[Rico]

Oh noes, here goes the wild and conspiratorial speculations again. If you come to these conclusions about camera testers I can't imagine what wild scenarios you make up for sports, politics or news. I see this kind of "logic" elsewhere and it's not pretty.

1. They're a bit lazy and have only one camera.
2. They don't have unlimited time to spend on shooting Pentax studio scene.
3. They use the same set up and tripod for all cameras and consider this part of the "even field" of testing
4. They have a standard lens per mount to enable cross body comparisons and don't want to change.
5. The lack of electronic shutter means they can't do the usual work around

It's unlikely that dpreview will get to the bottom of the shutter shock and find out how wide range of scenarios are affected or do any serious investigation. Unless there are no big releases in the coming weeks that is.

All dpreivew K-3 III sample galleries have a lot of soft shots. The recent DA2040 gallery being perhaps the exception. I think it's pretty clear that at least the camera at dpreview suffers from it.

There is nothing conspiratorial here house it's just the facts. DPR only uses "embedded Pentax DNG ACR profiles" i.e. DPR alters the images in ACR only for Pentax cameras. As far as I can tell all the other brands they first are using the native brand raw format nef, CR3, RAF etc plus they just use the "camera raw default" settings from the camera for all other brands to output a JPG but for Pentax they use DNG instead of PEF then set their own image settings in ACR which completely alters the image striping details.

Again in the ISO 800 image I downloaded if you output JPG's using the DPR "image settings" and the "camera raw defaults" the image produced by the DPR ACR image settings stripes out all kinds of detail compared to the camera raw default JPG.

And the only way to really understand what the comparison scene really looks like you have to download or compare any Pentax pixel-shift image they created of the comparison tool scene to actually see what the objects really look like. A good reference is the print of the people looking at the painting on the easel to the left of the Fab Four patch. The Pentax pixel shift images render more faithfully what it really looks like more than any other brand. An example is the highlight on the woman's dress just right of the dog. There is horizontal cross hatching in the highlights which doesn't show up without pixel shift. Or the highlight area behind the people standing the wall behind them has detail that only shows up correctly using pixel shift. Fuji Canon Nikon Sony all just smudge out all the real detail because they can not handle the actual cross hatching in the print. They are all just a simulation of what it actually looks like.

[Qwntm]

And still nobody else seems to be able to re-create the problem... That's the problem. Until it's reproducible, it's not a problem.


DPReviews response: No Comment.

[c.a.m]

FWIW, here's an introduction to DPReview's Studio Test Scene, and a glimpse into their approach, dated August 2016.

An introduction to our studio test scene: Digital Photography Review

Of note:

"Cameras are mounted securely on a macro rail on a heavily weighted-down tripod, to minimize external vibrations. Self timer and any available anti-shock modes are also employed to minimize the impact of shutter shock. When possible we shoot tethered using Capture One Pro, for precise focus adjustment."

"Interchangeable lens cameras are shot using prime lenses that offer around 85mm equivalent field-of-view – a decision that stems from our historical use of each brand's 50mm lenses on APS-C, which are generally very sharp and consistent across the frame when stopped down a little. The aim is to remove, as much as possible, the impact of the lens. Our testing has shown the use of dedicated own-brand primes to be more reliable than using multiple copies of third-party lenses."

"Experimental Error
Like all processes there are sources of variation (error), including differences in chart alignment, focus and lens performance over time. While we have done everything possible to minimize the impact of these errors (including using a large, easy-to-align chart, careful manual focusing and selecting copies of lenses that are used only for studio testing), it is impossible to eliminate experimental error altogether.

Our comparison tool makes it possible to identify differences that are within the realms of well-controlled error, so we trust our readers not to read too much into very slight differences in apparent performance."


- Craig

[Class A]

DPReviews response: No Comment.

That's not true.

Richard Butler responds to a user observing that he thinks there is a problem with the user's results as well.

I agree that the slower shutter speed image from the user looks a bit blurry but that could be due to slight defocus or a different sharpening level being applied.

It puzzles me that the slow shutter speed images of the DPReview scene look that bad. Yes, there is a small amount of vertical movement blur which is probably due to shutter shock but if a bit of sharpening is applied, most of the scene looks fine (the blur then only manifests itself as text appearing "distorted" and not as clear as it could be).

It seems there is definitely more going on than a bit of shutter shock (whether the latter is abormal or not) and the fact that they are using ACR is not helping at all. Earlier ACR support (or rather lack thereof) for the K-1 II caused magenta cast issues that were not a problem of the camera but rather a problem of ACR not being adapted to properly deal with K-1 II files.

[Rondec]

But it's not perfect in the corners.

The DA *55 should be fine on an APS-C camera. They aren't shooting at f1.4 and from f2/2.8 it should be nicely sharp. Regardless, I don't think the issue is with the lens, but something with their specific camera. If it really is shutter shock, then they should just extract a single image of their pixel shift series and use that for the time being. The point of the widget should not be to identify shutter shake or even lens flaws, but rather to compare performance of the sensor at different iso settings. Those sorts of issues can be mentioned in a review (as I'm sure they will), but not here.

[Kunzite]

The DA *55 should be fine on an APS-C camera. They aren't shooting at f1.4 and from f2/2.8 it should be nicely sharp. Regardless, I don't think the issue is with the lens, but something with their specific camera. If it really is shutter shock, then they should just extract a single image of their pixel shift series and use that for the time being. The point of the widget should not be to identify shutter shake or even lens flaws, but rather to compare performance of the sensor at different iso settings. Those sorts of issues can be mentioned in a review (as I'm sure they will), but not here.

I'm talking about this, to be specific:
Studio shot comparison: Digital Photography Review
You can check other corners, too. The Fujifilm appears "superior".

Relevance? Not so much; in real life the DA* will get excellent results.
It's only that some people would select the softer part of these images, ignoring the actual cause, and blaming the camera.

[house]

Pentax is in a troublesome place with the age of its glass. Even most lenses we consider "newish" are actually old and perform as such.

You can still get very nice images but aberrations tend to be more than the competition. Looking at that da2040 gallery is a bit painful when you zoom in. Looks great at low res though.

[LeeRunge]

Whether this is or isn’t a real issue the data will come from all the users eventually. I actually doubt this will impact sales much at all.

I watched the dpreview review on the K3 3 and found it very fair. Actually it basically sold it to me other than the price, I’ll need a Black Friday sale or something before I bite as 2k+ is just too much currently for me. But the camera looks great. Near D-500 AF-C! (I’m sure people are argueing which is better here somewhere but who cares if it’s near that it’s great).

I’ve been very critical of AF-C on Pentax’s since the K-3 and it stopped me from replacing that camera, but they did it (I said I’d be happy with D750 AF and they seem to have surpassed that).

That review sold a future Pentax to me once the price adjusts, the samples won’t have any effect, I’ll just tweak the settings to avoid it if it’s a real issue.

[Rush2112]

I agree wholeheartedly.

DPR really shouldn't have published the test shots without knowing if they were generally representative of the camera (clearly, they're not)... but I don't believe there's any malevolence at play, and - with respect - we're not presenting ourselves as positively as we might with some of the assumptions and doubts aired here (though, given DPR's history with Pentax and potential reputational damage from flawed testing, I do understand). Perhaps there's a weakness in their testing conditions, or in operation of the camera... or perhaps it's a problem with their camera unit, or simply one that exhibits shutter shock or mirror slap with their specific copy of the DA*55 lens, set up on their specific tripod and head, at the shutter speeds concerned. We don't know.

Rather than assuming the worst in DPR, I think we're better off disproving there's a general problem by replicating their tests, showing our results and presenting our evidence (as some here, and on DPR, have done). I'm fairly confident they'd re-do the tests given sufficient evidence that their initial results aren't the norm. I'm sure they did that with an earlier Pentax model where the lens was at fault (was it the FA77?)...

Think that is the main point. DPR should not have posted those results. Even with the disclaimer. Whether they accept it or not, people hold them to a higher standard. As far as technical review sites DPR and IR are and always have been the go to, top level sites, for comprehensive analysis. The most they could have done is issue a simple statement that they are working on getting images posted for the K3 Mk III but seem to be having some issues with the raw converter they are using, as well as some testing inconsistencies.

That's it. Nothing more.

Instead, they knew fully well the shit storm they would create by doing what they did, and later on have to reel it all back later after posting such controversial findings. They knew it, there is 100% certainty they knew it. How many times has it happened before? And controversy creates traffic. Both on the front end and back end.

It's so irresponsible and dishonest. And the "editor" BB, should be fired for it.

[Kunzite]

It's so irresponsible and dishonest. And the "editor" BB, should be fired for it.

Let's not exaggerate - we risk running out of reviewers, as Imaging Resource also had front focus issues with their samples

I guess the difference between the K-3iii and the D780 is that they were able to use EFCS for the latter, so the samples were shock free.
Of course, I don't support publishing such incorrect "image quality" samples. Mitigation should've been a simpler matter of avoiding the respective shutter speeds.

[texandrews]

It doesn't looks like dpthoughts' posts are held for moderation.

As someone pointed out, he only seems to comment about Pentax, and always negatively. Makes me wonder if it's Rishi Santal....

[Kunzite]

As someone pointed out, he only seems to comment about Pentax, and always negatively. Makes me wonder if it's Rishi Santal....

I'm sure he isn't. Of course, I have no proof - but I have no reason to believe it, either.

[BigMackCam]

they knew fully well the s*** storm they would create by doing what they did, and later on have to reel it all back later after posting such controversial findings. They knew it, there is 100% certainty they knew it. How many times has it happened before? And controversy creates traffic. Both on the front end and back end.

It's so irresponsible and dishonest. And the "editor" BB, should be fired for it.

I don't know what they knew... These days, I'm erring on giving them the benefit of the doubt and trusting that, in light of sufficient evidence, they'll post new test shots. The damage may already have been done, but all the great photos and user comments here on PentaxForums (not to mention the excellent independent reviews and even DPR TV's own positive video review) will be enough to convince most potential buyers how good the K-3III can be. Let's be honest, none but the truly uninformed might use DPR as their primary reference when researching Pentax gear

[robbiec]

As someone pointed out, he only seems to comment about Pentax, and always negatively. Makes me wonder if it's Rishi Santal....

Same thought has occurred to me more than once. I highly doubt it but it would be quite funny. I thought he was tracked down to Northern Europe at one stage? Had a bit of bad luck with a lower end Pentax and it unbalanced him...

[Class A]

I guess the difference between the K-3iii and the D780 is that they were able to use EFCS for the latter, so the samples were shock free.

If they used electronic shutter mode for the D780, why did they not publish the single image they extracted from the K-3 III PixelShift shot?

Publishing a fully processed PixelShift image would not be fair, but they managed to extract a single image from a PixelShift sequence (which is obtained using electronic shutter) so if they compensate for shutter shock for the D780, why not for the K-3 III?

I guess the argument is that while it is easy enough to select electronic shutter mode on a D780, it requires knowledge and extra work to obtain an image from the K-3 III that was taken using electronic shutter mode.

However, unless the D780 images are clearly flagged as being obtained using the electronic shutter mode, it is not adequate to assume that people will use this mode to avoid shutter-shock. They might not be aware of the issue or may be facing scenarios in which the use of electronic shutter is contraindicated.