Sigma/Foveon Questions
in [Digital Cameras]
|
Prev: SI Facescape
Next: FF camera with mirrorless design
From: Martin Brown on 2 May 2010 04:19 Alfred Molon wrote: > In article <010520101118269360%nospam(a)nospam.invalid>, nospam says... >> so does the human eye. chroma is 1/10th of luminance, so bayer actually >> has more than the eye can resolve. it's not a problem. > > You are mixing up things. Besides the human eye is *not* the reference. But it is exactly the point. Bayer mask can give colour images that the human eye cannot ordinarily distinguish the difference from a full colour image. You have to construct very artificial test cases to show any improvement and that "improvement" is only visible when you go pixel peeping at high magnification. The only situation where I have hit the limit in practice was with monochromatic H-alpha imaging of the sun - not in normal photography. > > For instance, the human eye cannot see colours at night. According to > your logic, camera sensors should switch to black and white at night. Non sequitor. His point is that there is no point in recording any information that the human eye cannot see when the image is being viewed normally. The human eye has limited colour resolution and much finer detail capability in luminance. It makes good sense in any data compression or measurement system to exploit that feature. If the eye cannot see the difference in practice then why bother to waste resources on acquiring detail that cannot be seen*. * without photographing special testcards with fine complimentary colour detail and zooming in until pixels are large obvious squares. BTW For a very long time - in conventional colour film era photographs of astronomical objects were hopelessly wrong -all pinks and blues. That was because the strongest visible emission line sits on the deep green safelight wavelength for panchromatic emulsions. It was in the mid 70's before anyone managed to produce true colour astronomical images that showed what faint objects would look like with a more sensitive human eye. First published in SciAm circa 1971 it made the cover page. Regards, Martin Brown
From: Jeff R. on 2 May 2010 04:26 "nospam" <nospam(a)nospam.invalid> wrote in message news:020520100406536192%nospam(a)nospam.invalid... > if someone is taking photos to be viewed by humans, then it *is* the > reference. > > if you can't see it, then there's no need to capture it. Haven't done much astrophotography, huh? One of the genuine attractions of photography in astronomy is revealing what the eye, with or without artificial magnification, cannot possibly see. Similar argument, though not as clear cut, for long telephoto and maybe even for photomicroscopy. ....but then, if I can't *see* the Horshead nebula, I guess there's no need to capture it. -- Jeff R. (who occasionally uses his superzoom to read distant streetsigns)
From: Alfred Molon on 2 May 2010 04:42 In article <020520100406536192%nospam(a)nospam.invalid>, nospam says... > if you can't see it, then there's no need to capture it. Then don't capture colours at night, because the human eye can't see them. -- Alfred Molon ------------------------------ Olympus E-series DSLRs and micro 4/3 forum at http://tech.groups.yahoo.com/group/MyOlympus/ http://myolympus.org/ photo sharing site
From: Alfred Molon on 2 May 2010 04:44 In article <gAaDn.60702$4W2.58498(a)newsfe01.iad>, Martin Brown says... > Non sequitor. His point is that there is no point in recording any > information that the human eye cannot see when the image is being viewed > normally. By that logic we do not need 60MP MF cameras, because the human eye cannot see so much resolution. Hint: it all matters on the enlargement you make. -- Alfred Molon ------------------------------ Olympus E-series DSLRs and micro 4/3 forum at http://tech.groups.yahoo.com/group/MyOlympus/ http://myolympus.org/ photo sharing site
From: David J Taylor on 2 May 2010 05:18
"Alfred Molon" <alfred_molon(a)yahoo.com> wrote in message news:MPG.26474df14f77f73c98c2cc(a)news.supernews.com... [] > For instance, the human eye cannot see colours at night. According to > your logic, camera sensors should switch to black and white at night. > -- > > Alfred Molon I find that low-light night-time shots sometimes look a much more realistic when converted to monochrome - it also avoids those nasty issues of mixed lighting sources.... Cheers, David |