small aperture test
in [Digital Cameras]
|
From: Rich on 4 Nov 2009 10:39 On Nov 4, 2:17 am, "Dudley Hanks" <dha...(a)blind-apertures.ca> wrote: > I've heard a lot about how the cropped sensor cameras are defraction limited > to around f/8 - f/11, so I thought I'd see what kind of an image my XSi puts > out at a small aperture. > > I snapped on my 50mm f/1.8 lens and set it up to take a picture at f/22, > with a shutter speed of 1 sec. > > How did it turn out? > > http://www.snaps.blind-apertures.ca/images/SmallWinterPortrait.jpg (quick > download) > > http://www.snaps.blind-apertures.ca/images/SelfPortraitWinter.jpg (full > size) > > Take Care, > Dudley With an APS-C sensor, you would lose about 1/2 your resolution (provided the lens is good) going from f8 to f22.
From: Chris Malcolm on 4 Nov 2009 12:49 Dudley Hanks <dhanks(a)blind-apertures.ca> wrote: > "David J Taylor" > <david-taylor(a)blueyonder.not-this-bit.nor-this.co.uk.invalid> wrote in > message news:t6aIm.1501$Ym4.551(a)text.news.virginmedia.com... >> "Dudley Hanks" <> wrote in message news:4U9Im.50459$Db2.29545(a)edtnps83... >>> >>> I've heard a lot about how the cropped sensor cameras are defraction >>> limited to around f/8 - f/11, so I thought I'd see what kind of an image >>> my XSi puts out at a small aperture. >>> >>> I snapped on my 50mm f/1.8 lens and set it up to take a picture at f/22, >>> with a shutter speed of 1 sec. >>> >>> How did it turn out? >>> >>> http://www.snaps.blind-apertures.ca/images/SmallWinterPortrait.jpg (quick >>> download) >>> >>> http://www.snaps.blind-apertures.ca/images/SelfPortraitWinter.jpg (full >>> size) >>> >>> Take Care, >>> Dudley >> >> Difficult to say, Dudley. Yes, the image isn't "tack sharp" (a term I >> loathe), so there could be some diffraction visible, but I'm also not >> convinced that the subject didn't move within the 1 second exposure! >> >> Cheers, >> David > Thanks, David, I'll try it again with an inanimate object, or a faster > shutter speed. > I suppose, if the test is to be useful, I should also take an equivalent pic > of the subject using a wider aperture so the two images can be compared. The diffraction limit of aperture is usually taken to be the last aperture in a decreasing series of sharper apertures, i.e., stopping down further makes the image softer because of diffraction. But that's not a fixed aperture, it depends on such things as the exact sensor pixel size (or crop factor) not just the nominal "1.5", on the resolution of the lens, and whether you're looking at the centre of the image or the edges or some compromise between the two. Why should it depend on those? Because the point at which an extra stop's worth of diffraction softening becomes larger than how much other kinds of lens aberration are being improved by stopping down obviously will depend on the size of those other errors. In other words better lenses will have larger sharpest apertures. I find for example on my Sony A350 that my general purpose zoom is usually sharpest at f8, but at its soft extremes that becomes f11, and my 50mm prime is sharpest at f5.6. This can only be established for your camera and each of your lenses by taking a comparative series of shots while varying the aperture. On zooms it may change with focal length. -- Chris Malcolm
From: Dudley Hanks on 4 Nov 2009 14:20 "Chris Malcolm" <cam(a)holyrood.ed.ac.uk> wrote in message news:7ldt5gF3cimtaU1(a)mid.individual.net... > Dudley Hanks <dhanks(a)blind-apertures.ca> wrote: > >> "David J Taylor" >> <david-taylor(a)blueyonder.not-this-bit.nor-this.co.uk.invalid> wrote in >> message news:t6aIm.1501$Ym4.551(a)text.news.virginmedia.com... >>> "Dudley Hanks" <> wrote in message >>> news:4U9Im.50459$Db2.29545(a)edtnps83... >>>> >>>> I've heard a lot about how the cropped sensor cameras are defraction >>>> limited to around f/8 - f/11, so I thought I'd see what kind of an >>>> image >>>> my XSi puts out at a small aperture. >>>> >>>> I snapped on my 50mm f/1.8 lens and set it up to take a picture at >>>> f/22, >>>> with a shutter speed of 1 sec. >>>> >>>> How did it turn out? >>>> >>>> http://www.snaps.blind-apertures.ca/images/SmallWinterPortrait.jpg >>>> (quick >>>> download) >>>> >>>> http://www.snaps.blind-apertures.ca/images/SelfPortraitWinter.jpg >>>> (full >>>> size) >>>> >>>> Take Care, >>>> Dudley >>> >>> Difficult to say, Dudley. Yes, the image isn't "tack sharp" (a term I >>> loathe), so there could be some diffraction visible, but I'm also not >>> convinced that the subject didn't move within the 1 second exposure! >>> >>> Cheers, >>> David > >> Thanks, David, I'll try it again with an inanimate object, or a faster >> shutter speed. > >> I suppose, if the test is to be useful, I should also take an equivalent >> pic >> of the subject using a wider aperture so the two images can be compared. > > The diffraction limit of aperture is usually taken to be the last > aperture in a decreasing series of sharper apertures, i.e., stopping > down further makes the image softer because of diffraction. But that's > not a fixed aperture, it depends on such things as the exact sensor > pixel size (or crop factor) not just the nominal "1.5", on the > resolution of the lens, and whether you're looking at the centre of > the image or the edges or some compromise between the two. Why > should it depend on those? Because the point at which an extra stop's > worth of diffraction softening becomes larger than how much other > kinds of lens aberration are being improved by stopping down obviously > will depend on the size of those other errors. In other words better > lenses will have larger sharpest apertures. > > I find for example on my Sony A350 that my general purpose zoom is > usually sharpest at f8, but at its soft extremes that becomes f11, and > my 50mm prime is sharpest at f5.6. > > This can only be established for your camera and each of your lenses > by taking a comparative series of shots while varying the aperture. On > zooms it may change with focal length. > > -- > Chris Malcolm Thanks, Chris, that's good info to have. This is a pretty cheap lens, and I think its a bit soft to begin with. I've always been a fan of mildly soft portraits, and this lens has worked well for that purpose. But its also given me a few nice and sharp pics as well. It'll be interesting to see how it performs across its full range... Take Care, Dudley
From: Better Info on 4 Nov 2009 17:47 On 4 Nov 2009 17:49:36 GMT, Chris Malcolm <cam(a)holyrood.ed.ac.uk> wrote: >Dudley Hanks <dhanks(a)blind-apertures.ca> wrote: > >> "David J Taylor" >> <david-taylor(a)blueyonder.not-this-bit.nor-this.co.uk.invalid> wrote in >> message news:t6aIm.1501$Ym4.551(a)text.news.virginmedia.com... >>> "Dudley Hanks" <> wrote in message news:4U9Im.50459$Db2.29545(a)edtnps83... >>>> >>>> I've heard a lot about how the cropped sensor cameras are defraction >>>> limited to around f/8 - f/11, so I thought I'd see what kind of an image >>>> my XSi puts out at a small aperture. >>>> >>>> I snapped on my 50mm f/1.8 lens and set it up to take a picture at f/22, >>>> with a shutter speed of 1 sec. >>>> >>>> How did it turn out? >>>> >>>> http://www.snaps.blind-apertures.ca/images/SmallWinterPortrait.jpg (quick >>>> download) >>>> >>>> http://www.snaps.blind-apertures.ca/images/SelfPortraitWinter.jpg (full >>>> size) >>>> >>>> Take Care, >>>> Dudley >>> >>> Difficult to say, Dudley. Yes, the image isn't "tack sharp" (a term I >>> loathe), so there could be some diffraction visible, but I'm also not >>> convinced that the subject didn't move within the 1 second exposure! >>> >>> Cheers, >>> David > >> Thanks, David, I'll try it again with an inanimate object, or a faster >> shutter speed. > >> I suppose, if the test is to be useful, I should also take an equivalent pic >> of the subject using a wider aperture so the two images can be compared. > >The diffraction limit of aperture is usually taken to be the last >aperture in a decreasing series of sharper apertures, i.e., stopping >down further makes the image softer because of diffraction. But that's >not a fixed aperture, it depends on such things as the exact sensor >pixel size (or crop factor) not just the nominal "1.5", on the >resolution of the lens, and whether you're looking at the centre of >the image or the edges or some compromise between the two. Why >should it depend on those? Because the point at which an extra stop's >worth of diffraction softening becomes larger than how much other >kinds of lens aberration are being improved by stopping down obviously >will depend on the size of those other errors. In other words better >lenses will have larger sharpest apertures. > >I find for example on my Sony A350 that my general purpose zoom is >usually sharpest at f8, but at its soft extremes that becomes f11, and >my 50mm prime is sharpest at f5.6. > >This can only be established for your camera and each of your lenses >by taking a comparative series of shots while varying the aperture. On >zooms it may change with focal length. Your test won't work. Unless the widest aperture of your lens is the very sharpest, that means that the lens is not of diffraction-limited quality, the very best there is. If you lose detail at any time that you open up the lens, then that means the lens is not diffraction-limited. If it's not diffraction limited you can't tell when diffraction is what is causing the softening. As in all DSLR glass, the softening you see is due to lenses not being of diffraction-limited quality, poor lens manufacturing. Usually only one stop is adequate because the defects are overridden by that particular aperture stop. Anything above and below it is showing lens-figure defects, not diffraction problems.
From: Paul Furman on 4 Nov 2009 22:17
Dudley Hanks wrote: > I've heard a lot about how the cropped sensor cameras are defraction limited > to around f/8 - f/11, so I thought I'd see what kind of an image my XSi puts > out at a small aperture. > > I snapped on my 50mm f/1.8 lens and set it up to take a picture at f/22, > with a shutter speed of 1 sec. > > How did it turn out? The only way to know is to do a test in series. It looks soft to me. Sometimes that's OK though, sometimes DOF is more important. This photo: http://www.flickr.com/photos/edgehill/4075980407 had DOF as a priority though I didn't exceed the diffraction limits and the lens is near optimum wide open. The power lines in the upper left are soft due to tilting the focus plane. > http://www.snaps.blind-apertures.ca/images/SmallWinterPortrait.jpg (quick > download) > > http://www.snaps.blind-apertures.ca/images/SelfPortraitWinter.jpg (full > size) > > Take Care, > Dudley > > -- Paul Furman www.edgehill.net www.baynatives.com all google groups messages filtered due to spam |