PDA

View Full Version : Is my 17mm f2.8 prone to poor contrast?


Zuiko
30th September 2012, 11:18 PM
In my recent thread of pictures (http://e-group.uk.net/forum/showthread.php?t=23231)taken with the 17mm f2.8 I mentioned that the lens seemed to have poorer contrast than many other lenses. Ulfric, who has used a 17mm for much longer than me replied that his lens has plenty of contrast. So is it just my perception, or my copy of the lens? Are other users of this lens able to comment?

Here are a couple of examples to illustrate what I've noticed about the contrast of this lens:-

http://e-group.uk.net/gallery/data/500/CRESSING_TEMPLE_BARNS_110_resized_ORF_no_processin g.jpg

This first picture is an ORF resized and saved as a JPEG wih no other processing. To me the contrast looks pretty low and makes the image very flat. The histogram is very good, indicating the need for only a tiny tweak in Levels (which I have not done), so this isn't an issue of poor exposure.

The next picture is the same Orf file but processed with a 35% increase in contrast. This was the only adjustment and it has noticably improved the picture in my opinion. 35% seems a large adjustment; normally to get the look I want I increase contrast in the range of 15 to 20%, using more only if I want to use it for particular dramatic effect.

http://e-group.uk.net/gallery/data/500/CRESSING_TEMPLE_BARNS_110_contrast_35_.jpg

Zuiko
30th September 2012, 11:32 PM
The second example is a little more severe. It was taken against the light (without a lens hood) and I suspect that this is responsible for much of the poor contrast. Once again it is an ORF resized and saved as a JPEG with no other processing.

http://e-group.uk.net/gallery/data/500/CRESSING_TEMPLE_BARNS_028_ORF_resized_no_processin g.jpg

And here it is with contrast increased by 40%, which seems a lot to me.

http://e-group.uk.net/gallery/data/500/CRESSING_TEMPLE_BARNS_028_Contrast_40_.jpg

Finally, the same image with some Levels adjustment as well.

http://e-group.uk.net/gallery/data/500/CRESSING_TEMPLE_BARNS_028_Levels_contrast_20_.jpg


So is there a contrast issue, or am I being too critical? Is it something to be concerned about as the files seem to respond well to processing anyway?

Zuiko
30th September 2012, 11:39 PM
Even more intriguing, why is there an obvious difference in these pictures when I view them in Windows Gallery but it becomes barely noticable when I post them on the forum? Surely the forum software doesn't apply any processing? Has anyone else noticed this peculiarity? :confused:

JohnI
30th September 2012, 11:43 PM
Might be interesting to see the results of a head-to- head comparison (same shot, same exposure values) with the lens mounted on both the E-P2 and E-M5. Yes, the two sensors are entirely different, but the relative lack of contrast - if it is the lens - should show up in both.

John

Zuiko
30th September 2012, 11:45 PM
Might be interesting to see the results of a head-to- head comparison (same shot, same exposure values) with the lens mounted on both the E-P2 and E-M5. Yes, the two sensors are entirely different, but the relative lack of contrast - if it is the lens - should show up in both.

John

Good suggestion, John, thanks for that.

Nick Temple-Fry
30th September 2012, 11:55 PM
Even more intriguing, why is there an obvious difference in these pictures when I view them in Windows Gallery but it becomes barely noticable when I post them on the forum? Surely the forum software doesn't apply any processing? Has anyone else noticed this peculiarity? :confused:

Well I can see the difference - I suspect that scrolling down rather than clicking between tends to obscure things a bit.

The same psychologists that allowed babies to plunge off chequered tables onto chequered floors (a test of the development of depth perception) probably have a theory to explain it.

And OK the early 1950s introduced chequed linoleum, and it was a great idea for wipe clean surfaces. But why did they leave a baby on a table unrestrained?

Sorry

Nick

Zuiko
1st October 2012, 12:08 AM
Well I can see the difference - I suspect that scrolling down rather than clicking between tends to obscure things a bit.

The same psychologists that allowed babies to plunge off chequered tables onto chequered floors (a test of the development of depth perception) probably have a theory to explain it.

And OK the early 1950s introduced chequed linoleum, and it was a great idea for wipe clean surfaces. But why did they leave a baby on a table unrestrained?

Sorry

Nick

You're right, Nick, opening another browser and clicking between images does show the difference; that's an interesting illusion when scrolling. Not sure about bouncing babies off lino, though - seems a harsh way to test perception! :eek:

Nick Temple-Fry
1st October 2012, 12:11 AM
I really shouldn't comment as I haven't got the lens or camera.

But the first shot of the reeds looks a bit overexposed, there are areas under the facing reeds where detail has been lost in shadow, but the image still shows them as quite light. To a degree the histogram shows that with a bit of a lhs tail.

But it's a dull light image, difficult to draw a conclusion.

Nick

Ulfric M Douglas
1st October 2012, 07:14 AM
I can't see you getting to the bottom (or even the top) of the problem without side-by side examples of the same scene taken with a lens which gives you higher contrast ... on the same body and at the same time.

Edit ;
I just had a look back at one of my lens test pages and I can detect a slightly lower contrast for my mZukio17mmF2.8 versus the Lumix14-45, which seems to be my contrast king, and other forum threads elsewhere have shown it too.
My 4/3rds 14-54MkI has exactly the same contrast as my 17mm, in my view, and that lens also has excellent contrast.
My 17mm has at least as much contrast as my collapsible mZuiko14-42 lenses.

Howi
1st October 2012, 10:30 AM
converting an ORF to JPG won't normally apply any level changes (depending on software of course), so raw files will tend to be flat compared to an OOC JPG.
Could be a combination of flat light conditions, slight over exposure etc rather than the lens.
as has been said, only true way to find out is comparison with other lenses on same shot.

snaarman
1st October 2012, 10:48 AM
That gate shot seems to be suffering from some general flare: What they sometimes call non-image forming light. In the old days it would be dust or grease on the uncoated lens, and it would have been much much worse. These days we expect lenses to be near perfect and that sort of shot is a harsh test indeed.

I find the otherwise excellent 45mm f1.8 lens needs my hand held lens hood* in many circumstances to avoid this problem. I forgive it because any lens that performed perfectly under heavy backlight would be beyond my financial means :)

Pete

*You hold your hand where the lens hood should be...

Zuiko
1st October 2012, 02:23 PM
Seems I've got some more controlled shooting to do if I want to get to the bottom of this. On the other hand nobody is saying that they can see a contrast problem, so it may just be me being hyper-critical based on a small number of shots. I'm certainly happy with how the shots from the day processed, which I suppose is the main thing. The bottom line is that, with a possible cautionary note regarding contrast, the 17mm performed rather better than I might have expected from it's reputation. :)

Bikie John
1st October 2012, 02:32 PM
John - have you tried other lenses on the same body? I wonder if maybe it is the body making the difference rather than the lens. Also you say that you have processed the ORF without any adjustments - I think that with some raw processors (Oly Viewer for example) they are sensitive to camera settings and will apply them by default, so perhaps you have contrast set to low on the body or something.

Sooo many variables to consider, the only way to get to the bottom of it is to change one at a time.

Ciao ... John

Ulfric M Douglas
1st October 2012, 06:12 PM
I'll just keep quoting this ;
... side-by side examples of the same scene taken with a lens which gives you higher contrast ... on the same body and at the same time...

Zuiko
1st October 2012, 07:19 PM
John - have you tried other lenses on the same body? I wonder if maybe it is the body making the difference rather than the lens. Also you say that you have processed the ORF without any adjustments - I think that with some raw processors (Oly Viewer for example) they are sensitive to camera settings and will apply them by default, so perhaps you have contrast set to low on the body or something.

Sooo many variables to consider, the only way to get to the bottom of it is to change one at a time.

Ciao ... John

That's a very fair point about the sensor, the E-P2 is new to me too.

Zuiko
1st October 2012, 07:20 PM
I'll just keep quoting this ;

I'll make sure that I remember it, Ulfric! *chr

Ulfric M Douglas
2nd October 2012, 07:34 AM
So why not actually do it John? <-- Not meant to stress you, really a straight question. :o
Have you no other lenses?
We KNOW you have lenses and an e-M5 too, it won't take long, just make sure the subject of the photo is not into the sun or a grey dog in the mist.

Howi
2nd October 2012, 10:12 AM
That's a very fair point about the sensor, the E-P2 is new to me too.

I think you are cluching at straws - is it the lens? - is it the sensor?
It is more likely to be light conditions - sometimes you just can't do much about it.
What about other photos you have taken with this camera/lens combination in better light conditions?

Zuiko
2nd October 2012, 05:08 PM
So why not actually do it John? <-- Not meant to stress you, really a straight question. :o
Have you no other lenses?
We KNOW you have lenses and an e-M5 too, it won't take long, just make sure the subject of the photo is not into the sun or a grey dog in the mist.

Well, the main thing is there seems to be a concensous that my lens doesn't have a contrast problem, which is reasurring as I may want to sell it (although I haven't fully decided at this stage). However, it would be an interesting exercise and should settle the matter beyond all doubt, so I'll probably do it during the next few weeks when I have time.

It's odd where all the time goes when you are retired! :confused: This week, for example: Yesterday I took my daughter to an appointment with the doctor, then onto school, followed by taking my wife shopping which involved brunch. After collecting my daughter from school I took her to Drama Club and basically that was Monday gone! Today I spent the morning sitting as a volunteer in the local gallery, then was called to the school to collect my daughter who was suffering quite badly with her asthma. Not really time to start much now before dinner. Tomorrow I have an early start and a full day at Hatfield Forest with another photographer, researching locations to take a group on Sunday. OK, this is playtime and an undeniable benefit of retirement! Thursday I visit my Mum and on Friday I'm meeting up with Peterbirder for coffee. Saturday we sadly say goodbye to our cats and take them to a new home (following the discovery that Freya has now developed an allergy to them that affects her asthma) and Sunday I'm back at Hatfield Forest with the group from our camera club!

I'm exhausted just writing about it! :D

Zuiko
2nd October 2012, 05:56 PM
I think you are cluching at straws - is it the lens? - is it the sensor?
It is more likely to be light conditions - sometimes you just can't do much about it.
What about other photos you have taken with this camera/lens combination in better light conditions?

That's the thing, so far it's just my observations and impressions based on a single day's shooting. See this thread: http://e-group.uk.net/forum/showthread.php?t=23231