[Ron Kruger]

ER1 is right: it is a subjective thing. I switched from Nikon to Pentax a few years ago because I liked the Pentax look. The Kx and Kr look like Canon/Nikon to me. Now this is a popular look, mostly through familiarity, but I suspected you wanted a richer, more slide-like look, and for that the K20D beats the Kr/Kx like a stepchild.

[dvest]

ER1 is right: it is a subjective thing. I switched from Nikon to Pentax a few years ago because I liked the Pentax look. The Kx and Kr look like Canon/Nikon to me. Now this is a popular look, mostly through familiarity, but I suspected you wanted a richer, more slide-like look, and for that the K20D beats the Kr/Kx like a stepchild.

Keeping the K-r and getting a K20D in a couple of months as a second body and using each to its strength might not be a bad idea. I would like that "rich, slide like look" you talk about. I'm also coming from Nikon and was hoping for this Pentax look I've heard about and seen in images before. Used K20D's are pretty reasonable. I usually like to have a second body anyway.

[Rondec]

i find it odd actually as i am shooting the k7 and it's not the dog i was led to belieeve. actually in most ways i find it better than my k10
a little (and i mean little) nosier at low ISO, above 400 though it beats the k10 hands down.
I still don't know how it did so poorly in DxO
don't care really either as it is doing the job for me which is all that really matters
jpeg performance i don't care as i don't shoot jpegs just raw. as long as the info is there i'll deal with any issue in post

I would agree with those who like the K10 for low iso shooting. The problem that I always had was that I got quite a bit of banding in shadow areas and that reduces the amount you can tweak shadows.

I like my K7 too and certainly don't think the sensor lacks that much in normal shooting situations.

[UnknownVT]

ER1 is right: it is a subjective thing.
but I suspected you wanted a richer, more slide-like look, and for that the K20D beats the Kr/Kx like a stepchild.

I always bow/cower to personal experience....
especially on the beating of step-children!

Probably the best bet is to look at standardized test shots like at Imaging-Resource.com.

Again as stated, this is obviously all subjective -

Using the ISO100 samples of the Indoor Portrait, No Flash

I thought -

K-r ISO100 sample (click on image to display full-size) seemed the most saturated and slightly dark (underexposed?) - so in that way the most slide-like.

K20D ISO100 sample (click on image to display full-size) seemed too pink and kind of artificial.

K-5 ISO100 sample (click on image to display full-size) looked a bit too yellow and just slightly over-saturated - almost like an attempt to make things look "sunny". Same applies to the K-5 ISO80 sample

K-x ISO100 sample (click on image to display full-size) examining this individually it looks well balanced. BUT in comparison with above looks almost pale - slightly under-saturated - but this is in comparison only - the most "natural"?

K-7 ISO100 sample (click on image to display full-size) of all the renditions I liked the K-7 ISO100 the best - it looks the most detailed without looking unnatural - could it be simply because of more "grain"?

So if I had to rank these based on these samples K-7, K-x, K-r, K-5 then K20D.

Of course I'll hear cries of -
not only do we beat our step-children - badly -
we don't shoot no stinking JPGs!

Ah!

But isn't the whole point of RAW being able to give almost any color balance/renditions one wants?

So if one likes a slide look -
then could that not be achieved with RAW?

In fact don't Fuji actually make cameras with various FujiChrome slide renditions - like Velvia?

OK go back to beating your stepchild.......

---

Last edited by UnknownVT; 02-15-2011 at 12:32 PM.

[eddie1960]

I always bow/cower to personal experience....
especially on the beating of step-children!

Probably the best bet is to look at standardized test shots like at Imaging-Resource.com.

Again as stated, this is obviously all subjective -

Using the ISO100 samples of the Indoor Portrait, No Flash

I thought -

K-r ISO100 sample (click on image to display full-size) seemed the most saturated and slightly dark (underexposed?) - so in that way the most slide-like.

K20D ISO100 sample (click on image to display full-size) seemed too pink and kind of artificial.

K-5 ISO100 sample (click on image to display full-size) looked a bit too yellow and just slightly over-saturated - almost like an attempt to make things look "sunny". Same applies to the K-5 ISO80 sample

K-x ISO100 sample (click on image to display full-size) examining this individually it looks well balanced. BUT in comparison with above looks almost pale - slightly under-saturated - but this is in comparison only - the most "natural"?

K-7 ISO100 sample (click on image to display full-size) of all the renditions I liked the K-7 ISO100 the best - it looks the most detailed without looking unnatural - could it be simply because of more "grain"?

So if I had to rank these based on these samples K-7, K-x, K-r, K-5 then K20D.

Of course I'll hear cries of -
not only do we beat our step-children - badly -
we don't shoot no stinking JPGs!

Ah!

But isn't the whole point of RAW being able to give almost any color balance/renditions one wants?

So if one likes a slide look -
then could that not be achieved with RAW?

In fact don't Fuji actually make cameras with various FujiChrome slide renditions - like Velvia?

OK go back to beating your stepchild.......

there are also several plugins for post like Nik and Dxo that let you emulate film types
I use the dxo film pack for this

for instance converted to emulate PanF with a gold tone added

[Emacs]

I believe this would now be the K-5, which has been the biggest game-changer for Pentax IMO.
The K110D/K100D/*ist D series all had the wonderful Sony 6Mp CCD, which they redeveloped into the 10Mp CCD in the K10D/K200D - all of these models offered amazing DR and image brilliance at ISO 100-200. The K20D and K-7 moved to CMOS, which apparently lost some DR but also performed worse in terms of long exposures and had slightly different colour profiles - more blues, less yellows. This is what kept me going with the K10D for so long. Not even the K-x or K-r sensors could compare to the Sony CCD sensor in this way.

But now the K-5 is out, the DR is simply amazing (rivalling the D3X and almost on par with the D700 according to DxOMark: DxOMark - DxOMark review for the Pentax K5) and given good rep here: Pentax K-5 best dynamic range of ANY digital camera!! - Steve's Digicams Forums

My pleasure was in seeing the K-5's noise handling in long exposures. It effectively creates smooth long exposure results easily meeting the K10D's ability to do so. Also the colours out of the 16Mp CMOS sensor are excellent and sharpness also great at 100%.

I guess you need to learn some math )))
14.1 of K-5 > 13.7 of D3x and, of course, more than 12.2 of D700

[Ron Kruger]

I appologize to all the stepchildren out there.
It's just a saying.

---

Last edited by Ron Kruger; 02-15-2011 at 01:51 PM. Reason: adding

[UnknownVT]

there are also several plugins for post like Nik and Dxo that let you emulate film types

In which case does it make any/much difference at Low ISOs between any of the Pentax dSLRs?

If we only go strictly by the numbers then the K-5 is tops followed by the K-x and K-r - so where should one go from the K-r ? -

[er1kksen]

i find it odd actually as i am shooting the k7 and it's not the dog i was led to belieeve. actually in most ways i find it better than my k10
a little (and i mean little) nosier at low ISO, above 400 though it beats the k10 hands down.
I still don't know how it did so poorly in DxO
don't care really either as it is doing the job for me which is all that really matters
jpeg performance i don't care as i don't shoot jpegs just raw. as long as the info is there i'll deal with any issue in post

For that matter, why did the GX-20 stomp all over the K20D in DxO ranking? And why do so many other cameras that have the exact same sensors and circuitry as each other show such disparities?

Answer, imo: DxO is screwy and doesn't accurately represent real-world results.

Better idea is to get some RAW test files (Imaging-resource is a good source) taken in identical conditions and compare their characteristics in the RAW converter of your choice. Don't trust the measurements.

[eddie1960]

I don't know if DXO ranks differently based on the category they drop a camera in? it would explain some disparities, but then they've introduced bias into the equation again and invalidated all the data interpretations. For the GX20 vs the K20 you think they would twig to maybe having a problem body and request a second to confirm the result given they are the same camera

Real world the K7 is a pretty fine camera IMHO after a couple of months of steady shooting. It did take a little getting used to after the K10, and i'm still discovering features and exploring its potential. i haven't shot a great deal of high iso, but what i have shot blows the k10 out of the water. it's no k5 but then again i can pick up a couple of them at current prices for what a k5 costs.

[UnknownVT]

For that matter, why did the GX-20 stomp all over the K20D in DxO ranking? And why do so many other cameras that have the exact same sensors and circuitry as each other show such disparities?

Probably because sensors are close -
and due to some margin of experimental error.

If we look at the test panels of the Samsung GX20 and the K20D side-by-side -
despite the seemingly wide gap of ranking #33 (GX20) and #50 (K20D) -
there really isn't that much difference -

of course the differences are measurable - that's why there's a difference -
but a difference of 0.1 EVs in Dynamic range -
gimme a break - are we likely to see any of this in real-life visual results?

[dvest]

Could be just a sample to sample difference. It's the same with lenses. You can take 2 samples of the same model and one will be a little better, or a lot better depending on quality control many times.

DXO is useful in some ways but there are some aspects that can't be measured in my opinion. I've seen it said "Not everything that matters can be measured and not everything that can be measured matters". I think that is true. You can't put numbers on the "Look" of and image. That is one of the things that matters to me. The look of an image.

[Rondec]

For that matter, why did the GX-20 stomp all over the K20D in DxO ranking? And why do so many other cameras that have the exact same sensors and circuitry as each other show such disparities?

Answer, imo: DxO is screwy and doesn't accurately represent real-world results.

Better idea is to get some RAW test files (Imaging-resource is a good source) taken in identical conditions and compare their characteristics in the RAW converter of your choice. Don't trust the measurements.

This is just sample variation. It is like the differences in measurement between the K5 and D7000 sensors -- measurable, but not significant. You make it sound like there are huge differences between the K20 and GX-20, but they really measured very closely.

[er1kksen]

This is just sample variation. It is like the differences in measurement between the K5 and D7000 sensors -- measurable, but not significant. You make it sound like there are huge differences between the K20 and GX-20, but they really measured very closely.

Perhaps the language of my earlier post exaggerated a little; I was a little riled up by the claim (which struck me as ridiculous) that the K20D would beat the K-r "like a stepchild." I mean, really?

Anyways, 3 points in overall ranking isn't entirely minor by the site's own metric, and I also cited the many other comparisons between different cameras that also have identical sensors and circuitry that also show disparities. 3 points is more than enough for a bunch of forumites to get excited over, at the very least.

Of course it's sample variation, which begs the question; why not test several bodies of each, if DxO aims to be a paragon of objectivity? I stand by my statement that DxO ratings are only roughly useful, and that it's better in the end to get some actual RAW files and work with them in the RAW converter of your choice. That gives you a better feel for the characteristics of that camera's files, something that involves subtler differences than a half stop of measurable DR or color sensitivity here and there, but which may nonetheless be more important to the individual.

The Canon 40D measured better than the K20D in several key metrics according to most, but I found that I hated the files. The K20D was much more pleasant to work with. And then the K-x came along.

[DaveC]

How much of an iso upgrade are the K10, K20 and K7 from the K100?