Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-24-2010, 11:37 AM   #46
Site Supporter




Join Date: Dec 2010
Location: Oregon
Posts: 44
Thanks PLANDIN85020 those shots are encouraging, but I hope someone can setup a scene in front of a bright window or lamp so that we can get a real comparison of the cameras. Again, these shots are encouraging, but the contrast in the second shot with the k5 is far less than the first one that showed CF.

12-24-2010, 12:11 PM   #47
Senior Member




Join Date: Jun 2010
Location: Northern Nevada
Posts: 198
QuoteOriginally posted by Dan Quote
There is no evidence that any camera can create purple fringing. Purple fringing is created by the lens. The concern that I have is that the K-5 exaggerates the effect.
Thank you, Dan, for clearing up (sharpening/post-processing ) my misunderstanding of this thread.
Prior to noting your terms "exhibits" and "exaggerates", I was inserting creates/generates/produces after the comments "prone to", "less prone to", "more prone to", "considerably more prone to", etc, in previous posts.
RayGunn
12-24-2010, 12:29 PM   #48
Senior Member




Join Date: Jun 2010
Location: Northern Nevada
Posts: 198
I do not understand purple fringing...

QuoteOriginally posted by PALADIN85020 Quote
For example, this is a shot of an M4 carbine which is backlit. ... (VERY strong backlighting).

(I would insert Paladin's first photo here but I do not know how.)
To my eye, it appears the backlighting is coming from the back and the left.
Why is there purple fringing only on the right side of the Aimpoint scope?
All the purple fringing I can remember seeing always exhibits this one-sided effect.
Is there an explanation of this purple fringing one-sidedness to which someone can direct me?
RayGunn
12-24-2010, 12:34 PM   #49
Senior Member




Join Date: Jun 2010
Location: Northern Nevada
Posts: 198
Hey, Paladin...

I REALLY like your second M4/Aimpoint photo.
RayGunn

12-24-2010, 12:48 PM   #50
Dan
Senior Member




Join Date: Dec 2007
Location: Pacific Northwest
Posts: 239
QuoteOriginally posted by PALADIN85020 Quote
My personal take on this is that PF is just a characteristic of CA with certain lenses when a high contrast subject is posed against backlighting. .... Note the purple fringing around the Aimpoint scope mounted on the rifle. ...
That photo shows lateral chromatic aberration, not purple fringing. Purple fringing is a separate phenomenon. Why do I say this?

(1) PF and lateral CA look very different. With lateral CA the fringes are different colors on different edges, either Red/Cyan or Blue/Yellow or some combination of the two, as seems to be the case in your example. With purple fringing there is only one color along the edges--purple. For an example of PF, see the photo in the Photozone link in the next paragraph.

(2) With lenses that have both problems, the two problems often behave very differently as the lens is stopped down. Consider, for example, the FA 35/2. It has lots of PF wide open, but little lateral CA. As with every lens, PF diminishes as the lens is stopped down. But with this lens lateral CA increases as the lens is stopped down. By f8 the PF has gone away but the CA is peaking. This behavior is noted in the Photozone review (page 2):

Pentax SMC-FA 35mm f/2 AL - Review / Test Report - Analysis

Klaus notes that "you can experience some purple fringing in extreme contrast conditions at wide-open aperture" but that "Lateral chromatic aberrations (color shadows at harsh contrast transitions) are very low at f/2". Lateral CA increases as the lens is stopped down, and by f/8 the problem is "on a medium level". But by this point the PF is gone. In short, lateral CA and PF can behave very differently as a lens is stopped down, suggesting that they are not simply different manifestations of the same phenomenon.

(3) Software that corrects for lateral CA does not remove purple fringing. This is why some software programs, such as Capture One Pro and DxO Optics Pro, contain separate corrections for purple fringing. The PF tool does not remove lateral CA, and the lateral CA correction does not remove PF. In Photoshop, the routine for eliminating PF is quite different (and less automated) than the routine for correcting lateral CA.

(4) There are some lenses that are free of purple fringing but which nonetheless exhibit lateral CA. For example, the Voigtlander 180mm APO-Lanthar that I mentioned above is "absolutely free" of purple fringing (according to Photozone, and my own tests), but it does have lateral CA (enough that my RAW profile for this lens contains a CA correction). On the other hand there are lenses with small amounts of lateral CA that are very prone to purple fringing. This is why the better test sites--such as Photozone--test for these problems separately.

From a practical standpoint, lateral CA and PF are very different phenomena, with very different behaviors, and with different corrections in post-processing.

Dan

Last edited by Dan; 12-24-2010 at 01:03 PM.
12-24-2010, 04:36 PM   #51
Senior Member




Join Date: Jun 2010
Location: Northern Nevada
Posts: 198
Thank you, Dan...

I have a much better understanding, thanks to you.
RayGunn

Well, OK, except for CA one-sidedness.
RG
12-25-2010, 09:31 AM   #52
Forum Member




Join Date: Jan 2010
Location: South SF Bay, CA
Posts: 71
I just wanted to post these photos as an illustration of the difference between PF and CA. This photo was taken with a K-7 and DA70 Limited at f2.4. First, the full-image:



Then some crops showing PF:







The last image also shows CA if you look at the holes in the brake-disc as well as the areas around the spokes of the wheels. You can see it as orange and blue glows on opposite edges.

Overall, the CA is barely noticeable. And was easily correctable in Lightroom. The PF is pretty bad though.

If the K-5 "emphasizes" PF more than the K-7, I'd hate to see what this image would have looked like if taken with a K-5.

(Is the DA70 known as having more PF than other lenses?)

Greg
12-25-2010, 02:35 PM   #53
Pentaxian
JohnBee's Avatar

Join Date: Sep 2008
Location: front of computer
Posts: 4,620
Original Poster
Well I'd agree that this does look bad.

However, I think its important to highlight that this is taken wide open and the condition that we could consider idea for PF are also well and accounted for in this particular scene.

However, I guess a good test might be to fit this same lens on a K-5 and K-7 and compare the CA/PF limits provided by each system. Or better yet... against a K20D

12-27-2010, 03:34 PM   #54
Veteran Member




Join Date: Nov 2008
Location: Trabzon/Turkey
Photos: Gallery
Posts: 1,010
QuoteOriginally posted by jamesm007 Quote

I have noticed Klaus of Photozone is having his AA filter removed however so he can get max sharpness.
Wondering if he put pictures somewhere he is taken with K-5 without AA filter somewhere on the internet you know of?
12-27-2010, 07:25 PM   #55
Site Supporter




Join Date: Jul 2008
Location: Detroit MI, USA
Posts: 503
DPR says the K-5 has a relatively weak AA filter -

"The only issue to be aware of as far as image quality is concerned, is the occasional appearance of jaggies in very fine diagonal lines, where they appear 'stepped'. This is almost certainly a consequence of the K-5's light AA filter. We saw this clearly in our studio test images (and it is reported here) but it should be stressed that in the hundreds of 'real world' pictures that we took with the K-5, we barely noticed it."

I believe if memory serves (what) that Falk L also found the AA filter on the K-5 weaker versus the K20D. So...

So why would Klaus remove the filter completely. I know he is a reviewer and tests lens not dSLRs so...

Why doesn't the D7000 show this, or does it? Did Pentax produce a tad more resolution (with weak AA filter) for its extra or really just a "tad" more NR; this is how Pentax pulled it off, So...

Why do I care? I hate PF, one of my first posts on DPR is when I bought the Tamron 70-300mm and could not believe what it was doing (loads of PF/CA). This was some years ago. I always remove any PF/CA that I can find. I find very little with the K20D and my current lens. To tell the truth my GX10 (same as K10D) only had a tad more PF/CA versus the K20D that I noticed in day to day shooting. So...

How can a sensor produce PF/CA, well they have little lens in front of each pixel and PF/CA is an artifact of sensors just as lens from my research. But to what extent. Is the K-5 producing extra PF? I don't know. And with 14 months of Pentax warranty and a free check-up I still have not used for my K20D I am in no hurry to try the K-5. So I am impressed but want to know everything before spending my hard earned money I am saving. I have also learned sometimes things take time to notice, like longer than the return period and it may not be a defect... so...

How much is too much PF, it's subjective I think.
12-27-2010, 08:10 PM   #56
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by cbaytan Quote
Wondering if he put pictures somewhere he is taken with K-5 without AA filter somewhere on the internet you know of?
I will not reveal internal information here. But let me say this much: I don't think Klaus is publishing images with the AA filter removed at the moment. Don't ask me for reasons though
12-27-2010, 08:14 PM   #57
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
Regarding the topic of the thread:

As long as nobody does a controlled side by side test with identical lens and subject and lighting and comparing 100% crops from, e.g. a K-5 and K-7, as long this entire thread is useless and should die.

The default assumption should be that a modern CMOS sensor doesn't amplify PF.
12-28-2010, 02:07 AM   #58
Veteran Member




Join Date: Nov 2008
Location: Trabzon/Turkey
Photos: Gallery
Posts: 1,010
QuoteOriginally posted by falconeye Quote
I will not reveal internal information here. But let me say this much: I don't think Klaus is publishing images with the AA filter removed at the moment. Don't ask me for reasons though
Ok Falk, I guess one of the reasons is:

From photozone forum:
--------------------------------------------------------------------
Klaus, on 23 November 2010 - 10:06 PM, said:

First impressions:

The JPEG quality is terrible (in terms of pixel-level-sharpness).
The RAW quality is Okayish - 25% blur - which is significant. Based on these initial tests I'd conclude that it has a rather aggressive AA filter.

The max. LW/PH are barely higher than the ones of the K10D. So far for the fanboy criticism that I should have migrated earlier ...


Just to clarify (because the former comment is already cited out there).

K5 @ 16mp:
max. 2500 LW/PH (RAW)
max. 2050 LW/PH (JPEG ****)

For comparison:

A33 @ 14mp:
max. 2850 LW/PH (RAW)
max. 2500 LW/PH (JPEG)

Other than that the K5 seems to be a very fine camera from a user's perspective. Please note that I'm simply requiring a camera for a lab environment and the K5 is not substantially "better" for the lens tests (and only for the lens tests) than the old K10D.
------------------------------------------------------------------

So Ogl was right eh? I also second Klaus, Pentax should offer a K-5 model without AA filter too. I would only buy that one.

Original post is at
Photozone - Pentax K5 ... ordered -

Last edited by cbaytan; 12-28-2010 at 02:13 AM.
12-28-2010, 03:13 AM   #59
Veteran Member




Join Date: Nov 2008
Location: Trabzon/Turkey
Photos: Gallery
Posts: 1,010
QuoteOriginally posted by falconeye Quote

The default assumption should be that a modern CMOS sensor doesn't amplify PF.
How'bout jpeg algorithm?
12-28-2010, 05:38 AM   #60
Site Supporter




Join Date: Jul 2008
Location: Detroit MI, USA
Posts: 503
QuoteOriginally posted by falconeye Quote
Regarding the topic of the thread:

As long as nobody does a controlled side by side test with identical lens and subject and lighting and comparing 100% crops from, e.g. a K-5 and K-7, as long this entire thread is useless and should die.

The default assumption should be that a modern CMOS sensor doesn't amplify PF.
I have over 80,000 plus pics on my past GX10 (same K10D) and now K20D including a lot of bird pics with my DA55-300mm (I used it on both). I have never seen, nor can I find any example with as much PF/CA as the pic above except with my old Tamron lens. Other people have also taken notice to this. Not saying there is a problem or some type of alarmist. But I don't like shutting the door on things and pretending it's not there.

Here is a list of some causes that included causes that a CMOS sensor can generate.(copied and pasted from http://wiki.panotools.org/Chromatic_aberration)
  • By sensor overflow visible as blooming. This blooming often has a purple color because of the bayer pattern of the sensor: If a sensor cell overflows to the neighboring cells all cells are affected the same. There are 50% green but only 25% red and 25% blue sensor cells. Hence blue and red are weighted more resulting in purple fringes around overexposed areas. Since CMOS sensors are not prone to overflow this affects only CCD sensors.
  • By chromatic aberration and other errors in the micro lenses that are in front of the sensor.
  • By reflections between the sensor and the protective glass (that might be coated and hence reflects colored).
  • By interpolation errors or anti moire filters.
  • By partial color saturation: Blue sky f.e. might be clipped to pure white because all color channels are saturated. If there is a blurred dark object in the image the sensors in the blur region receive only part of the light and hence don't saturate. The blur region appears blue. This is the only effect that applies to analog film as well.
A lens is a lens is a lens. CMOS sensor as well as CCD sensor have lens in front of each pixel well. The AA filter interacts with these lens according to all literature I can find. No one has said they have hard proof, only curiosity so why dismiss it? Especially another's (the OP) post or who ever has curiosity. The answer it not that the K-5 has more mega-pixels. The K20D and K-7 have 14.6 so they are pretty close. Something else is at play. I can show test pics from Imaging Resource of the same scene showing the K-5 as having more PF than the K-7. But just a tad more (but not being back lit!)

I don't think the K-5 has enough of a PF/CA problem that it would stop me from buying it. But it just might have more than the K-7 or K20D pixel-peeping, some only ask why. There are many tech/gear heads here that just want to know and get curious and its not a negative thing.

Thank you if you read it

kind regards.

Last edited by jamesm007; 12-28-2010 at 06:11 AM.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, dslr, k-5, k-5 ii, k-5 iis, k5, pentax k-5
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
lens evaluation houstonmacgregor Pentax SLR Lens Discussion 9 08-11-2010 03:08 PM
personal test and evaluation: Vivitar S1 90/2.5 + FD/K adapter Pentaxor Pentax SLR Lens Discussion 6 07-11-2010 07:59 AM
New lenses - pickup evaluation tips JVM Pentax SLR Lens Discussion 19 11-26-2009 06:36 PM
Taking the K-7 Where it Belongs... (my evaluation) Duplo Pentax DSLR Discussion 33 07-28-2009 01:46 PM
TEST SHOTS for LENS EVALUATION gkopeliadis Pentax SLR Lens Discussion 12 10-16-2008 08:39 AM



All times are GMT -7. The time now is 08:00 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top