Newbe question 35mm vs MF and dof

[David Klepacki]

Thanks Nick.  Regardless of its cost, the only point of looking at the Achromatic+ is that it is a clear example that shows the Bayer versus non-Bayer resolution differences.  

The theoretical differences in luminance resolution are quite substantial as I have shown; however, the realized gains in resolution will of course depend on the individual lens capabilities as pointed out by Phase One.  With some modest lenses, even a 5% - 10% gain in resolution is still 2x - 3X more than what was claimed earlier in this thread.   As Phase One also points out, reproduction lenses will typically show the highest resolution gains.

[Sheldon N]

The bottom line here is that Bart has made claims here .... that are not supported by anyone in the photographic industry.  

LOL!

[David Klepacki]

Sheldon,

I know you would love to think that somehow my views are somehow "radical".  However, the fact is that none of my claims are inconsistent with that of the field of scientific digital imaging.  Different CoC values are simply defined in different ways to suit different objectives.  And, this is ground that has already been covered here.

The difference in the current debate is that Bart's claim about luminance resolution is not supported either in the field of scientific digital imaging or in the photographic community.  And, in fact specific examples have been provided here where respected people in the photographic community, such as Mark Dubovoy, do not agree with Bart's claim.  

On this topic of luminance resolution, no one has come forth to prove wrong the conclusions of Mark, Claus Molgaard and Phase One.  If you are also a believer in Bart's claim about luminance resolution here, please show us your evidence and prove us all wrong.

[ErikKaffehr]

Hi,

I don't argue your point. This is what Mark wrote:
-------------------------------------------------------------------------------------------------------------------------
"All images were shot with an Alpa SWD focusing on the groundglass with my good old eyes. I used the Rodenstock 100 mm APO HR lens at F/8 which is very close to its optimum aperture. The files were converted to TIFF in the latest version of Capture One with no adjustments and no sharpening. The files were then opened in Photoshop CS4 and enlarged for the attached screen shots.

Conclusion: The Achromatic back definitely has more resolution than a standard P45+ back. Compared to a P65+, however, the Achromatic back looses, but not by much.

You need to look carefully, and of course I have the advantage of the original files, so if you cannot see it in your computers, you will have to trust me that the P65+ has slightly superior resolution to the Achromatic back. I would say that the resolution of the Achromatic back is about 2/3 of the way between a standard P45+ and a P65+.

_________________________

I have personally made comparisons between the standard P45+ and the P65+ on the Phase One 645 camera with Phase / Mamiya lenses. Although the "look" of the P65+ is much more pleasing, with better microdynamics, better color shading/color saturation and better dynamic range, when using the Phase One camera, I can see no difference in resolution between the two backs.

On the other hand, when I use either the Alpa or the Linhof with HR lenses, there is a very noticeable difference in resolution between a P45+ and a P65+.

It all fits together. Hurrah for the laws of Physics!"
-------------------------------------------------------------------------------------------------------------------------------
The sample from Dr. Klaus shows some jaggies on the hands of the clock. Dr. Klaus also point out the "black dots" on the "tomato soup can" being more visible on the achromatic back. Quite visible in the samples.

Best regards
Erik

Erik,

First, we cannot include the P65+ in any comparison with the Achromatic+, since the sensor is different with different size pixels.  It is the direct comparison of the P45+ and the Achromatic+ that proves wrong the claims made by Bart van der Wolf in this thread.

[David Klepacki]

Thank you Erik.  That's all I have been trying to say.  I haven't yet had a chance to look at Lloyd Chamber's images, but I trust that his claim of seeing up 10% more resolution in his own images is consistent with Mark and Claus.

[ErikKaffehr]

Hi,

I revisited the article here on LuLa David recommended.

Mark Dubovoy wrote that the Achromatic back was 2/3 on way from P45+ to P65+, that would be around 16% on a linear scale.

In the same article there are also some images and words from Dr. Claus (Mölgaard). The image from the Achromatic back is cleaner.

Now, some of the advantage may come from the raw-converter, which doesn't need to interpolate colors from surrounding pixels. Obviously there will always be some benefits with using filterless backs for B&W. The filters absorb a lot of photons.

I presume that the number of pixels is still going up and many of the issues we see now will be much reduced once the lens is clearly the limiting factor. (For you with the S2 it may be a long wait until sensor outresolves lenses, but I don't feel any pity! )

Best regards
Erik

Lloyd Chambers has some good examples of the extra resolution of the achromat backs on his DAP site. It's clearly a step up in res, but not earth shattering, maybe 5-10% better (subjectively). At this end of the market, each few extra percent costs an arm and a leg!

[Sheldon N]

Sheldon,

I know you would love to think that somehow my views are somehow "radical".  However, the fact is that none of my claims are inconsistent with that of the field of scientific digital imaging.  Different CoC values are simply defined in different ways to suit different objectives.  And, this is ground that has already been covered here.

The difference in the current debate is that Bart's claim about luminance resolution is not supported either in the field of scientific digital imaging or in the photographic community.  And, in fact specific examples have been provided here where respected people in the photographic community, such as Mark Dubovoy, do not agree with Bart's claim.  

On this topic of luminance resolution, no one has come forth to prove wrong the conclusions of Mark, Claus Molgaard and Phase One.  If you are also a believer in Bart's claim about luminance resolution here, please show us your evidence and prove us all wrong.

I don't claim to have any level of understanding of Bayer aspirin, let alone a Bayer array. You guys totally lost me about 2 pages ago.

I just found your phrasing ironic, considering our prior discussion (sounds a lot like something I said). Plus, I love to inject a little levity now and then.  If you can't laugh at a good discussion about DOF and luminance values in a Bayer array, what can you laugh at? 

[Nick Rains]

You gotta larf or else u cry...

Unfortunately a similar thread has kicked off on the Leica forums, must have been something I said!

[hjulenissen]

In comparing achromatic to bayer sensors, is it "fair" to use OLPF for one or both? How strong should it be? (it is easy to "prove" that one technology has more acutance than the other by removing the OLPF of one, but that comparision may not be fair because you will also get aliasing).

The discussion about luminance resolution may not be fair without a discussion on:
1. How the world typically behaves chromatically (if the world tends to be spectrally flat or erratic)
2. How the spectral sensitivity functions of the pre-filters work (the amount of overlap between r, g and b.) If (in the limit) these filters are flat within the visible spectrum, a bayer sensor would equal an achromatic sensor. If the filters are very sharp and peaky, the difference between bayer and achromatic should be large
3. How demosaic algos weigh luminance resolution vs other niceties (a B&W algo could possibly improve acromatic performance, and could be a fairer way to compare to an achromatic sensor).
4. For practical B&W photography, spectral selectivity is regarded as important by many photographers. For a bayer camera, this is easily done in post processing. For a true achromatic camera it would have to be done using color filters (Argh).

-h
Examples of cameras where the OLPF is removed:
http://www.maxmax.com/hot_rod_visible.htm

40D spectral sensitivity:
http://www.maxmax.com/canon_40d_study.htm

"'a sensor with an equivalent number of pixels has a √2 resolution advantage over Bayer for the green channel and about 2 times for other colors. Clever demosaicing reduces the gap but you still have an anti-aliasing filter with Bayer. And, if you look at the maximum resolution of most Bayer cameras, they give a figure that's around 70% (1/√2) of the potential maximum value (Nyquist), whereas we can get right up towards the Nyquist frequency."
http://www.dpreview.com/news/1010/10100504Sigmainterview.asp