[kari]

The generation of *ist cameras had strong AA filter
The generation of 10 MP CCD had weak AA filter
The generation of K20D/K-7 had semi-thick AA filter
K-x has even weaker filter than 10 MP CCD generation
K-5 has rather strong AA filter close to *ist generation.

Where did you find this info?

[Fl_Gulfer]

I thought that anyone that cared about photography worried about light not sharpness. All the best photo's have nothing to do with sharpness. It's speed, light & Clearity.
I'm not sure why you would want very sharp photo's.
To see all the flaws in your picture? Maybe if you are a designer, or a astronomer you would need sharp edges.

[Mscott]

Wait, im honestly confused by this subject... From what I understand these cameras are NEAR identical in IQ output, and were all tested on the default settings...

So, to increase the sharpness couln't we just change the sharp setting to +1 and move on?

I mean sure its softer than some of the compitition (or not). It seems to me its a non-issue.

[Parallax]

I thought that anyone that cared about photography worried about light not sharpness. All the best photo's have nothing to do with sharpness. It's speed, light & Clearity.
I'm not sure why you would want very sharp photo's.
To see all the flaws in your picture? Maybe if you are a designer, or a astronomer you would need sharp edges.

No, anybody that truly cares about photography knows that what it is really all about is picking apart equipment that you don't have, to make yourself feel better about what you do have.

For anybody that does put a lot of importance on the minutia, I would suggest paying far more attention to Falk's measured, quantifiable data than to others' interpretations of collected opinion.

---

Last edited by Parallax; 11-25-2010 at 09:04 AM.

[ogl]

Where did you find this info?

In test and reviews.

[ogl]

and you don't buy K-5 ? Ohh... it's not good

I'm waiting for DNG from K-5 + FA43. Then I will think.

[Porga]

Pentax cameras with 10 MP sensors never called SOFT. K-x too.

When K-x was out, it was same as now with K-5. Soft images, mirror slap, bla bla.

[JamieP]

I'm waiting for DNG from K-5 + FA43. Then I will think.

No need to think further. Just trust these opinions:

As pixel-peeper, I can say that 100% full-sized photos from K-5 is better than 14.6 MP.

MP doesn't spoil IQ. The crops are close to 10-12 MP pictures from Pentax in terms of sharpness. AA filter seems weaker than K-7. I'm really impressed. Better sensor?
PRIME II is the same as in K-7.

For me, IQ of K-5 kills Canon 7D.

It's very good. I'm impressed that IQ has no any traces of softness and "soapy" like 7D.

[robgo2]

As DSLRs mature and the gap between film and digital outputs narrow, life gets harder for those who aspire to be opinion shapers through the use of bench test and academic explanations. To stay relevant, they need a "man bites dog" storyline.

In the case of the K-5, unlike times past, they are late to the party. Their main talking points and the issues may have given me pause have been dissolved by examples posted by others for all to see. The cheese has moved. All that remains are a few fly-specs and crumbs to examine then blather endlessly about. Pretty soon even that will be devoured.

Anyway, if you happen to cross the path of a fellow "loose in the wild" enjoying his K-5 (or its successor), I hope you will take a moment and say "Hello..." It might just be me.

Best of success and enjoy the season...

I think that we have reached the point in this discussion where pixel peeping becomes counterproductive. As far as producing sharp, detailed images with great DR and low noise is concerned, the K-5 is a superb tool. Whether or not it is as sharp on a per pixel basis as some other camera is simply irrelevant to that end. To echo what others have stated: Stop all this measurebating and try taking some pictures that have impact and artistic merit. That is the proper goal of photography. My guess is that some (not all) of the measurebators don't have an artistic bone in their bodies, so they spend all of their time obsessing over technical minutiae. Jeeez!

Rob

[paperbag846]

As a fan of landscape photography, I can see where 1:1 output would be every bit as integral as that of reduced printing.. I do think the same sort of requirements come-in on various studio and media grade photography as well...

... differences in detail due to AA filtration may not be recoverable by post processing. Having said that... it would take a considerable amount of effort to setup a reliable testing method...

... seen or observed at 1:1(or 100%) where both the advantages and consequences are visible.
...
"good photo's" for some time now. However, as needs and requirements continue to move forward, I think everyone is in there right to expect a likely progression along the way also.

Personally, I'd like to see Pentax build a prosumer grade body.
IOW. A crop sensor system that would command the outmost in IQ and performance for the working professional that 'would not' cost 10 thousand dollars.

I've yet to see a 100%, in focus crop that was not sharper than my eye could see. Anti aliasing is a very important part of digital signal processing. Look at it this way: you have two choices (this goes with any digital sensor). You either lose a little detail, for argument's sake, bumping your 16 MP camera down to 15 MP. Or, you could keep it at 16MP and watch as your digital sensor generates all sorts of artifacts that do not exist in the original scene. As a landscape photogtapher, do you want colour moire to attach itself to the treeline? Would you like to process it out of the waves?

If you insist on judging files at 100%, judge both the product, and the alternative. I have seen pictures of this camera at 100% and people going "boo hoo" (likely people who are not used to owning a 16 MP ASP-C). But they aren't comparing it to anything. They compare it to a lower-MP camera and don't equalize the resolution, or the D7000, which is sharper and shows more colour moire.

If you are a working professional, colour moire would be worse. It would take you longer to process it out (costing you sharpness, by the way), and that costs you money.

Furthermore, exactly how large are you printing these landscapes? If sharpness at wall-sized prints is your thing you are going to have to stop being cheap and purchase yourself a FF or MF camera. You want your cake and to eat it too, and still complain. This argument is actually very hard for me to understand because people are posting pictures that are very sharp and then complaining that they are soft. The, someone shows me some Russian who shoots a vodka bottle in a poorly-lit room at ISO 800, and it doesn't even look like the poor Russian knows how to properly focus his camera!

Maybe one of the problems is that people forget that more megapixels means you are zooming in more, which is misleading when you are looking at pixel level sharpness. Print K10D, K20D, K5 prints exactly the same size, processed directly from raw. Tell me if you can see that the K5 is noticeably softer. Until you do that, stop complaining that the camera is not "professional".

And for god's sake stop talking about JPEGs. If you are this particular I can't understand for the life of me why you would shoot JPEGs.

I certainly do not hope that the K-3 has a weak AA filter which forces me to clean colour moire out of my photos one by one.

[Fl_Gulfer]

Amateurs worry about Cameras, Pros worry about Lenses, Photographers worry about Light.

[Asahiflex]

Ogl, I don't understand why you didn't reply to my question???

Anyway: I can offer you DNG's of my K-5 with FA43 and FA77 attached at any f/stop you want, but only after resizing them to K200D resolution. You do agree with this? Only then it makes sense to compare the K-5's IQ to the K200D.

[cbaytan]

Amateurs worry about Cameras, Pros worry about Lenses, Photographers worry about Light.

Well then you have to admit some of us not only amateurs, besides some of those obsessive-compulsive ones about sharpness, like me. I can't help myself as a close-up, macro shooter.

[JonPB]

In summary, the stock K-5 isn't a significant upgrade from the K10D for people who want to shoot nothing but test charts.

If you take pictures of things other than test charts, Klaus' findings don't seem to be relevant. Perceived sharpness of a final photographic image is a lot more complicated than an LW/PH value. And overall image quality is a lot more complicated than perceived sharpness, at least for me.

My experience with Pentax is that they make engineering decisions based on photographic needs. I have no doubt that they'll lose some sales to other camera companies who make engineering decisions based on expected tests by internet reviewers at the expense of photographic quality. And maybe they went too far in the photographer's interest, like with how they designed white balance for the K10D. [The K10D was designed to have an "automatic" response to light much like film does, where filters/manual white balance needed to be used to remove color casts from light sources like incandescent lamps.] But that's a preference, and I like that Pentax designs cameras for photographers rather than for photographic equipment testers.

I'm not knocking people who want to be able to have lots of sharpness in their images. I'm one of them. I'm just saying that Klaus' test isn't overly relevant to that pursuit.

That said, I'm very pleased that Photozone will soon be testing more Pentax lenses.

[ogl]

When K-x was out, it was same as now with K-5. Soft images, mirror slap, bla bla.

Mirror slap - yes.

AA filter -
Falk Lumo: Lumolabs: Pentax K-x sensor quality
The anti alias filter is very weak or absent as it allows for moiré down to three times the Nyquist limit. Color moiré is visible for textures at the Nyquist frequency.

No any bla-bla now. A lot of DNG and JPEG photos of K-5 to prove.