06-23-2012, 02:47 PM
[quote name='photonius' timestamp='1340362408' post='19023']
Well, from own empirical observations, when taking pictures through poor window glass with tele, results where clearly affected, distorted, while with wide-angle image was passable. I think because of the narrow angle, if the not-flat glass distorts rays, they end up further away from the point where they should "land" on the sensor than with wide angle lenses. I.e.let's say the piece of glass is a little like a convex lens (because it's not flat, it's a bit thicker in the middle of the filter), say 0.1 dioptre. A given dioptre has a stronger effect on long focal length lenses than on wide-angles, where there is less effect.
[/quote]
Actually, the real reason for less effect on a WA lens is that we are resolution limited by the sensor in cases like this. The larger AoV makes that the sensor has much fewer pixels available for a given subect, and distortion of that subject, than with a longer lens, and aberrations will therefore be less visible. OTOH, a WA lens has more DoF, so the whole pane surface may get into DoF, where with a long lens this may not be visible.
Having said that, even the worst filter I have ever come across is a lot better than the window pane you descrbe. I have shot with the 100-400L through double window panes, with good results. I lost a little contrast, but that was easily fixed in postprocessing. Check this:
![[Image: songthrush01.jpg]](http://www.lumifer.com/fs/songthrush01.jpg)
This is cropped to 60% or so, BTW - I had no choice. I only had time to take 4 shots before the songthrush disappeared again, 1 was unusable because of rapid subject movement, the other two were all similar to the above one. Taken through a (dirty, if not too bad) double glazing window pane, at an angle of about 60 degrees, with a filter mounted on the lens.
Regarding bokeh, I have a filter mounted as a standard, and only take it off if I use the 500D close-up filter for macro shots. I use B&W MRC filters, and here is another shot with the 100-400L with filter mounted as per default:
![[Image: flower-side-01.jpg]](http://www.lumifer.com/fs/flower-side-01.jpg)
Personally, I can't complain about the bokeh.
Do note that if one takes shots of objects with thin, elongated and/or angular objects in the background, like grass stalks or grain stalks, yes, the bokeh can get a little busy, and not necessarily pleasing. However, very few lenses are capable of handling that well in the first place, although the 100-400L does remarkably well at that (as both these pictures show).
In short, IMO, to parpahrase Mark Twain, the rumours of the effects of filters on bokeh have been greatly exaggerated.
Kind regards, Wim
Well, from own empirical observations, when taking pictures through poor window glass with tele, results where clearly affected, distorted, while with wide-angle image was passable. I think because of the narrow angle, if the not-flat glass distorts rays, they end up further away from the point where they should "land" on the sensor than with wide angle lenses. I.e.let's say the piece of glass is a little like a convex lens (because it's not flat, it's a bit thicker in the middle of the filter), say 0.1 dioptre. A given dioptre has a stronger effect on long focal length lenses than on wide-angles, where there is less effect.
[/quote]
Actually, the real reason for less effect on a WA lens is that we are resolution limited by the sensor in cases like this. The larger AoV makes that the sensor has much fewer pixels available for a given subect, and distortion of that subject, than with a longer lens, and aberrations will therefore be less visible. OTOH, a WA lens has more DoF, so the whole pane surface may get into DoF, where with a long lens this may not be visible.
Having said that, even the worst filter I have ever come across is a lot better than the window pane you descrbe. I have shot with the 100-400L through double window panes, with good results. I lost a little contrast, but that was easily fixed in postprocessing. Check this:
This is cropped to 60% or so, BTW - I had no choice. I only had time to take 4 shots before the songthrush disappeared again, 1 was unusable because of rapid subject movement, the other two were all similar to the above one. Taken through a (dirty, if not too bad) double glazing window pane, at an angle of about 60 degrees, with a filter mounted on the lens.
Regarding bokeh, I have a filter mounted as a standard, and only take it off if I use the 500D close-up filter for macro shots. I use B&W MRC filters, and here is another shot with the 100-400L with filter mounted as per default:
Personally, I can't complain about the bokeh.
Do note that if one takes shots of objects with thin, elongated and/or angular objects in the background, like grass stalks or grain stalks, yes, the bokeh can get a little busy, and not necessarily pleasing. However, very few lenses are capable of handling that well in the first place, although the 100-400L does remarkably well at that (as both these pictures show).
In short, IMO, to parpahrase Mark Twain, the rumours of the effects of filters on bokeh have been greatly exaggerated.
Kind regards, Wim
Gear: Canon EOS R with 3 primes and 2 zooms, 4 EF-R adapters, Canon EOS 5 (analog), 9 Canon EF primes, a lone Canon EF zoom, 2 extenders, 2 converters, tubes; Olympus OM-D 1 Mk II & Pen F with 12 primes, 6 zooms, and 3 Metabones EF-MFT adapters ....