens array makes the effective focal plane of a digital sensor quite shallow... perhaps shallower than the effective focal plane of the film given its different emulsion layers. Your statement is once again a bold claim without any support.

Why do you think it would be a misaligned element, i.e. a manufacturing/quality control error?

</blockquote>

Um, you're suggesting the very vague 'quality control error' whereas I'm suggesting a basis for that quality control error. Please try & make your point clearer, in more rigorous scientific terms.

-Rishi

rishij · April 19, 2011

I got some numbers if anyone's interested:

CCD pixel depth can be on the order of 5 microns.

Film base is ~100 microns for 120 film, & in Fuji's diagrams it appears that the color emulsion layers together are on the order of thickness of the film base... so maybe we're talking about 50 microns depth per color layer?

So I think those numbers give Q.G.'s statement 'digital sensors are shallower than film' some credence (you're welcome Q.G.).

But it's hard to assess what this means in the real world in terms of lens sharpness tolerance (or focus tolerance). In film, it's not like light rays falling within the a certain pixel pitch all get binned to the same pixel. If focus is off, the light ray may travel further or less through the emulsion & therefore expose a grain at a different location (in the X-Y plane of the film). So technically that should introduce loss of sharpness... but if you invoke the theory that an area of film has to be large enough to resolve thousands of levels of tones before it can be thought of as a 'resolving element' (like a pixel), then, yes you can kind of treat a certain square area of film as a 'bin' that will average whatever photons strike the emulsion within that 'bin'.

But this is all conjecture at this point & I don't even know if this process is well understood. After all, weren't theories of how film really works put forth after people developed film itself?

Rishi

zack_zoll · April 21, 2011

Dude. Rishi. (Dudette?) You're about three steps from suggesting that we're all, like, molecules in some giant's fingernail, man. I actually think I'd need to go back to college and smoke some of that 'green stuff' to understand what you just said.

Lemme' sum up what others have said, and hopefully it will help. I tried to read that last post, but I honestly couldn't understand it. I don't mean to be offensive, but it's 10:00 here and I've had a loooong day. I hope my summation clears things up.

Anything with an aperture will have a depth of field. Even enlarging and projecting lenses have a depth of field, although the variations are so slight that it doesn't usually matter to most people. The image that leaves the back of the lens also has a depth of field. Since film thickness varies (both base and emulsion), camera and lens designers generally design their gear with a little tolerance, or 'wiggle room' so that a loose lens mount, or a different thickness of film, or a loose film plate, won't drastically impact the negative quality. Since the receptive part of the digital sensor is so much thinner than film, exact alignment becomes more important; it's basically like trying to park the same car in a much smaller parking space. It still fits, but there's not as much margin for error anymore. This doesn't generally have a huge effect on focus, but colour alignment is even pickier, so you might see an issue there.

To rephrase the vacuum back statement: film flatness issues due to film that's been bent around a roller probably isn't a huge issue, or else more people would produce a fix for it. A good analogy for this is the audio/cable company Monster, which produces a 'power cleaner.' If you live in a building with poorly or long-ago wired electricty, the claim is that it can affect the performance of a stereo, recording device, hot lights, or anything else that requires a constant voltage or is suseptible to 60-cycle hum. If Monster was still the only company that made such a device, we could assume that the claim was bunk. Since many other companies make such a device, and since a lot of audio components build the circuitry into their components now, we can assume Monster's claim about 'dirty power' is valid.

Since Nikon, Canon, Pentax, Hasselblad, Leica, etc. haven't gotten on board with vacuum backs, we can assume the arguement is bunk.

Sign In

15 Exposure 120 Film Back for Mamiya 645 AFD

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in