Chamonix 45N-1 Testing Procedure

[Steve Hamley]

May I suggest that the hypothesis is that the fresnel lens does NOT introduce significant focus error (innocent until proven guilty). In this way your experiment can hope to find evidence to contradict this hypothesis. If it does, we have all learned something (as Darin B pointed out above).

You may suggest that an additional lens inserted in the optical path between the image-forming surface and the taking lens doesn't introduce a (significant) focus shift, but the focus shift is there according to the laws of optics. If you don't "see" the focus shift, then it means your test is not sufficiently sensitive, or in practical terms, some other factor like focus error or depth of focus compensates for it.

And the hypothesis is contrary to several Chamonix users posted images and the manufacturer's posted information.

Cheers, Steve

[BradS]

You may suggest that an additional lens inserted in the optical path between the image-forming surface and the taking lens doesn't introduce a (significant) focus shift, but the focus shift is there according to the laws of optics. If you don't "see" the focus shift, then it means your test is not sufficiently sensitive, or in practical terms, some other factor like focus error or depth of focus compensates for it.

And the hypothesis is contrary to several Chamonix users posted images and the manufacturer's posted information.

Cheers, Steve

I am afraid you misunderstood what I wrote. I suggested a Hypothesis - not a violation of laws of physics.

This is how science works, when properly done. You start with a hypothesis which is consistent with the status-quo and conduct an experiment that will, hopefully supply evidence contrary to your hypothesis (supporting the alternative). The criminal justice system in the USA works this way too. However, in science, we often iterate, modify the hypothesis and do other experiments. Notice that in both cases, an experiment (or trial) which does not contradict the hypothesis is less informative, less powerful than one that does.




As an aside, I do not pretend to know how the results of the experiment will turn out...nor do I much care. I am encouraged that some of our numbers recognize the folly of investing too heavily in the opinions thrown casually about on internet fora and the value of careful independent testing. Finally, it is only mildly interesting to me that the Mfgr. has publicly admitted a design error...that fact does not detract from the value of independent experiment.

[stealthman_1]

I don't doubt the laws of optics, I doubt the real world issue. Besides, we're adding a 3rd lens when we look through a loupe to focus!
From a couple weeks ago...
Chamonix 45N-1 bought last fall (2008) All 100% crops of 1200dpi scans on an Epson V700 with Ilford Delta 100 film.
Schneider 90mm f5.6 @ 5'10"


Schneider 90mm f5.6 @ 15' Note the focus problem is most likely caused by the film not being flat, which is one of the main focus issues in any view camera.


Rodenstock 210mm f5.6 @ 5'10"


Rodenstock 210mm f5.6 @ 15'

You judge.

[stealthman_1]

And for your artistic enjoyment, again, my beautiful Goerz Berlin, Doppel Anistigmat, 12 inch, f7.7

[Darin Boville]

>>And for your artistic enjoyment, again, my beautiful Goerz Berlin, Doppel Anistigmat, 12 inch, f7.7<<

Those old lenses sometimes don't even *look* like lenses.

--Darin

[vinny]

stealthman, we met at whitney portal last year.
Your shot with the 90mm doesn't show me much. The ruler should be at a much steeper angle for this type of test. As for the 210mm, that's not an issue since the problem is with short focal lengths. For what it's worth, my tests with my 90mm proved the need to remove the fresnel.


vinny

[Jeremy Moore]

Yes, I think that is probably true....but...every good experiment raises more questions than it answers and this is a good experiment.

And every good experimenter knows scope creep is the death knell of a good experiment. I want to keep this simple.

So, what is your hypothesis?

May I suggest that the hypothesis is that the fresnel lens does NOT introduce significant focus error (innocent until proven guilty). In this way your experiment can hope to find evidence to contradict this hypothesis. If it does, we have all learned something (as Darin B pointed out above).

From a scientific standpoint you're coming at this the wrong way as you go with proven theory until you falsify said theory and not the other way around. This isn't some test at the edges of science where we will rework the nature of optics, but a practical/real world experiment looking at focus shift.

Proven optical theory: a fresnel is a lens and will change the plane of focus. Fresnels have already been "proven guilty" of this so there's no reason to assume otherwise.

What we're testing is whether this shift is substantial enough to be detectable visually under the conditions I have set which are the same conditions in which the problem is said lie (wide angle lens wide open at a mid-distance). You can semantically state the hypothesis in the positive or the negative, but due to the nature of the experimental model I will have learned something--visually detectable focus shift or no visually detectable focus shift for these conditions.

[Larry Gebhardt]

What I think you are actually testing is whether the ground glass was set at the correct distance for A) use with a Fresnel in front, B) use without a fresnel in front of it, or C) some other distance.

[Jeremy Moore]

What I think you are actually testing is whether the ground glass was set at the correct distance for A) use with a Fresnel in front, B) use without a fresnel in front of it, or C) some other distance.

I'd been working off the assumption (may be an incorrect one, see below) that the GG was properly positioned at the film plane and the fresnel was added later.

I just went to Chamonix's website and the .pdf they have up states:

Based on this, I do not recommend that users of 45N-1 camera to change the position of the Fresnel lens as all cameras are tested for the accuracy of the Fresnel position before leaving our factory.

This won't change the basics of the test tomorrow--to test the amount of focus shift between with the fresnel between lens and GG and with no fresnel--but it is going to add a third test where I do not focus by eye, but by using the Betterlight's digital focus verification on a focus target: http://www.betterlight.com/viewFinder_FocusTab.html

This does come as a surprise to me, though, because if it's true then switching to an Ebony or Maxwell screen wouldn't fix the problem as the ground glass/fresnel combo would be positioned differently than the fresnel/ground glass combo factory installed on a Chamonix as the fresnel is recessed closer to the lens in a ledge at a different plane than the GG. By my understanding, an Ebony or Maxwell screen would rest on the same plane as the GG--NOT as the fresnel. Will someone with a Maxwell or Ebony screen please help illuminate this quandry.

[BradS]

From a scientific standpoint you're coming at this the wrong way as you go with proven theory until you falsify said theory and not the other way around. This isn't some test at the edges of science where we will rework the nature of optics, but a practical/real world experiment looking at focus shift.

Proven optical theory: a fresnel is a lens and will change the plane of focus. Fresnels have already been "proven guilty" of this so there's no reason to assume otherwise.

What we're testing is whether this shift is substantial enough to be detectable visually under the conditions I have set which are the same conditions in which the problem is said lie (wide angle lens wide open at a mid-distance). You can semantically state the hypothesis in the positive or the negative, but due to the nature of the experimental model I will have learned something--visually detectable focus shift or no visually detectable focus shift for these conditions.

OK. Have fun. I hope you find what you're looking for.