Pentax K3 Gets Its First Test Shots

I see the Pentax K3 as yet another piece of excellent innovation that neither Canon nor Nikon did. Because they don't do anything anymore except sit there and vacuum up money.

Regardless! The K3 uses its "ultrasonic" sensor cleaning system, which really does nothing more than vibrate the sensor very quickly, to shake and thus blur the image slightly. This gives it the effect of a low-pass filter.

An increasing number of cameras recently have been ditching the low-pass filter for two reasons. It absorbs a small amount of light, meaning that cameras without the filters do slightly better in high-ISO situations. It also adds a small degree of sharpness. There are many photogs who see this as "critical" sharpness, or that last bit of a good photo that gives the image an edge or "bite."

Generally, I would make fun of such ridiculous and idiosyncratic language, but it does actually make sense. That last little bit of sharpness in an image, when multiplied over an entire photo, can actually make a noticeable difference, even to an untrained eye.

So, how does the new K3, which gives you the on-the-fly option, hold up? Imaging Resource has posted its images, and we can actually draw a few conclusions.

First, the JPEG ISO performance is exactly where every other competitive camera is in the segment, making me further think that, at least currently, the ISO wars are over. In case you missed it, Fuji won. The X-Series of cameras is so far ahead of the pack as to really be... huh?

I captured Imaging Resource's images and did a huge blow-up of their corner resolution, where the effects are going to be smallest. For me, this is the most interesting part.

Make sure to view the full image.

It's impressive that even here, there is a difference. There is a small difference, to be sure, but one of the most eye-catching elements of the comparison is the slight increase in edge contrast along the black bars, as opposed to merely an increase in resolution. Right there, right in those black bars, is the "pop" that people talk about in images taken without the AA filter. That increased overall contrast is noticeable. As is easily visible even in the low-res preview above, the corners, but not in the center of the image, also suffer an increase in chromatic aberrations.

That is a very interesting aspect to an AA filter that I had never thought of. It acts as a small band-aid over slight variations in the alignment of wavelengths of light within the lens, thus hiding what chromatic aberrations would be there. It hides small defects in the lens design.


I recommend going over and viewing the full series of images at IR. There is a similar difference in the center of the image, where resolution is highest, and well worth a look.

Is Pentax's new K3 A Game Changer?

Hats off to Pentax. They have done something different. In case you haven't already heard, Pentax has introduced a new optional "low-pass filter." I put that in quotes because it's not technically a filter.

I don't mean to take away from the, um, coolness, of what they have pulled off, though. It's impressive and something I would have expected a bit sooner. Basically, the K3 shakes the sensor very quickly on a sub-pixel level to add a tiny amount of blur to the image, effectively providing a low-pass filter effect. This can be turned off whenever maximum detail is desired.

It's probably a rejiggering of an extant technology. Most IL cameras have a sensor that sports some sort of "sonic" dust removal system to keep the sensor clean. By sonic, they mean that the sensor shakes the dust off. So to me, this is an implementation of technology that's been long in the making. It's still good, don't get me wrong, but not a huge shift... pun unintended.

I have also been waiting for a camera company to use the sensor to take multiple, quick exposures while moving the sensor to set locations to increase resolution. I could have sworn one of the MF camera companies was doing this already, but now I can't find it. Regardless, we are already doing it for HDR photography where the camera quickly takes three shots in succession. One sub-pixel shift to the left or right would theoretically double resolution.

But forget all that. Thank you Pentax for giving us something new. We now have Sony, Pentax, Fuji, Olympus, Panasonic, and especially Blackmagic all giving us new, or at least new-ish, stuff. Now, who's missing... who's missing... oh right! Canon and Nikon! Who are off doing fuck all.

Is APS-H the future?

I made an off-handed mention about Sony, and by connection the entire industry, ditching APS-C except for their entry-level cameras.

The more that I thought about that idea, the better it became. Canon makes two, TWO, lenses worthy of an enthusiast's attention for their APS-C cameras. At most, they make three. Those are the 60mm F2.8 macro, which is excellent, and the 17-55mm F2.8. The 10-22mm F3.5 is pretty good, so I'll play my own Devil's advocate and include it.

Three lenses. They make over sixty. It's obvious that APS-C is not something that they care about. It's obviously something that NONE of the companies care about. The only company out there who's wholly behind APS-C is Pentax, which is, regardless of the quality of the new K5, a bit player at best.

One of the major advantages to APS-C is the decreased size of the optics and the increased depth of field at any given aperture. If I am forced to use full-frame lenses, all advantages are eliminated. The only remaining advantage is increased pixel-density, which is only kinda'-sorta' an advantage.

There will be an increase in price, since sensors of larger size will obviously have smaller yields and thus higher prices. But Canon and Nikon alike both makes APS-C cameras that are just as cheap as 4/3 cameras, so with the advantages of scale, the price increase might be negligible.

Enthusiasts would jump all over this. No one has a lens set that they want to keep for APS-C, so by shifting formats, the companies aren't doing any harm to customers. Instead, the well-heeled customers are given a big boost in image quality and access to lenses that their format can actually use to their potential.

This potential should also be of great interest to the marketers and bean-counters at the various corporations. Sales of SLR cameras, after climbing by huge percentages for the past ten years, are flatlining and starting to decrease. The market is becoming saturated and the best that can be hoped for is selling upgrades to those who already have cameras. Everyone who wants a camera has one.

By introducing a new format to a wide audience, the companies net three benefits. One, they can sell new cameras as upgrades, generate new interest, and possibly lure new enthusiasts into the market. Two, they also lure more people away from cheaper APS-C lenses into their respective systems of full-format lenses, which will then lead to increased sales of their full-format cameras. And three, they finally have some real differentiation between cameras, as opposed to engineered differences where one camera is inferior to another simply because it needs to be, not for an financial reason.

Think about how great that would be. APS-C cameras cost up to $1,000. APS-H goes from there up to around $2,000. And from there on up, it's full-format. It would be a beautiful thing.

APS-C will always have a place for those who want cameras that cost well under $1,000, but for those who want more, give them something worth wanting. Sony was able to do the A850 for under $2,000. Everyone else can sure as hell manage a 1.3x sensor for a lower price.

Pentax Has Gone Insane

I, like everyone else, just got to see the official shots of Pentax's upcoming super-teeny-weeny Q-Mount camera.

As with Panasonic and Olympus, Pentax seems to think that the money is to be made in trying to net people who are upgrading from point-&-shoot cameras. I have discussed why this is wrong. But, that shouldn't be taken to claim that there is no money to be made in that market. I'm saying that there isn't a lot, and that the market is very fickle.

If you go to Flickr's Camera Finder page, you can see which cameras are the most popular. The iPhone is now the most popular camera on Flickr. Followed by four rather expensive D-SLR cameras, including a $2,500 full-format camera. The top P&S camera doesn't even crack the overall top-30! Does this sound like a market that wants crappy pictures, even in a small body?

But, you say, Flickr is the domain of photographers and enthusiasts, and of course they are primarily concerned with image quality. That's why even the "most popular P&S cameras" chart has one of the largest cameras in the category, the Canon G11. Even the iPhone is explained by photographers liking Apple products a lot.

All true.

So if enthusiasts and photographers won't be swayed, then those who buy P&S cameras must be the ones that want these things, and the Micro 4/3's cameras that have so far failed to connect with that market must be missing that je ne sais quoi that the P&S buyers want, that when introduced, will turn the market into a massive, money-generating monstrosity.

Totally false.

If that assertion were true, then the GF2, which was significantly smaller and less pro-oriented than the GF1, should have sold better, or at least as well. It did not. It sold very poorly and is positively invisible on Flickr. Panasonic's most popular camera is still the GF1, which is now two years old.

The only reason why cameras as cameras do well is because of their association with serious photography. Cameras as toys do well for other reasons. Making a super-tiny camera won't get a company anywhere. They need to appeal to all of the other things that people in that market want. They want colors and style. They want flashy advertising and the ability to slide it in anywhere, be it purse or pocket. This will NEVER HAPPEN with an interchangeable lens camera (ILC) system. It's just mechanically impossible.

But still, there are many out there who want in on the ILC game, but can't afford the crazy-expensive lenses. Even Micro 4/3's lenses can exceed $1,000 for a single lens. That is most definitely the enthusiast domain. And that's a viable point. I think a toy-like ILC might be tenable. So maybe that's what Pentax has in mind. They don't want either market. They want a new market of people who are curious about playing around with cameras.

Yeah! That's a great idea! Lenses that don't cost much more than $100 each, and a camera that costs maybe $300 would be... wait. What? The camera is going to cost $800? Are you serious? I'm Googling this...

Yes! It DOES cost $800! Is Pentax fucking insane!

Yes. They are. It's the only way to explain this.

Nikon D5100 DxO Mark Review

DxO Mark has just reviewed the sensor in the new Nikon D5100. It's the same 16MP sensor from Sony that's used in the Pentax K-5, a camera that I enjoyed enormously, and the D7000. Aside from some goodies like a large buffer, high-speed shooting, and some more pro-oriented controls, the picture quality of the D5100 looks like it would be identical to the D7000. Which is to say, excellent. DPReview did a test between the D7000 and Canon 60D which illustrated the low noise floor of the new sensor. Spoiler: The Nikon D7000 hands the EOS 60D its ass on a platter. These new sensors are on a different plane than the competition.

It makes me all the angrier that Olympus had the gall to release the E-5 with a three-year-old sensor and try to sell it for full price. If the Olympus E-3 hadn't been such crap, no one would even be interested in the E-5. The only thing Olympus had going for it, and truly likely the only reason that they sold any cameras at all, is its lens selection. Olympus' semi-pro and pro lenses are among the absolute best on the market. And focal-length to focal-length, the Oly lenses are half the weight (or less) of larger-sensored cameras.

I'm not changing my opinion that your best bet would be 4/3's and then a full-frame camera if you want the best quality, but with cameras like the Pentax K-5 being so small, that's primarily because of the lenses. I once wrote that Canon's creation of the 7D might indicate a newfound focus on the enthusiast cameras, and that a collection of new, high-quality EF-S lenses was on the way. That was wishful thinking. It's obvious that Canon would rather try and protect their cash cow than actually blaze a trail into any new markets.

So we are stuck with excellent APS-C cameras that have crap for lenses. One of hte advantages of a smaller sensor is, with lenses DESIGNED FOR THE SENSOR, you get a broader depth of field for any particular aperture. That allows more things to be in focus. But you only get that benefit with lenses designed for the 1.6x crop factor. Simply slapping an ordinary EF lens onto the back of the 7D nets you no benefits except for slightly better vignetting, and in return provides lower resolution. Yay.

Sony has their NEX Cameras, which might finally result in good, APS-C glass. As might the rumored Fuji mirrorless system camera coming next year. Until then, m4/3's is the only system that's pushing the boundary of general photography, and for that, it gets my money.

New Pentax and Nikon Mirrorless Cameras

The rumormill turns ever-onward. Apparently, both Nikon and Pentax's upcoming mirrorless camera systems will use very small sensors. Point-and-shoot small. If this is true, it's an interesting perspective on the market, because there is only one way to read this. Basically, Nikon and Pentax are saying that the real money is to be made in super-cheap interchangeable lenses. Olympus, in its marketing incompetence, seems to be bumbling in that direction, seeing as all of the m4/3's lenses that they have available are all aimed at the mid-to-low end.

Perhaps of greatest importance is that it seems to indicate a desire to avoid cannibalizing sales of their own SLR cameras. I've talked about how we should avoid buying Canon and Nikon because they don't want to take risks. Well, this is an example of them not taking a risk. They have their established SLR cameras, with which they command 40% of the market, and instead of competing with themselves to better future-proof their market share, they're content with letting Sony, Panasonic, Olympus, and Samsung try to dethrone them.

Still, even though I think it's an idea born of a desire to not compete with themselves, as opposed to blazing a new trail, this holds some interesting promise. First, we must change our perspective when looking at cameras. I come from an enthusiast/pro perspective, as such, seeing lenses for m4/3's that cost $300-$500, I think "what a deal!" But if I put myself in the shoes of someone coming from any other perspective, paying as much for a single lens as it costs to buy an entire, high-quality, point-and-shoot camera is ridiculous! In this sense, Penkon stands a better chance of capturing sales from the club-shooting, Facebook crowd.

But even then, I find the whole idea of a super-small interchangeable lens system no more than a novelty. I think that Olympus and Panasonic were correct to target enthusiasts and not the club crowd, because they will always be interested in size and style over quality. No matter how small they make the lenses, a compact camera will be smaller. So who will buy this system? Buyers of current super-zooms? That seems like a lot of work to capture a very small part of the market. And if they think that there are super-zoom owners with enthusiast aspirations, why not just make better super-zooms? And doesn't Nikon already sort of make a camera for them in the P7000?

Fundamentally, Olympus and Panasonic created cameras that act as replacements for SLR cameras, Penkon want to create cameras that act as replacements for point-and-shoots. With small sensors, they can create lenses that are super-super-compact, and sell them for $100. While I would never consider buying one, and don't think that their target demographic would either, it will be interesting to see to what degree the market bites. It's quite a gamble.

Nikon D7000 Dukes It Out With Pentax K-5

Holy crap! The Nikon D7000 has been ranked near the tippy-top of APS-C sized SLR cameras! With an overall score of 80, that puts it with or above all of Sony and Canon's full frame cameras. Excellent work, Nikon!

The D7000 is fighting with Pentax's K-5, which even more shockingly mustered an 82, which matches the Nikon D3s. If these numbers are to be believed, sensor development in the APS-C arena has kicked into turbo. I'll wait for further information, but this is making me reconsider my previous ideas about Micro Four-Thirds.

I still say that m4/3's is the best system to buy into for someone wanting a family shooter. It's compact, fantastic lenses cost very little and weigh even less, and Panasonic has proven a complete dedication to the format. You can buy a m4/3's camera and lens for less than $800, buy two more lenses for $500 to $1000 a piece, and you'll have a complete kit that fits into a small camera bag and will do for any situation an average person could imagine. Perfect.

But I used to also argue for the four-thirds format over APS-C because the increased sensor size didn't seem to net much benefit. Going up to full-frame resulted in a significant difference, but notsomuch APS-C. The smaller sensor's 2X crop factor meant that zoom lenses positively sung, with greater length and deeper depth of field. Colors were somewhat better on the larger sensor, as was dynamic range, but I felt that if that was a serious concern, you should save up for a full-frame camera. APS-C just didn't provide enough of a quality boost to warrant the increased size and cost.

These results change that perspective. The best 4/3's sensor on the market is the Panasonic GH1/2, which has a best score of 64 on DxO Mark. That plopped it smack in the middle of most modern APS-C cameras, truly, trailing the EOS 7D supercamera by only two points. But trailing the leading APS-C camera by 18 points cannot be ignored.

As I said, I'll wait to pass judgment, but if these early results hold up, anyone with enthusiast or semi-pro aspirations cannot consider the 4/3's format any more. APS-C has just walked away.

Tests and reviews for the camera Nikon D7000 (DxO Mark)

Pentax 645D Information

Pentax is releasing what will be, as far as I know, far and away the cheapest medium format digital camera on the market. It's a complete camera, meaning no separate back and body assembly, so pop on a lens and you're good to go. At $10,000, it's half the price of the next cheapest medium format cameras from Phase One and Mamiya, which ring in at $20k. While this prevents you from swapping out backs, when the whole camera only costs this much, who cares?

I'm surprised to see SD cards being used for storage. When photo files can easily hit 100MB, where's the CompactFlash?! Other than that, all of the limitations and advantages common to medium format apply. This is very exciting for someone looking to either expand their studio or an enthusiast who wants to take their art to the next level. The price is still quite high, and buying this over a Nikon D3x or Canon EOS 1Ds/5D is something that's going to have to be mulled over. But if you like portrait photography and/or, as Pentax discusses in the interview, landscapes, this camera is just drool-inducing.

My skills are still nowhere near maxing my EOS 5D, but if yours are, what the hell. Go for it.

DPReview posts sample images!: Link

DPReview, the bestest camera site on the net, has posted sample pictures of the new 645 and even in their studio setting, the detail capture by it in comparison to every other camera they have is amazing. In a studio, it's pixel-peeping. The big differences will be revealed in complex nature and landscape shots, but even here, the difference is significant. For example, look at the small globe.

Obviously, noise levels are still far behind the DSLR's from Cankontaxony, but at only ten grand, and the Sony Alpha A850 now below $2,000, the new flagships from the likes of Canon and Nikon will have to stump up some serious performance to justify their likely-to-be over $7,000 prices.

Early Preview/Review with sample photos

Japanese site Digital Camera Watch has posted a review of the 645D along with multiple full-resolution sample photos. Dynamic range is .5 stops lower than most other medium format cameras at 11.5, but the images are still great. Without a direct comparison, I see no difference between these sample photos and samples available from Mamiya, Phase One, and Hasselblad.

From fō-tō-gră-fē

As with many (all?) medium format cameras, the Pentax has no low-pass filter. On such large, high-res sensors this rarely results in moire, and even the complex city scene I was only able to spot the smallest amounts. I did find one of their photos with a noticeable degree, and that was on some rough wood. I've posted the crop below.

From fō-tō-gră-fē

Interview and Photos

Luminous Landscape has an interview with a Pentax rep and a few photos. As with all Luminous Landscape posts, he doesn't provide anything at full resolution, thus rendering all of his photos useless. I don't get this. It's a behavior common along many photo websites. Fuck you. Give me the files.

DxO Mark Releases Test Results

DxO Mark, whose test methodology befuddles me, has released sensor information on the 645D. It's not at the top of its class, but it it definitely earns magna. It's beaten out by two Phase One backs and the Nikon D3X, but that's it! The Nikon is half the resolution but brings many other benefits to the table, it also costs a few grand less, and the two Phase One backs cost well over twice as much as the Pentax. Go Pentax!

Ken Rockwell Hates the 645D

As any good critic, it's not their job to explain why something is good or bad, but to say whether they liked something or not and to effectively explain why. Ken Rockwell does an excellent job with this.

Basically, he's arguing that Pentax is retarded for making a medium format camera aimed at the hobbyists who buy point-'n-shoots. Pentax has loaded the camera up with useless features that, if anything, get in the way of the pro shooter. He's also upset that this is not a true medium format camera. The sensor is too small. This criticism is truly damning since the value proposition of the Pentax is predicated on being a direct competitor to more expensive medium format cameras while being only slightly more expensive than Canon and Nikon's top-pro cameras.

I appreciate all of his criticisms, and understand them quite well, I'm still leaning towards liking the camera. Its sensor is larger than Canon and Nikon, has 40MP regardless of sensor size, good dynamic range, and is an excellent introduction to the 645 world. And for only $10k, I'd learn to work around its faults.

Luminous Landscape Full Review:

They like it! They really like it! If anyone was going to kvetch about the 654D it was the Luminous Landscape guys, and they like it enough to call it a "defining camera." Obviously, the price is amazing, but also the ergonomics of the camera were called "stellar."

Luminous Landscape Comparison:

After their review, they've posted a comparison of the 645D and the PhaseOne P40+, they also threw in a Canon EOS 1Ds Mark III and a Leica M9 for good measure. The Canon Performs very poorly, but I think that is a combination of both the lens and the camera. They suspect something might have been wrong. The Pentax performs almost identically to the P40+, even outperforming it at times, but is beaten in the dynamic range competition, but not by much. This means that Pentax was actually telling the complete truth when they said that their dynamic range would slightly trail its competitors.

They're very impressed by the Leica M9's resolution, which was undoubtedly helped by its total lack of a low-pass filter, but admit that for larger prints, you want to move up to the medium format camera. This was an epic comparison and a frank analysis that I'm not used to seeing on photography websites.