noflashplease: Isn't it amazing how a 1/2.3" sensor represented the standard size for largely extinct low-end, consumer point-and-shoot cameras, but suddenly is appropriate for "flagship" smartphones? Of course, the iPhone still has an even smaller 1/3" sensor, which goes to show you the level of current smartphone hype.
I'm not sure what you'd advise using in smartphones instead. Panasonic had a pop with a 1" sensor and all it got them was a bulky, unpopular and very expensive phone. The photos look fantastic but it's hardly a market-winning proposition.
Thanks, DPR. The mea-culpa is hugely appreciated.
WillWeaverRVA: Pixel shift is just not meant for use with moving objects. It's more of a studio tool that can work for landscapes, architecture, and other static scenes.
I kind of want to see what the results would be with Pentax's provided Silkypix software, just to see if it handles the motion artifacts better. Supposedly it does.
It works pretty well with motion in this image.
Roadrunnerdeluxe: "Waterfails", nice!I honestly didn't expect it to be able to handle motion as tested here. I suppose it depends on what you are hoping to shoot whith pixelshift. For architectural work where there will be mostly moving clouds in my case, I will just over lay with a photo shot without pixel shift.For nature shots with lots of movement as tested here that would be too tedious though.
So these would be waterwins:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
thegreat26: Guys get over it. We didn't expect miracles and we knew what pixel shift is for from the beginning. There are no new technologies without compromises. Pentax at least risks to bring new technologies for the consumers( and Olympus apparently)....
Here's what the new technology does with the correct supporting software:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
The sooner Adobe catch up, the better.
KonstantinosK: Good luck finding motion-free landscape scenes. As I see it, the pixel shift technology may be good for interior architecture, products and portraiture of dead people.
Or you can take waterfall photos if you process correctly:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
Seeky: This is a bit like kicking in an open door... Although it seems disappointing, it is totally expected that landscape shots like these suffer when shooting in PSR. No camera hardware or software can correct for random movement of multiple elements in multiple directions in one scene, unless you are plugged into a supercomputer. I seriously doubt the usefullness of PSR unless you have a specific controlled environment.
Apparently this conventional wisdom is not 100% accurate:
Probably cherry-picked, but it is possible.
howieb101: I really think the green note, at bottom of article by DPReview should be clearer that Adobe Camera Raw (ACR) is the cause of the color cast. It just says that it is using ACR without clarifying its negative impact on the images.
On a side note, I have read many user reviews at BHPhotovideo and actual buyers of the camera seem to like pixel shift resolution. Each to their own, I guess.
I think this article is really a side track. You have to know when to use pixel shift and what its limitations are. If you can work within its limitations you'll get real benefits otherwise its a feature on the camera you can choose to ignore.
Personally I can see it useful for still life, certain types of landscape, macro, product photography and architecture.
Yet this image of a similar environment looks fantastic:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
TN Args: As expected. No disappointment here.
Except it's not even accurate:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
Franka T.L.: it does not take a rocket scientist to lnow that what's being asked for basically is not physically viable. Its not the camera or Pentax, its just a n in build limitstion of the sort regsrding thr techniques.
I'mma keep reposting this link wherever someone says this "cannot be done":https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
Greg VdB: A test I would like to see is the following: shoot a completely static scene (e.g. architecture, but with plenty of fine detail), once without and once with pixel shift. Then print both images at increasing sizes (or crop to different levels to keep the costs down ;-) ), and hang both sets on the wall from "small" to "big", randomizing the pixel shift vs non-pixel shift position. Note for yourself first in which print you see a clear difference. Then, invite ten office workers to (individually) evaluate from a viewing distance of at least 1 meter the prints from small to big. Record at what point they can distinguish "pixel shift" from "non-pixel shift", and let us know what the equivalent print sizes are.
Of course we want the best possible quality as a starting point, but often I get the impression that due to too much pixel-peeping, we have lost a sense of perspective on real-world advantages regarding the minute differences we see in studio tests or comparisons like this one.
Motion Compensation can actually do this sort of subject when it's being used properly:https://www.facebook.com/photo.php?fbid=10154298669699617&set=o.130482700345756&type=3&theater
But the DPR FUD is making people doubt even that. How lame.
gravelhopper: Let us sum this up:
- Pixel shift motion correction applied to compensate for fast moving subject linear movement.
- Long exposure times (1/4 sec) applied to fast moving subject.
- Applying software with limited capabilities while ignoring Pentax provided software.
What is the purpose of this test? To see if Pentax pixel shift motion correction can compensate for a completely flawed photographic technique applied to a subject that is explicitly out of scope of the tested technology?
This test / article is a complete break-down. Fortunately, this doesn't happen too often with DPR.
The worst bit is there are links in this article that show Pixel Shift /can/ cope with this sort of subject if you know what you're doing.
If this was a hit piece then they shot themselves in the foot. Methinks it's just plain poor journalism, though, which is no better.
Timbukto: These are true real-world field results of pixel shift of someone using gear they purchased in pursuit of their art - http://www.dpreview.com/forums/post/57854722
Coming up with headlines, clickbait controversy, shooting scenes that couldn't possibly be more ill-suited for PS, and reviewing an entirely different product (i.e. ACR) on features it does not have while faulting Pentax and the technique in general is just bizarre.
That post deserves to be appended to the article. Thanks for sharing it here.
bdbender4: In 10 years, or maybe only 5, we will look back on the days when cameras did such "poor post-processing" that we insisted on doing it ourselves, using computers and phones, taking huge amounts of time using complex software that was expensive and not very well written. Back in the days when there were "JPEG" photos, seen as inferior since they had "modified the data from the sensor", and there were "RAW" photos, where we modified said data ourselves. Never mind that there were several types of RAW data, all of which were highly artificial constructs that modified the data from the sensor. ;)
Many of us are pretty familiar with what RAW actually means. The thing is, when creating a digital image you're always taking a large amount of data and removing chunks of it to make the final viewable result, which will always involve compromise. This is fine, but an automated compromise (aka in-camera processing) will never know what your intent as a photographer was - it will always be a best guess. It will also always be time constrained - done quickly rather than for maximum processing quality. This latter point could get better over time but determining the user's intent is not a problem to be solved.
So respectfully, I disagree. For those of us who know what we're up to, we'll always want to take as much data as possible to produce the final image that we wanted.
MrTaikitso: It's funny, because after 3 or so months of using the excellent Panasonic GX8, I hardly ever popped the screen out (except for video 'selfies'), using the VF most of the time. Unless it was plain obvious an image was tripe (missed shot etc), I did all my reviewing back at base on a superb Dell 27" 4K monitor, where I can pixel peep properly. Even with the GX8's superb VF, unless you're staring an image in the face at 'full size', it is never the same. Leica may well be onto something here for still photogs.
Surely it's still nice to have, though? I mostly use my camera's rear screen for two things:
1) Periodically checking I've set my exposure correctly2) High/low angle photos
No screen would leave me SOOL on both of those, whereas a fold-in screen gives the best of both worlds.
Wye Photography: I don't use the screen to "chimp" on my Leica M8, I forced myself to pre-visualise instead, to see the final result in my mind. So, I would love this camera. Amazing. A brave move. I would love to afford it. I would love to afford a Ferrari as well.
I just wonder if on car forums Ferrari is slagged off or on watch forums Rolex so slagged.
What is the root of all this hate? Jealousy perhaps, or dissatisfaction with life, or job or wife or perhaps feelings of inadequacy???
My Leica M2 and M8 were not worth the money I paid for them, neither is this camera. Neither is a Ferrari for that matter. That's life. Ain't it a bitch!
Well, It's been entertaining reading the comments, keep it up chaps.
I don't get why people resort to the "jealousy" narrative. It's quite simple. Leica charge way more for their products than they're worth and justify it with nonsensical press briefs - hence the sneering.
It's not changing the world, but it's nice to chuckle of an afternoon.
More power to you if you can afford these baubles, but that's what they are - baubles.
Spunjji: I only wonder what happens when Apple release a phone with a different shape a year from now? It's not tremendously expensive, but it seems like this device will have a shelf life not-much longer than a 2 year phone contract and that seems, to me, to be a sad waste.
Good points there, although on the final one, I would argue that the D5 has a shelf life of much more than 2 years! It will be superseded, but not by something dramatically better for another 4-5 years.
I'm definitely glad for the transition from 2-year contracts too. They're a dismal way to buy phones and encourage rampant waste. The thing is, though, phone companies are still geared up to sell their products in that fashion - they haven't had any incentive to make the devices last very long.
Points above accepted, I still feel weird about products like this... I guess it just doesn't strike my imagination. The way I see it, if you're carrying the grip, you might as well carry the camera. I guess it can apply to people semi-affluent enough to afford a $600 iPhone but without the cash to spare for a decent Point and Shoot camera as well, but that sounds like a small market. I would be more inclined to bet that most of the grips end up doing not-much.
bernardly: A really inexpensive way to experiment with swirly bokeh is the Russian Helios 44-2 or variant legacy manual focus lens. They were made in vast quantities and a copy can be found on eBay for very little dollars. Best used on FF camera.
I'm afraid I don't have any specific tips - in my case I would simply shoot wide-open. These are the best examples I could find without going on a long dig:
In those instances it's admittedly not as dramatic as the lensbaby, but the price difference ought to be borne in mind. This is also on a Micro Four Thirds camera so you miss out the edges of the frame where the effect is strongest.
I only wonder what happens when Apple release a phone with a different shape a year from now? It's not tremendously expensive, but it seems like this device will have a shelf life not-much longer than a 2 year phone contract and that seems, to me, to be a sad waste.