Updated Graphic - Max. print size for 5 formats

[bglick]

While participating in another concurrent thread, I had updated my graphic that shows final print size, for DSLR and P65+ MF back, as well as 3 film formats. The formats are:

15MP DSLR
24 MP DSLR
60MP P65+
6x7 Mamiya 7 (Velvia)
4x5 Velvia
8x10 Velvia


Instead of a bar chart, I re-did the graphic to show the true aspect ratios of the max. final print sizes (to scale) while holding 5 lp/mm to all prints. The purpose of the graphic is to demonstrate how apt. diffraction effects max. print size, (at 5 lp/mm) at 4 different DOF scenarios,


1) No DOF,
2) Min. DOF
3) Avg DOF
4) Max. DOF.


This illustrates where each format excels, as it relates to recorded resolution. All f stops are scaled to their respective formats. The numeric table at the bottom. 1/R was used, at 1500/f stop for aerial resolution and for recording media - digital used pp/mm and Velvia 80 lp/mm.

[Dan Fromm]

Um, Bill, why go to all this trouble? If you'd posted your magic formulas we could evaluate your work and find the effects of changing your assumptions. Also, it would be useful to know the sizes you assume for the three chips and the units of measure for the prints.

And, now that I think if it, what's become of your perfectionism? The generally-accepted rule of thumb for the minimum resolution in the final print that will let it look sharp at normal viewing distance (10") is 8 lp/mm.

Finally, what's the point? The people who believe that digital is better than film will smell fudge in your results, those of us who think it depends of the size of the final print will wonder whether you made silly assumptions.

I can't help wondering whether you've assumed that technique will be perfect. This is unrealistic as can be. What's interesting is how sensitive final print quality is to shake, focusing errors, ...

[jb7]

I think it's a worthwhile exercise- and for the sake of argument, I'm prepared to assume that all other variables are taken care of-

The one thing that becomes apparent is how it might inform your choice of equipment for a particular type of shot-

I think it's a good illustration of your arguments in the other thread, and if anyone wants to print at other sizes, I don't suppose you'll be stopping them-

Valuable work, I think, although I'm not qualified to question the details-
but you wont be found wanting for those who can-

j

[sanking]

I would personally find the exercise more useful if some of the assumptions were stated so that I could compare to my own work. For example, the final print sizes for the Mamiya 7 and Canon 50D are a bit smaller than I had calculated for my work, using the same value of 5 l/mm in the final print. Perhaps we are just starting with different maximum resolution values but it would clarify things if that kind of thing were stated up front. Other information would also be useful, for eample are you figuring optical printing with film, or scanning and digital printing?

Inquiring minds like details!!

Sandy King

I think it's a worthwhile exercise- and for the sake of argument, I'm prepared to assume that all other variables are taken care of-


j

[Joanna Carter]

I take it all these measurements are in centimetres or metres?

Don't forget, not everyone in this forum is American, even Canucks are metricated now

[Jim Michael]

I could build a simple Mathematica application that would allow you to manipulate the parameters as you think they should be while conforming to the physical limitations of 1/r etc. You would just need to install a viewer to use it. If bglick is interested in collaborating on this he can PM me. There are different ways of approaching this sense there may be different problems one is attempting to solve in an analysis. It's easier to build a manipulate routine for solving a specific problem vs. trying to multiple problems with the same application. This problem I interpret to be "find the maximum print size one would expect good results from using a particular source (film or digital image having specific physical parameters)". Of the physical parameters in determining these final dimensions you should also decide which of those are fixed (lambda?) and which may be manipulated. Any final product would be published on the Wolfram Demonstrations Project page or other public-accessible site. I'm open to ideas for collaborating on other demonstrations as well.

[Dan Fromm]

Jim, why Mathematica? All of these calculations can be done in a simple spreadsheet.

[jb7]

Well that sounds interesting too-
It would be good if that information proved to be as accessible to people like me,
who don't necessarily find advanced math to be as knee slappingly jolly as some,
but who might have a basic grasp of the implications of it.

It would also be good if it was universally available- that is,
available to people who use a mac rather than a pc,
but as I've only heard of that program and never used it,
I don't know if this is the case.

If that is the case, then the broader brush of the illustration above is certainly more useful than nothing.

I did make the assumption that the units were inches-
though stating that to be the case might have saved a lot of obvious disgruntlement-
or whatever interpretation you'd like to put on it-

[sanking]

I sure hope those measurements are in inches. Standard is sure high if they are in centimeters.

Canucks and brits can check their calculations at http://www.worldwidemetric.com/metcal.htm

Sandy King

I take it all these measurements are in centimetres or metres?

Don't forget, not everyone in this forum is American, even Canucks are metricated now

[Jim Michael]

Jim, why Mathematica? All of these calculations can be done in a simple spreadsheet.

Go for it.

I can put an equation in Mathematica and generate a fully functional application with manipulate components in a few minutes. Also, it has superior graphics capabilities, symbolic math support, greater accuracy, works on any OS, and the viewer is free.