It comes from the physiology of the human visual system.
<http://www.ndt-ed.org/EducationResources/CommunityCollege/PenetrantTest/Introduction/visualacuity.htm>
This source states that the "normal" resolution of the human eye is
0.00349" at 12" viewing distance. Work that out and you'll find it's
286.5 dpi. Close enough to 300 dpi.
The same sort of computation is what's behind our DoF tables to compute
the CoC. IIRC, the first work of that sort was done by Leitz in the
30s. But as near as I can figure the typical CoC of about 0.03mm
assumes something more like 200 dpi at 8X magnification and may be based
on performance of old lenses. But, as you have seen and stated, it very
much depends on the individual viewing the image, the exact distance of
the eye and the type of image. I think you'll find that laser printers
typically produce text at at least 300 dpi and high quality dithered
text at 600 dpi because otherwise it's easy to see the dots at high
contrast ratios.
I continue to produce my prints at 300 dpi even though some labs print
at only 250. Your suggestion that viewing distance be that which allows
the whole image to be taken in at once is probably a very reasonable
choice. But I'll continue to make my images at higher resolution if I
can. Detailed images invite close inspection at reading distances of
10-12".
Chuck Norcutt
On 10/27/2010 12:08 AM, Moose wrote:
> On 10/25/2010 2:31 PM, Chuck Norcutt wrote:
>> 300 dpi is generally considered what's required in order to not see
>> dots at typical reading distance (about 10-12 inches or 25-30cm).
>
> Where do you suppose that came from? It just doesn't match my
> experience, and that of others I've read on the web. Sometimes rules
> arise at one stage of development of a technology, then persist even
> after the reason(s) for their existence no longer apply.
>
>> That works for everybody but Moose who has Superman vision.
>
> Yeah, but even with 20/8-10 acuity in the eagle eye, I don't agree.
>
> I have a print made with the Epson 1270 of a shot of one of my
> granddaughters taken with the 1.9 MP Canon S110. It's slightly
> cropped, so the effective input is about 150 ppi. It's a terrific
> looking print. Closer than about 10" or so, eagle eye can start to
> see unsharpness and artifacts. Beyond a foot or so, it has a 'sense'
> of sharpness and clarity that is excellent.
>
> My photobook has pages about 8.25x10.75". The native resolution of
> images to be uploaded to fill a page is 1200x1600. Again, input ppi
> of about 150. At about 15=18", I can see that there is a texture in
> low detail areas like sky. By 8- 10", I can see the pattern of the
> dots. But as with the print above, get out to normal viewing distance
> of a couple of feet, and there is a sense of great clarity and
> sharpness.
>
> I guess the point for me is that I simply don't notice dots in these
> images. the only time I've noticed the dot pattern in the book is
> when the issue has come up here, and I've gone looking for them. Not
> a single person who has gone through my book has ever said anything
> about this. IF I get right down to approaching nose distance, yup, I
> can see where the limit of apparent resolution is the color dots.
>
> I can also see where the DOF that was great at normal, even at
> semi-close, viewing distance, doesn't hold up. But that's exactly
> what DOF calculations predict! So what? I didn't make the image to be
> looked at that way. I want people to enjoy the whole image,what it
> may say to them, how it makes them feel - and you can't see the whole
> image from that close.
>
> Hmmm, maybe that's a good criterion. Can you see the flaws from the
> distance at which you can comfortably view the whole image at once?
>
> D P I Moose
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|