[D1N0]

DPreview seem the think the pixel shift feature should be a forget about and shoot feature, and when they get bad results they blame the feature. Camera's should be dumbed down to a level where any n00b can get an award winning shot. Where is the fun in that? You have a time difference because you need different shots. That has it's limitations, that can to some extent be corrected, but you have to take limitations into account in order to get good results en be prepared to throw away a lot of pics that fail for some or other reason. Like this guy pixel shifting butterflies: Few butterflies and dragonflies close up with Pentax K3 II: Pentax SLR Talk Forum: Digital Photography Review

the conditions were just all wrong. A lot of movement, too much nearby foliage, low shutter speed (1/4 second) small aperture (f16 diffraction is going to negate pixel shift). The K-1 is not a magic box you can demand great pics from without some effort of your own and a bit of use of your mental facilities. It's not just pentaxians railing at dpreview. A lot of interest from other brandly shooters as well.

[dafbp]

They must be upset because not only is there a Pentax in their "Most Popular Cameras" list, but the K-1 is 2nd place.
And to think that once upon a time I would go to their site looking for Pentax news...

[Kunzite]

The K-1 in 2nd place? They mistakenly started counting some K-1-related article, it seems
(The worst offender is Flickr, with their camera statistics... they just can't be bothered to count selected Pentax models).

[LensBeginner]

Look, this F1 car won't go well on gravel, I want my money back!!!

[monochrome]

Basic sociology. You have to have an 'Other' to identify a group as not Other. Pentax is the universal Other.

Dpreview either believes it's meme and cannot see their faults, or intentionally supports the meme by arranging these false premises for tests and letting Pentax fail them.

Either way, CaNikOny users feel secure and Pentax user outrage raises their Google Profile hit rate.

Ignore it and use your camera.

[Sandy Hancock]

Ignore it and use your camera.

^This.

[Rondec]

Your shutter speed has to be fast enough, whether or not you are using pixel shift. The idea that you can use 1/4 second shutter speed with a fair amount of breeze is silly. Even if you aren't using pixel shift, you will still get a bunch of blurred foliage.

On the other hand on a reasonably still day, the amount of detail you get is pretty impressive.

This is a shot from a couple of mornings ago.

[carolina_sky]

I did a very explicit and relatively controlled test last night shooting Palouse Falls in Washington. A tricky subject that I wouldn't have tried with my K3ii. I think PS with motion correction worked beautifully. I took a PS and non-PS pic back-to-back (so say within 1 minute), same exposure, etc.
Taken with Pentax 24-70 @24mm, ISO 100, F18, 1.0 second exposure
The Pixel Shift was processed first by importing to DCU5.5.1 (aka SilkyPix?). No settings changed except for ensuring that PS Motion Capture was turned on. Exported as TIF. Imported TIF to Lightroom.
Both pics processed in Lightroom with *identical* settings (slight boost to exposure and boosted shadows, slight adjust to tone, upped sharpening to 100 in both).

The first pair is the non-PS and a 1:1 crop of a small area.
The second pair is the PS with same (well, close, couldn't be bothered to exactly line them up) 1:1 crop.
The rendered JPGs are 50% size and 90% quality. Same settings for both.

I noticed two things:
1) some difference in brightness and color depth at full size - not too visible here by the time it's sized reduced and uploaded
2) significant difference in detail level at 1:1 -still visible here
(bonus observation: DCU5.5.1 did a great job removing motion artifacts from the water while preserving detail in the rest of the pic - exactly as advertised)

[CDW]

DPReview has apparently just pulled the 'review' of K1 pixel shift. Lots of complaints were posted and it is obvious the test was inept and bogus.

[bobell69]

In the interview on DC Watch the Ricoh engineer said motion correction doesn't work on detail like leaves. On the seminar recorded from CP+ 2016 Tanaka Kimio uses a waterfall scene to demostrate motion correction. The video can be found in the events section on the K-1 special site or on youtube. Even with these 6th+ generation shots you can still see quite a bit of detail on the static rocks. So yeah, motion correction ain't gonna work all that well with a scene that is mostly plants and leaves.

I just shot a series of up close images of waterfalls this past week. All shots were in RAW, I didn't use pixel shift. ISO was 100 and shutter speed was on average a third of a second or so in order to smooth the water. But guess what? I had motion in the leaves because in any waterfall picture you are going to experience movement of the greenery around the falls when using a slow shutter speed. The falls produces its own climate, there is always wind. DPReviews test was a failure because no camera under those conditions would be able to capture freeze frame action using slow shutter speeds.

[kh1234567890]

Funny, posting a front-page article bashing a camera's feature before the camera itself is reviewed.

It's gone 404 now

[Not a Number]

It seems DPR did not use DCU or another RAW processor that can develop Pixel Shift with Motion Correction. FAIL!

https://www.pentaxforums.com/forums/190-pentax-k-1/322518-looks-like-dpr-drop...ift-study.html

[Mock]

It seems DPR did not use DCU or another RAW processor that can develop Pixel Shift with Motion Correction. FAIL!

https://www.pentaxforums.com/forums/190-pentax-k-1/322518-looks-like-dpr-drop...ift-study.html

Man, we figured this out weeks ago. I'm so disappointed.

DPR should invest in some espionage, infiltrate the so-called "other forum" and study what those lowly creatures are talking about, they could have avoided this whole fiasco.

[luftfluss]

I just shot a series of up close images of waterfalls this past week. All shots were in RAW, I didn't use pixel shift. ISO was 100 and shutter speed was on average a third of a second or so in order to smooth the water. But guess what? I had motion in the leaves because in any waterfall picture you are going to experience movement of the greenery around the falls when using a slow shutter speed. The falls produces its own climate, there is always wind. DPReviews test was a failure because no camera under those conditions would be able to capture freeze frame action using slow shutter speeds.

It's gone 404 now

This has been posted by Barney Britton:

Yesterday we published an article examining the performance of the Pentax K-1’s Pixel Shift mode when shooting waterfalls – a common landscape photography subject. We found various issues, including movement artifacts and false colors. It has become clear that some of these issues (and others related to Motion Correction) are unique to Adobe Camera Raw. When this became clear, we pledged to update our article with a more detailed look at performance in other Raw converters, including the bundled Raw converter, SilkyPix.
However, given that this is a rather time-consuming job, we’ve decided to temporarily pull our original article instead of updating it piecemeal. We’ll be publishing an updated analysis as soon as possible. We’d like to apologize for any confusion that the original article may have caused.

[monochrome]

I don't go over there very often, but I have the impression Barney Britton is a pretty straight shooter. I'm fairly astonished, though, that a purportedly competent review site could make such a glaringly obvious error; and even admitting the error, do it with weasel excuses about testing other RAW converters, which is 'time consuming.'

I want to say, "Boo Hoo Hoo RTFM," but I won't because that would just be a childish retort. But geez, ya know, it says DCU 5 in the User Manual.

Just tell the truth. "We didn't read the manual. We used the wrong converter. Our bad. We have to redo the test. We're taking down the article."