[telfish]

I'm sorry if i misted it, but what's the cure against the soft image?

VALIUM!

[dgaies]

VALIUM!

I think I could use one as well... some of the posts in this thread are bordering on ridiculous.

[JohnBee]

Here are a few more interesting samples that I think may prove useful in nailing this down.

1. D7000, 1.40s, f/5.6, ISO100
[IMGWIDELEFT]http://www.bertin.ca/tmp/K-5/D7000_CRP4.jpg[/IMGWIDELEFT]
2. K-5, 1/25s, f/5.6, ISO100
[IMGWIDELEFT]http://www.bertin.ca/tmp/K-5/K-5_CRP4.jpg[/IMGWIDELEFT]

The question now is... what would causes a camera to produce purple instead of black?
I'm thinking a very poor lens!

BTW. I have DPReview samples that are taken with better glass and look much better than these(I'll post them shortly).

[FullertonImages]

All the bad/soft images all look more like bad focus or bad optics to me. I seem to be getting pretty sharp results out of mine. But I do still think the K-7 is sharper at the pixel level.

[JohnBee]

All the bad/soft images all look more like bad focus or bad optics to me. I seem to be getting pretty sharp results out of mine. But I do still think the K-7 is sharper at the pixel level.

I am tempted to agree.
But what I find strange is that the K-7 was apparently tested with exact the same lens...
Which obviously raises more questions.

[Parallax]

The question now is... what would causes a camera to produce purple instead of black?
I'm thinking a very poor lens!

Poor lens and/or not focused. When doing the focus adjustments for my lenses with a black and white target, under fluorescent lights they will show either that exact same purple, or a green edge, depending on whether FF or BF. (I don't remember which color goes with which condition). That K5 sample just looks out of focus, to me.

[i83N]

Ok, same lens (35 macro limited) on K20D is more sharper, and image looks soft even on K-5 display so it's not raw file converter.

[FullertonImages]

I am tempted to agree.
But what I find strange is that the K-7 was apparently tested with exact the same lens...
Which obviously raises more questions.

So is it out-resolving the lens? I find it hard to believe that 1.4 megapixels would make a huge difference in that respect.

I'm ok with the reality that the K-5 is a bit softer than the K-7. In fact, that is what I was expecting based on the results of the K-x. But all these test show a lot of blooming, purple fringing and glowing whites, which typically are the result of poor focus of optics being used outside of their sweet spot. That is what make me want to say that the camera is capable of more than we are seeing.

The K-5 is still well within the range of acceptable sharpness. I did some environmental portraits for a law office the other day, and I'm still getting sharpness at the level of being mercilessly unforgiving on skin texture, and bring out every subtle skin flaw. Not that that is a bad thing, I'll just smooth it out in post. But what I'm saying is that, the camera is for damn sure sharp.

What I'm finding more than anything else, is what seems to be some pretty considerable AF inconsistency, which might be related to all these tests. In that very same portrait sesh, there were a lot of OOF, back-focused and front focused shots. I'm having a hard time getting all my lenses fine tuned in. And it seems that the lens-to-subject difference is affecting focus in a weird way. I'll correct a lens for BF from a certain distance, then shoot from a different distance and it will be FF'ing.

[nanok]

yes. that's a known fact. it is also known that the k20d was way behind "the competition" in high iso, unsatisfactory build quality compared to "the competition", and so on. the k-7 had the same problems as the k20d and a few others on top (compared, again, to this illusive "competition" ). if i could only find the time, i would.. let me try this way (we like polls, don't we):

1. print some test charts and start shooting them with my k-5 (and pay somebody to shoot them with my k20d, so i don't have to touch that outdated unsatisfactory piece of gear)
2. take the k-5 out with my favorite lenses, maybe bring the k20d along as a second body (k-5 wide, k20+grip+tele.. hmmm), and shoot landscape ALL DAY.

now, which do you think? (hint: i've had the k-5 since monday and this



is all i have to show for it (yes, yes, my dog grasing, i've been told: very inspired) . if i had a roll of film, it would still be in there, waiting to be finished!; so which do you think?)

just my two cents worth..

[robgo2]

I just looked at the K-5's mannequin photos (jpegs, not raw) on IR. If those photos are soft I will eat the mannequin. I also compared them very carefully to the D7000's mannequin photos (not in the comparometer.) Without any doubt whatsoever, any differences in sharpness have to do with different focal planes and, possibly, variations in the lenses used. Parts of the D7000 images appear sharper (e.g. the brooch) while parts of the K-5 images appear sharper (e.g. the lips and the flowers.)

In summary, I do not believe that the K-5 suffers from any fundamental problem with sharpness. Apparent softness is the result of the testing procedures.

Rob

[dgaies]

In summary, I do not believe that the K-5 suffers from any fundamental problem with sharpness. Apparent softness is the result of the testing procedures.

My opinion as well.

[JohnBee]

I just looked at the K-5's mannequin photos (jpegs, not raw) on IR. If those photos are soft I will eat the mannequin. I also compared them very carefully to the D7000's mannequin photos (not in the comparometer.) Without any doubt whatsoever, any differences in sharpness have to do with different focal planes and, possibly, variations in the lenses used. Parts of the D7000 images appear sharper (e.g. the brooch) while parts of the K-5 images appear sharper (e.g. the lips and the flowers.)

In summary, I do not believe that the K-5 suffers from any fundamental problem with sharpness. Apparent softness is the result of the testing procedures.

Rob

Sounds very good.
Where can we find the D7K mannequin photo's?

BTW. the extreme PF example provided above was taken at center image. Which would make the lens very bad in terms of optical quality.

To tell you the truth, I have no idea why people conduct tests if the tests only confuse of raise more questions than when they started with.
I mean... what's the point?

[JeffJS]

As long as I get the focus correct and the exposure correct, I'm not seeing any soft images (at the focus point).

http://farm5.static.flickr.com/4018/5156599330_f316cc8fde_o.jpg

http://farm2.static.flickr.com/1070/5159189543_b80a3c258c_o.jpg

[glanglois]

Sounds very good.
Where can we find the D7K mannequin photo's?

BTW. the extreme PF example provided above was taken at center image. Which would make the lens very bad in terms of optical quality.

To tell you the truth, I have no idea why people conduct tests if the tests only confuse of raise more questions than when they started with.
I mean... what's the point?

I have the answers:

#1. It's a hobby, for some to test and for others to discuss the test. I'm not being snarky; some people just really like the investigation.

#2. It's a way to kill time owing to inclement weather/time of day/disinterest of subject.

#3. It doesn't remind me how bad my photos really are.

#4. It might give me an excuse for how bad my photos really are.

Wait for it ...

#5. It can drive reputation points up, points that can be used for free airfare anywhere in the world!

[dgaies]

Wait for it ...

#5. It can drive reputation points up, points that can be used for free airfare anywhere in the world!

Pep points awared... hopefully thay will get you one step closer to your ticket