Need technical issue explained in lay terms

[RustBeltRaw]

I am amazed no one's said "diffraction-limited" yet.

[Manfred M]

I am amazed no one's said "diffraction-limited" yet.

Actually I was thinking it, but with all of the other wonderful issues of lenses of that era. Tthe original Chevalier lens used by the commercially produced Dageurre cameras were pretty awful by today's standards and had a maximim aperture of f/15. The somewhat later (and famous) Petzval portrait lens was f/3.5 but had pretty signifcant softness as one got away from the "sweet spot".

[Manfred M]

I am amazed no one's said "diffraction-limited" yet.

Actually I was thinking it, but with all of the other wonderful issues of lenses of that era. Tthe original Chevalier lens used by the commercially produced Dageurre cameras were pretty awful by today's standards and had a maximim aperture of f/15. The somewhat later (and famous) Petzval portrait lens was f/3.5 but had pretty signifcant softness as one got away from the "sweet spot".

[inkista]

Assuming the D800 sensor site is about 4880nm and that the average size of the corresponding area on the daguerreotype is about 475nm, that means that more than 10 times as many sensitive "parts" can be found in every square inch of a daguerreotype than in every square inch of a D800 sensor. Thus, the daguerreotype's resolution is more than ten times greater than a D800 sensor. ...

Sorry, for chiming in so late, but I think your math is wrong. The 1/10th the size grain is the "pitch"--or measurement along one side. If you get 10 more pixels per side of an area, then you end up with 100x more pixels.

The D800 has 7,360 × 4,912 => 36MP (36,152,320 pixels)

Assuming we have silver halide particles that are 1/10th the size of those pixels, you'd get:

73,600 x 49,120 => 3,615,232,000 pixels, or 3.6GP.

[Mike Buckley]

Thanks, Kathy Li. Using your calculation, the resolution of the dageurreotype is 100 times greater than the D800. You may be right (I really don't know), but intuitively it seems incredible that it could be true. If it is true, Manfred owes us yet another revised graph.

[steve welle]

Kathy is correct of course about the increase in pixels being 100 rather than 10. But I wonder if that is what matters for clarity of the image. For example, if we had an image of a bunch of very fine fine vertical lines the thing that would help you resolve them would be the 10 fold increase in pixels in the horizontal direction. Having ten times as many in the other direction would not help resolve the lines. I suppose a diagonal line across the image would have 14 times as many pixels per inch rather than 10. Am I looking at this the wrong way?

[Manfred M]

Kathy / Steve - calculations are correct in the instance the the silver halides in a dageurrotype resemble a highly regular alignment like an electronic sensor, where everything is in a nice even order.

What if the alignment is more like a photographic film; where there can be all different sizes and a distribution that shows significant gaps between the halide crystals? In fact, part of the issue is we have no data on the density that the halides form.

http://www.optics.rochester.edu/work.../spr04/jidong/

[Mike Buckley]

I'm beginning to think that I wish I had never started this thread. That's because my capacity for understanding the technical details is not close to the standards of all you folks.

Understanding that context and now that I have seen the follow-up to Kathy's latest post, I take it that in all likelihood it is more accurate to say that the resolution of a daguerreotype is probably closer to eight times that of the Nikon D7100 sensor than 80 times. Have I got that right?

[RustBeltRaw]

Understanding that context and now that I have seen the follow-up to Kathy's latest post, I take it that in all likelihood it is more accurate to say that the resolution of a daguerreotype is probably closer to eight times that of the Nikon D7100 sensor than 80 times. Have I got that right?

In practical terms, I'd say yes. It would take a very precisely-manufactured film plate to have perfectly regular grains, and even if you consider only the average size of the grains on a given plate, the lenses of the time couldn't keep up with the plate's "resolution." A modern large-format lens with minimal air-glass interfaces, applied to daguerrotype plates or low-speed, finely-grained film will achieve the highest resolution a hobbyist can reasonably attain.

TLDR: For most applications, if you exceed 12MP, you'll be fine.