50mm 1.8D and distortion, and other issues

[vladvlaz]

This is not really a question, just things which stir up feelings of

uneasiness!

 

I've only been into photography for a year. I bought a Nikon F55 + 28-

80G. Three months into shooting I bought the 50mm f/1.8D. Then I

purchased the old 75-300 4-5.6...

 

After buying a film scanner and shooting some slides I've been

getting more and more worried about lens quality, bokeh, sharpness,

etc which was not a big issue when printing 4x6 or with the cheap

enlarger in the uni darkroom.

 

The first thing I noticed which bugged me was that at 3200 dpi, after

inspecting a scene I shot with the 50 prime and the cheap zoom at 50

there was very little difference in quality, which is not what I

expected. What is worse I could not make up my mind which lens was

which since I'd forgotten the order in which I shot them. Another

thing with the prime lens which has the respect of so many on this

site is distortion. Whenever I shoot something with the film plane

not parallel to the object, there is a pronounced case of buildings

being wide at the bottom and getting narrower towards the top. Is

this normal for 50mm? I thought this lens would be free from

distortion, perhaps it is a different type of distortion that people

are talking about? Sometimes pictures almost look like wide-angle...

 

My brother has an Elan 7NE with the "cheap" zoom 28-90 which

sometimes (perhaps I am paranoid) seems sharper at around 60-70mm

than my 50! Why is all this happening??

 

I also can't make up my mind on whether I want to buy the 24 2.8, 35

2, 85 1.8, or 105 macro next. I would like a portrait lens but can't

afford the DC ones.

 

If you have any thoughts/opnions on the above, please share them, I

would be interested to hear them!

 

Thanks,

Vlad

[vladvlaz]

And why does it always remove my paragraphing?

[ilkka_nissila]

What you're seeing when you shoot with a tilted camera is called keystoning, or converging verticals or horizontals. It's not distortion. Distortion is present when straight lines of the scene are mapped to curved lines on the image. And no, the 50/1.8 doesn't have distortion.

 

As far as the sharpness of the 50 vs. the 28-80, there are a number of issues which you need to fix before making comparisons.

 

1) Shoot from a solid tripod, or if you hand-hold (not recommended), use at least 1/(4*FL) s mm shutter speeds (1/200 s or faster with the 50 mm; for lens testing I'd err on the faster side).

 

2) There is more difference in sharpness and contrast at wide apertures (f/2.8-f/5.6) between the prime and the zoom. You need these apertures if you want to keep the shutter speed high.

[Karim Ghantous]

Vlad wrote: "And why does it always remove my paragraphing?"

 

If you select 'Plain Text' then your paragraphs will be preserved PROVIDED that you use TWO returns to separate paragraphs.

 

If you select 'HTML' you need either a <p>...</p> surrounding paragraphs or a <br /><br /> tag between them.

[john_stockdale2]

Vlad, welcome to the real world of optics. As a previous poster has suggested, the way to check for distortion in a lens is to photograph a rectangular grid at right angles to the axis of the lens. Your 50/1.8 will be pretty good in this regard, almost certainly much better than your zooms. The effect of distortion is often not noticeable in most subjects, though.

 

If you have seen beautiful architectural pictures of tall buildings that have been taken from ground level, and where the verticals are parallel, it means that the film was parallel to the verticals*. This is done traditionally with view cameras with lens movements, so the the film stays vertical while the lens is shifted up. There are a few lenses for SLRs that allow this, such as the Nikon PC 28/f3.5

 

When the camera is simply pointed upwards, the non parallel verticals are not distortion. It's actually what we see, but our vision system processes the image because it knows how we want to see what is plainly vertical.

 

*= There are also computer techniques for fixing converging verticals (and nearly everything else, it seems)

[vladvlaz]

Thanks for your answers guys!

 

 

Yeah, John! I already trained myself to see white walls as yellow in tungsten light...

 

I just thought about it, and ofcourse! To exaggerate the effect if you come up close to a building and look up you will see obvious convergence.

[john_carter_rhodes]

Yup. Convergence of verticals isn't distortion. It's just the natural perspective.

 

John.

[lachaine]

To avoid that perspective distortion of buildings, do whatever you have to do to keep the lens perpendicular to the building. Use your zoom if you can't get the whole building in with your 50mm without pointing the camera upwards. You can either back farther away and use the long end of the zoom, or you can use the short end of the zoom.

 

On sharpness. The difference in "sharpness" between a zoom and a prime lens is really overblown. There might be a difference in contrast, though, and other qualities of the prime lens - little or no linear distortion (where straight lines in the subject end up as curved lines in the picture). Really though, other than that, the most worthwhile advantage of the 50mm prime lens over the zoom you have is that you get a larger maximum aperture.

[Robert_Lai]

Check out this discussion on

<a href="http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=0090WQ" >35mm f/1.4 lens properties</a>.<p>This has plenty of views of barrel distortion, light fall-off, and curvature of field for the 35mm f/1.4 and 50mm f/1.4 lenses.