[Laurentiu Cristofor]

Would be very grateful for references on this

Their shots are proof of this. I looked at enough test shots on both dpreview and imaging-resource to realize that maintaining the same focus is not part of their goals. It may happen in some shots, but it is not guaranteed to happen by their testing methodology. Thus, it is important to verify that focusing does indeed happen in the same area, before you can use their test shots to compare 100% crops. At 100%, the perceived DOF is small even at large apertures and particularly for high ISO shots where details are lost to noise, it is important to make sure that you are actually seeing the differences caused by noise instead of differences caused by focusing.

[UnknownVT]

Their shots are proof of this. I looked at enough test shots on both dpreview and imaging-resource to realize that maintaining the same focus is not part of their goals. It may happen in some shots, but it is not guaranteed to happen by their testing methodology. Thus, it is important to verify that focusing does indeed happen in the same area, before you can use their test shots to compare 100% crops. At 100%, the perceived DOF is small even at large apertures and particularly for high ISO shots where details are lost to noise, it is important to make sure that you are actually seeing the differences caused by noise instead of differences caused by focusing.

I have studied test shots for a long time ....
in all honesty - cannot say I agree with you.

Let's just agree to disagree - on friendly terms -
have a very happy new year.

[mysticcowboy]

One thing that ISO tests don't measure is how much detail you can pull out of shadows. I'll often underexpose my K-5 in low light situations by a full stop, then push the exposure in Lightroom. The result is often better above 1600 ISO. The D7000 has similar dark area detail, with a tiny bit more detail but a tiny bit more chroma noise. The Kr doesn't do as well working this way.

[Laurentiu Cristofor]

in all honesty - cannot say I agree with you.

No problem, my goal is not for you to agree with me.

The dpreview K-5 shots are backfocused compared to the K-x/K-r ones. Anyone can see it in the ISO 200 shots when noise doesn't muddle everything - the bottom of the piece of paper is telling: in the K-5 shots the lines are already fading out of focus. That is why if we look in the back of the scene, on the Q card, for example, the K-5 shot will look sharper, but in the front, on the watch face, the K-x shot will show crisper detail. Again, you don't have to agree - I'm including the snippet for general reference for anyone interested in the argument.

Attached Images

 

[UnknownVT]

No problem, my goal is not for you to agree with me.

The dpreview K-5 shots are backfocused compared to the K-x/K-r ones. Anyone can see it in the ISO 200 shots when noise doesn't muddle everything - the bottom of the piece of paper is telling: in the K-5 shots the lines are already fading out of focus. That is why if we look in the back of the scene, on the Q card, for example, the K-5 shot will look sharper, but in the front, on the watch face, the K-x shot will show crisper detail. Again, you don't have to agree - I'm including the snippet for general reference for anyone interested in the argument.

Respectfully, still can't agree,
from the exact same test picture -
Most relevant is the inclusion of a line-grid card set at an angle
specifically for the purpose for checking front/back-focus
shown in the overall shot near bottom center, the feint outline




amazing that we can be looking at the same photo,
and coming to different conclusions -
hence my posting -

Photography is a visual media -
and more often than not our eyes are the final judge -
but I will take the point that one's own judgement may not necessarily apply to anyone else.

Since it was the HighISO sample comparisons that were called into doubt -
to be pedantically correct as every test picture is a separate photo -
the relevant comparisons for front/back focus checks:
ISO6400 Jpg -


ISO6400 RAW:


ISO12800 Jpg:


ISO12800 RAW:

[Laurentiu Cristofor]

amazing that we can be looking at the same photo,
and coming to different conclusions -
hence my posting -

It's not amazing, it's called confirmation bias. I look at an area where one shot is clearly getting out of focus in the front and I conclude that that shot is backfocused compared to the others. You ignore that and you're instead pointing at a tiny area where all shots appear in focus and you conclude that they are all focused on the same exact spot.

[UnknownVT]

It's not amazing, it's called confirmation bias. I look at an area where one shot is clearly getting out of focus in the front and I conclude that that shot is backfocused compared to the others. You ignore that and you're instead pointing at a tiny area where all shots appear in focus and you conclude that they are all focused on the same exact spot.

Interesting call about confirmation bias.

Out of the entire photo area -
the ramped line-grid is supposed to be
specifically for checking front/back focus errors -
this was the area I chose.

Whereas the area for your comparison
was hard to interpret (for me) -
and to be fair -
I picked the area that was specifically designed to check front/back focus.
(actually expecting to see the back-focus - but I didn't -
that's why I went to the trouble of looking at both the JPG and RAW samples at HighISO too)

Can we just agree to disagree, on friendly terms, please?

---

Last edited by UnknownVT; 01-05-2012 at 02:32 PM.

[Laurentiu Cristofor]

the ramped line-grid is supposed to be
specifically for checking front/back focus errors -

It does not matter what the grid is for, it matters how it is used in this particular case. And in this case, it is not used to determine focusing accuracy, it is just an element of the photographed scene - they try to focus on it, but they can't use it to determine the focus accuracy - that would require the grid to fill most of the frame.

Can we just agree to disagree, on friendly terms, please?

I'll have to disagree on this too, because I think that "agreeing to disagree" is just plain stupid. We don't have to agree on anything, so we don't need to appear as if we're agreeing on something. I just like to provide arguments for arguments, so don't take my replies as an effort to convince *you* - they are for other readers as well. I like to debunk fallacies. When I get bored or too busy, I don't bother anymore. But I never, ever, "agree to disagree". I am getting a bit bored though.

[UnknownVT]

I'll have to disagree on this too, because I think that "agreeing to disagree" is just plain stupid. We don't have to agree on anything, so we don't need to appear as if we're agreeing on something. I just like to provide arguments for arguments, so don't take my replies as an effort to convince *you* - they are for other readers as well. I like to debunk fallacies. When I get bored or too busy, I don't bother anymore. But I never, ever, "agree to disagree".

Hmmm... OK - I have not made any of my remarks personal -

I am getting a bit bored though.

Not to be disagreeable -
This is something I can agree with!