An unscientific lens test - any expert feedback?
 

I just spent my Sunday afternoon doing some rather unscientific testing of
my new Canon 17-40mm lens. Despite having a digital P&S for several years
(and using the manual controls quite a bit) I found myself like a fish out
of water with a new DSLR. Suddenly I had a camera where changing the
aperture actually does something!

So I set about a bit of experimentation to find out how best to use the new
camera and lens. Is it better wide open or stopped right down? Zoomed to
17mm or 40mm? Embarrassing as it is to admit, I didn't have a clue. So I
set about finding out. The results are here:

http://www.derekfountain.org/17_40.php (2 largish images on that page)

I'd be interested in any feedback any experts could give me. I know my
testing was unscientific, but I drew some conclusions from it anyway and
intend to put those conclusions into practise. If anyone thinks I've drawn
the wrong conclusions, or can offer input as to why things are so, and
whether they are so with most lenses (or not) I'd like to hear about it.
I've clearly got a lot of learning to do!

--
The email address used to post is a spam pit. Contact me at
http://www.derekfountain.org : <a
href="http://www.derekfountain.org/">Derek Fountain</a>

Re: An unscientific lens test - any expert feedback?
 

Seems to me that you may be missing the point here, a higher F number will
increase your depth of field so photographing a flat item may not prove what
should be loking at.

By the way I am Perth boy, ex CBC Leederville.

"Derek Fountain" <nospam@example.com> wrote in message
news:423d6110$0$2730$5a62ac22@per-qv1-newsreader-01.iinet.net.au...
>I just spent my Sunday afternoon doing some rather unscientific testing of
> my new Canon 17-40mm lens. Despite having a digital P&S for several years
> (and using the manual controls quite a bit) I found myself like a fish out
> of water with a new DSLR. Suddenly I had a camera where changing the
> aperture actually does something!
>
> So I set about a bit of experimentation to find out how best to use the
> new
> camera and lens. Is it better wide open or stopped right down? Zoomed to
> 17mm or 40mm? Embarrassing as it is to admit, I didn't have a clue. So I
> set about finding out. The results are here:
>
> http://www.derekfountain.org/17_40.php (2 largish images on that page)
>
> I'd be interested in any feedback any experts could give me. I know my
> testing was unscientific, but I drew some conclusions from it anyway and
> intend to put those conclusions into practise. If anyone thinks I've drawn
> the wrong conclusions, or can offer input as to why things are so, and
> whether they are so with most lenses (or not) I'd like to hear about it.
> I've clearly got a lot of learning to do!
>
> --
> The email address used to post is a spam pit. Contact me at
> http://www.derekfountain.org : <a
> href="http://www.derekfountain.org/">Derek Fountain</a>

Re: An unscientific lens test - any expert feedback?
 

Pete D wrote:

> Seems to me that you may be missing the point here, a higher F number will
> increase your depth of field so photographing a flat item may not prove
> what should be loking at.
>
> By the way I am Perth boy, ex CBC Leederville.

The photo I took this morning, just after sunrise, was from one of the beach
lookouts. It was a kind of landscape and I wanted a good depth of field to
get as much of the beach in focus as possible. I selected a higher f number
(maybe f16 or so), took the shot, and it came out distinctly soft. I didn't
understand why, but now it makes sense.

But I take your point. My testing looked at focusing on a fixed, flat
object, nothing else. I still need to research how to use those smaller
apertures to best effect. I suspect that, for smaller apertures, the lens
will perform better in stronger light than I had just after dawn, and in
larger DOF scenarios than a bit of newspaper taped to my wall. :)

--
The email address used to post is a spam pit. Contact me at
http://www.derekfountain.org : <a
href="http://www.derekfountain.org/">Derek Fountain</a>

Re: An unscientific lens test - any expert feedback?
 

Derek Fountain wrote:

> I just spent my Sunday afternoon doing some rather unscientific testing of
> my new Canon 17-40mm lens. Despite having a digital P&S for several years
> (and using the manual controls quite a bit) I found myself like a fish out
> of water with a new DSLR. Suddenly I had a camera where changing the
> aperture actually does something!
>
> So I set about a bit of experimentation to find out how best to use the new
> camera and lens. Is it better wide open or stopped right down? Zoomed to
> 17mm or 40mm? Embarrassing as it is to admit, I didn't have a clue. So I
> set about finding out. The results are here:
>
> http://www.derekfountain.org/17_40.php (2 largish images on that page)
>
> I'd be interested in any feedback any experts could give me. I know my
> testing was unscientific, but I drew some conclusions from it anyway and
> intend to put those conclusions into practise. If anyone thinks I've drawn
> the wrong conclusions, or can offer input as to why things are so, and
> whether they are so with most lenses (or not) I'd like to hear about it.
> I've clearly got a lot of learning to do!
>

Derek,
In general, I would say your tests are quite good.
Your page shows only the center of the image. What about the edges?
You may come to different conclusions if you test the edges.
The edges will likely be softer at wide apertures.
At higher f-stops diffraction dominates and becomes worse
at smaller f/stops. So there is definitely a compromise between
wide open (and all the aberrations) and stopped fully
down (diffraction). All my lenses, and those for which I've seen test
data, tend to be optimum at f/8 to f/11, but there are exceptions
(though I never seen an exception at the short focal length range).

Here is my lens test page:
http://clarkvision.com/imagedetail/r...lens-sharpness

I print multiple test targets, placing them target on a board so
that the full field of view is covered. You can keep your newspaper,
and add other test patterns as desired, but fill the field of view so
you can analyze the full field.

Does your DSLR have a depth of field mode? Depth of field mode
allows you to focus on near and far then the camera sets the
best focus and f/stop for the picture. In my experience, I find
I must stop down one more stop to make the picture sharp
(unless I'm above f/16, then I don't as diffractions just gets
worse).

Roger

Re: An unscientific lens test - any expert feedback?
 

In message
<423d7475$0$2746$5a62ac22@per-qv1-newsreader-01.iinet.net.au>,
Derek Fountain <nospam@example.com> wrote:

>The photo I took this morning, just after sunrise, was from one of the beach
>lookouts. It was a kind of landscape and I wanted a good depth of field to
>get as much of the beach in focus as possible. I selected a higher f number
>(maybe f16 or so), took the shot, and it came out distinctly soft. I didn't
>understand why, but now it makes sense.

When I use super-wide lenses in bright conditions, I generally use
aperture priority, set nice and low (3.5 - 5.6), unless I have a very
good reason to prioritize shutter speed. I used shutter-priority for a
while the first day I bought my 10-22mm, and the aperture was going down
to f/16 at times, and these images were totally blurred. Even f/10 was
visibly softer. You can't get sharp pictures with wide-lenses at small
apertures.

You're fighting three curves. The smaller the aperture, the deeper the
range that is close to maximum focus, but maximum focus itself also
softens, the smaller the aperture. On top of this, many lenses are also
a bit soft, especially in the corners, at the maximum aperture.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

Re: An unscientific lens test - any expert feedback?
 

Derek Fountain wrote:

>If anyone thinks I've drawn
> the wrong conclusions, or can offer input as to why things are so, and
> whether they are so with most lenses (or not) I'd like to hear about it.
> I've clearly got a lot of learning to do!
>

Tests look OK to me. What I do, rather than shooting newspapers, I shoot in
3D outside and manually focus on a detailed object about 20-30 feet from
the camera (on a tripod of course) and try to include something at the same
focus point in the edge of the frame as well if I can. Shooting this way
you can get an idea of what the whole frame looks like and also how the DOF
vs fstops looks so you can better judge where you have to shoot to get good
DOF, edge sharpness etc.

On my 11-22 zuiko from f4-f11 they all look almost identical, at f16 it
starts to get soft and is actually softer than it is wide open at this
setting. At f22 it's pretty soft and I'd never use it unless I had to,
which wouldn't be very often given the DOF this lens has. It also appears
to be as sharp at 11mm as it is at 22mm.

I did this same test on the 40-150 zuiko consumer zoom and it needed to be
stopped down to f4.5 before it got good on the short end, needed 2 stops
for the edge to get good. But needed 2 stops down to f8 on the long end and
even then the edge wasn't perfect. This zoom was better at the short end
than the long end like you noted on your lens. This one was returned..

I then tried the 50-200 zuiko pro lens and it was like the 11-22, very good
1 stop down on both ends and edges with no loss at the long end compared to
the short end and pretty sharp even wide open.

So you are right, there are comprimises to lens designs and performance.
Good lenses aren't cheap and is why the best wide zoom L lens is $1400, I
bet that one isn't softer/lower contrast on the long end like this cheaper
one is..
--

Stacey