[JohnBee]

Hello Pentaxians.

Having snuffed the K-5 soft charges, I was wondering if anyone here would be interesting in tackling the infamous K-5 optical deficiency charge? You know, the one where the K-5 is either harder to achieve perfect shots, and of course more prone to PF and CA than with other models.

I'm not sure what a good test would be to this effects, but... I'm thinking that a set of images from previous camera's such as, the K10D, K20 and K-7 could provide us with enough evidence to draw some pretty valid conclusions.

What do you think?

[JonPB]

Isn't chromatic aberration and purple fringing (a) due to lens imperfection, (b) practically correctable with proper raw development parameters, and (c) a function of the imaging plane being very thin? Issues (a) and (b) are not the fault of the camera. And I haven't heard that issue (c) varies between one digital camera and another, much less one CMOS sensor and another.

My opinion is that if--if--the K-5 is a more demanding camera, that is due to it being a more capable camera. Large format systems are harder to achieve perfect shots with than compact cameras, but when the image is nailed the more demanding system is the more rewarding one. Which happens to be the main reason I use a SLR: compact enough that I'll carry it, and produces image quality good enough to justify carrying it, which is a balance every individual must strike for themselves.

Frankly, anyone who says that a K-5 produces, say, 3.2 pixels of chromatic aberration where the K10D produces only 2 pixels of the same, and claims that this makes the K-5 an inferior camera, doesn't understand the nature of megapixels and can be either educated or written off without any worry.

Unless someone shows me some evidence that they say is convincing that the K-5 truly produces a disproportionate amount of CA or PF, I don't see any reason to put any effort into showing otherwise.

[ManuH]

Unless someone shows me some evidence that they say is convincing that the K-5 truly produces a disproportionate amount of CA or PF, I don't see any reason to put any effort into showing otherwise.

For CA I'd agree, that's lens dependent and more pixels will show more of it. Besides, it's relatively easy to correct with lens profiles.

For PF this is another matter. I've seen shots of K-7 and K-5 side by side and it seems the K-5 is definitively more prone to PF. And this is more difficult to get rid of. LR is not very good at it for example.

[dexmus]

For CA I'd agree, that's lens dependent and more pixels will show more of it. Besides, it's relatively easy to correct with lens profiles.

For PF this is another matter. I've seen shots of K-7 and K-5 side by side and it seems the K-5 is definitively more prone to PF. And this is more difficult to get rid of. LR is not very good at it for example.

Sorry I did not understand that. Purple Fringing (PF) is caused by Chromatic Aberration. What do you mean it is another matter?

Are you referring to sensor bloom? That should not cause purple fringing. Why would sensor bloom show purple color? As I understand, PF is caused by the failure of the lens to focus all colors at same point and because AF systems in more sensitive to green color. The other colors combine to give purple in out of focus areas.

Why are we testing the camera for CA?

[robgo2]

For CA I'd agree, that's lens dependent and more pixels will show more of it. Besides, it's relatively easy to correct with lens profiles.

For PF this is another matter. I've seen shots of K-7 and K-5 side by side and it seems the K-5 is definitively more prone to PF. And this is more difficult to get rid of. LR is not very good at it for example.

Capture One 5 has a tool expressly for dealing with purple fringing. This is separate from its chromatic aberration tool. I don't know if there are any other raw converters that have such a capability.

Rob

[Manel Brand]

the K-5 is either harder to achieve perfect shots, and of course more prone to PF and CA than with other models

Just to add to the discussion (and because I didn't see this mentioned):

As everybody else, I do have some lens (new and older alike) that can exhibit CA and PF in demanding light situations, and I take for granted that this can occur with any camera and any lens in the same situations. Maybe I'm wrong but I don't have high-end gear to clarify this; also admit that a high-class lens is less prone to this issues but it comes with a price that I'm not willing to pay. Anyway, I've come to a conclusion in my practical experience that this problem is much more related to the actual f/number than to lens quality or sensor deficiency, thus a problem related to exposure, not only to the hardware itself; meaning that a particular lens could produce CA/PF free images at f/8 with a K-7 body, but at the same f/8 value could be inducing some CA/PF in a K-5 body. Could this be the case?

This is simple common sense as I'm no engineer or gear tester. Furthermore, I find no hassle at all removing that small deficiencies in post-processing.

Thanks

[Gimbal]

This PF showing up in just about every K-5 shoot is starting to annoy me, I was all set to get the K-5 but for the first time I'm beginning to hesitate.
Is PF the k-5s Achilles heel?

[Parallax]

.

What do you think?

I think that unless the pictures taken with the different cameras are of the same subject, under the same lighting, and through the same lens (not same model of lens; the same lens), the results are as moot as the sharpness comparison tests.

[Staffan]

This PF showing up in just about every K-5 shoot is starting to annoy me, I was all set to get the K-5 but for the first time I'm beginning to hesitate.
Is PF the k-5s Achilles heel?

Gimbal, first you seem to have a K-5 giving you a lot of PF pictures, but then you write that you were all set to get it?
Generally speaking, where is all the PF proof to be seen in real picture samples, where it is possible to establish the decisive conditions that were prevailing?

[Arpe]

When I'm looking at a pic, what's the difference between pf and ca?

[Gimbal]

Gimbal, first you seem to have a K-5 giving you a lot of PF pictures, but then you write that you were all set to get it?
Generally speaking, where is all the PF proof to be seen in real picture samples, where it is possible to establish the decisive conditions that were prevailing?

I don´t have one (I have been trying to get one for weeks though) but there are plenty of pictures on the net. The last pictures I looked at where examples from the 18-135 zoom (from Ned) and they where awful.

[kittykat46]

This PF showing up in just about every K-5 shoot is starting to annoy me, I was all set to get the K-5 but for the first time I'm beginning to hesitate.
Is PF the k-5s Achilles heel?

Its not likely a K-5 issue, because it appears lens dependent.
I have not had any PF issue on the DA*50-135 and my DA21, DA40 and DA70s.

[JohnBee]

Thanks for all of the suggestions and responses so far(well... for the most part).
So is there anyone who owns a K-5 along with another body that would be willing to take some contrasting images?

Having read claims such as the following(from jamesm007 dpreview):

OK, let me be clear. Its the K-5 that lacks macro-contrast because it has too much CA/PF. That lens on a K20D or K-7 would show normal amounts is my bet. If the very good performing Sigma 70mm (in terms of low CA/PF) has it in high amounts with the K-5 that's a problem; and it does in IR tests. The problem is it ruins macro-contrast. It can never be restored. Its a scientific trait of lens design and now sensor design not IMO.
So stop judging the lens. Its the K-5!!! It seems none of today high-megapixel high ISO sensors have been perfect except the K-x (close) and Samsungs K20D, K-7 (close). Thats how its panning out so far. Remember DXO does not look at the PF/CA of a sensor. Hmm wonder it thats the key trait they are missing.

I am reminded that talk is cheap and where verified facts are the only means from which to get answers. Mainly where theories(such as these) may sound logical enough(even convincing) but that doesn't necessarily mean they are empirical either.

[jpzk]

Thanks for all of the suggestions and responses so far(well... for the most part).
So is there anyone who owns a K-5 along with another body that would be willing to take some contrasting images?

Having read claims such as the following(from jamesm007 dpreview):

I am reminded that talk is cheap and where verified facts are the only means from which to get answers. Mainly where theories(such as these) may sound logical enough(even convincing) but that doesn't necessarily mean they are empirical either.

This James fellow ... did anyone bother to ask him on what scientific basis is he "big talking" about? Or it might as well be left ... all alone?

JP

[jamesm007]

Thanks for all of the suggestions and responses so far(well... for the most part).
So is there anyone who owns a K-5 along with another body that would be willing to take some contrasting images?

Having read claims such as the following(from jamesm007 dpreview):

I am reminded that talk is cheap and where verified facts are the only means from which to get answers. Mainly where theories(such as these) may sound logical enough(even convincing) but that doesn't necessarily mean they are empirical either.

Talk is cheap! everyone is forgetting one thing. The AA filter. It can make a 16mp sensor give the resolution of a 10mp camera. It also does something else to the light. So we have users saying they do see some PF with their K-5s but talk is cheap... I have been looking at all RAW examples I can get my hands on. In most say 90% of the time the K-5 has higher PF/CA versus the D7000. OK so talk is cheap I can't verify what I am seeing. Can anyone do a controled test between the K-5 and say K20D with a lens like the DA18-55mm II, at f/5.6 or something?

I have noticed Klaus of Photozone is having his AA filter removed however so he can get max sharpness. OK an AA filter can disperse light more if its strong as in the case of the K-5. If the pixels are very sensitive (bloom easy) than this would explain why I see soft edges and CA/PF compared to the D7000? In softening detail (light) or filtering light it may be doing something strange per say... Why does not the D7000 show this? Or does it. I can't test any of my theories can or will anyone else?

BTW I base my claims on the fact that we have tons of pics from the K-5 showing high PF/CA. It was noticed right away. I can show many persons post speaking of it.

This from ManuH "or PF this is another matter. I've seen shots of K-7 and K-5 side by side and it seems the K-5 is definitively more prone to PF. And this is more difficult to get rid of. LR is not very good at it for example."

John I know you are smart, look and you will find the truth, whatever it will be.