Pentax K20D Report—Part II: The CMOS Sensor

By Carl Weese

Link to Amazon U.S. K20D page

Link to Amazon U.K. K20D page

We're about to look at first impressions of the new sensor in the K20D, but first, the result of some serendipitous testing. Monday morning I decided to repeat two tests (white clipping point and ISO/noise) from last week, this time using both K10D and K20D cameras one right after the other so the files can be compared side-by-side. For ISO I set the tripod about seven feet from a cluttered bookcase. Lots of detail, many bright whites and dark shadows to show off noise. Without thinking about it, I used my 21mm DA Limited, letting the cameras handle focus while I fiddled with the ISO settings. I've had the K20D less than a week and already forgot about "the focus issue." The 21mm has been diagnosed by the K20D as needing a +6 custom AF focus point adjustment. The files from the K20D were in perfect focus, but the set from the K10D was off by a country mile. So bad that I couldn't use them to evaluate noise at 100% view.

This presented an opportunity. The K20D has given my 40mm DA Limited a clean bill of health: perfect AF calibration. So I repeated the series using the longer lens. Both cameras nailed the focus. A lovely, if accidental, confirmation of the accuracy and usefulness of the AF calibration feature. I'd also been wondering to what extent a calibration derived from tests of a small target at just a couple feet would hold at longer distances. This showed it holding perfectly at seven feet, and real-world pictures I've been making the past few days show it tracking fine into the middle and far distance. Here's what a simple focus test target looks like, as seen by the camera during the test:

Angle of approach should be 45 degrees. Make the AF focus on the dark line in center, then, at high magnification, see if the sharpest plane lies in front or back. (This is Tim Jackson's focus test chart.)

The sensor
Now on to that new CMOS sensor. The pixel horsepower race being what it is, of course the first point is the pixel count. At 14.6 MP it's not only a 46% increase over the K10D, it's also out front of the price-class competition, at least for the moment. So my first impression is, these files are big! Each RAW capture takes up 23.2 megs of storage space, so an 8 gig SD card fills up with 334 captures. 165 shots fill up an archive DVD. A 16-bit file opened from RAW into Photoshop weighs in at 83 megs.

The main reason to seek more pixels is to make bigger prints with less up-interpolation of the data. At 360 ppi (my standard since it's native for the Epson pro-stylus printer I use) the print has grown from the K10D's 7.2 x 10.75 inches non-interpolated or native file to 8.62 x 12.98 inches. Probably more important, the amount of up-interpolation needed to make a print 22 inches wide has dropped from 204.5% down to 169.5%. I have not made prints yet, because I will first have to test to find out what combination of interpolation and sharpening works best for these files. The effect of the higher pixel count is plain to see on screen, though. I have not worked with a CMOS sensor before and will have to feel my way into getting the character of this thing, but my first impressions are favorable.

From Town Plot, Waterbury, Connecticut

Sunday till mid-day there was luminous light under slowly clearing clouds. This capture, made on auto exposure with no compensation, holds the full range of values with beautiful tonal transitions right on the defaults in ACR.

In particular, from my first pictures of real subjects, I get a sense of smoothness, of more seamless tonal transitions throughout the range, while the general feel of the color is similar to the K10D. At the same time, detail looks very "optical" to me. That is, edges and textures look the way I expect a lens to render them, not as though they've been artificially generated by a computer. The terms "plasticky" and "film-like" get tossed around a bit—so far my impression is the K20D results are in the latter camp.

The main concerns when engineers pack almost half again as many pixels onto the same size sensor are a potential loss of dynamic range and possible increase in noise. For dynamic range, I'm most concerned to know how far up into the highlights I can go before losing data to clipping. To test this I place a smooth piece of card stock in perfectly even light and aim the camera on tripod at it. I make a bracket around neutral gray, based on the camera's spot meter (good time to check whether the other meter modes agree reasonably). Then starting at 2 stops over-exposure (Zone VII to classicists) I move the exposure up 1/3 of a stop at a time. Download and open the set of exposures as a group in ACR and check densities with eyedropper and histogram. The card exposed as middle gray came in at level 128 for the K10D, 119 for the K20D, a difference of something like 1/8th or 1/6th of a stop. Both cameras held full detail, no clipping, with space between the sharp peak of narrow histogram and the right wall, up to 2.3 stops above middle. Each showed a touch of clipping at 2.7: the K10D file came back with a Recovery setting of 4. The K20D needed a Recovery of just 2. At three stops up, the K10D recovered with a setting of 10, the K20D with 6. At 3.3 stops up, the K10D couldn't pull back from clipping with any amount of recovery, while the K20D came clear at 16—not an extreme amount of Recovery. At +3.7 it blew through the right wall of the histogram.

So, even with nearly 50% more pixels, highlight retention is slightly improved. In addition, my real world shots are indicating that this test may underestimate the improvement. In the test, not only does the new camera reach a little higher, but it holds more data just before the cutoff. The older camera sort of slides into the clipping above +2.3 stops. This conforms to my impression that the highlights in real pictures are richer, improved more than the white card test indicates.

Noise
Now for the noise tests, second go-round. The K10D's noise performance up to ISO 400 is good enough that in walkabout shooting I make a practice of using the camera's auto ISO feature, limited to a range of 100-400. I've never had a picture ruined by excessive noise because the light demanded 400. Up to 400, I also don't see a real difference between the K10D and K20D—except that a 100% view of the new camera's files is more magnified on screen because of the higher pixel count. At 800, the judgment gets a little tricky, which calls for a digression.

The K20D has a new menu item that lets the user select one of four levels of high ISO noise reduction: Off, Weakest, Weak, and Strong. I shot a series at ISO 100 through 3200 using RAW+best quality JPEG. As near as I can tell by pixel peeping at 100%, the NR is entirely done by the software that makes the in-camera JPEG files, not hard-wired into the sensor. The differences are plain to see in the JPEGs, but I can see no change at all to the RAW files viewed in ACR—they all look like the JPEGs done at "Off." This is fine with me, because to me the JPEGs with NR look worse, not better. Any reduction in noise is more than offset by smearing of detail, and of the noise itself.

Both at 800 and 1600, the noise in the K20D RAW files viewed at 100% in ACR looks better to me than the older camera, though at 800 I would not say there is exactly "less" of it. When the two are compared, the K20D noise looks crisp and clean, while the K10D files at high ISO look, in comparison, as though they've had some smeary noise reduction done to them, even viewed in ACR. In the old days, there were films that had smeary, mushy grain and films (famously 35mm Tri-X and T-Max P3200) that had beautiful, tight, clean-edged grain. Good grain that people came to love—imagine Salgado's pictures without that grain! At least on screen, the K20D looks like it may have "good noise" at high ISO. The new camera's 800 files look better, and the 1600 files look a whole lot better, than the older camera. 3200 looks really coarse. These are first impressions that await actual print tests for confirmation, but I'm looking forward to printing some good pictures made in rotten light at 1600. A little bit of nice tight grit may be just the thing for "available darkness" shooting.

Liquor Store, Union City, Connecticut

Another shot from Sunday with a long range of tones held beautifully with just a little adjustment in ACR.

Next time, more on color and tone, and more on new features.

_______________________

Previously:
Pentax K20D Report Part I: Focus

_______________________

Carl

Carl's website 
Working Pictures
WPII

Comments

[Gary Sutto]

How do the dng files look next to the Pentax raw files?
I'm in favor of NOT having to upgrade my software program(s) to work with a new camera. That's one of the reasons I'm interested in the new Pentax.

[Dennis]

I've read a few of your posts, Carl, but never clicked through to your site until I saw the Waterbury, CT picture today. And now I just spent half an hour going (too quickly) through your blog looking for Torrington pictures. I spent the first almost-30 years of my life in Torrington and moved to Sharon 12 years ago. It's fascinating to see interesting pictures made from familiar scenes and even more fascinating to see how much is unfamiliar from the place where I grew up. Now your blog will have to be on my list of daily visits.

[Andreas]

100% ack. I have the K20D now (together with the 31 Ltd., hehe ;-) and found exactly the same things on the K20D. Better looking noise, usable 3200 I would say and fine tonal transitions.

Carl, it would be great if you could repeat what I tried: I shoot PEFs at high ISO (at least 1600) and load them with Pentax Image Browser, then convert them to DNG from there, and import them in Lightroom to continue my usual workflow. I find that there is a huge difference in the look of the noise compared to when I open the PEFs directly in ACR.
Would be interesting if you could confirm this.

Another thing I am intersted in is what you think about this D-range 200% thing. I have no conclusion yet, but I am not scientific enough either...

However, thanks for your efforts!

Andreas

[Carl Weese]

Gary, the .dng files can be read by early versions of ACR (I've checked with my aging iBook G4 running PSCS2) but to fully support the new camera you need the latest (4.4.1) ACR and therefore PSCS3. I'll deal with this in more detail next posting.

Dennis, something about Torrington makes it the most productive nearby territory for me to go hunting for pictures.

[Shadzee]

Ohhh come on Carl... I don't want to upgrade from my K10 yet ;-)

You're making me regret reading these posts ;-)

Now... I love the way you explain and put in words these technical highlight clipping terms. I've compared my K10's raw files with Canon 30D, and found a smother transition in tones near the highlight clipping points. I guess these nice transitions in lighter range of tones are one of the reasons I find the K10 images more pleasing to my eyes.

[Robert Roaldi]

Every few months MJ writes a piece, or in this case published a piece by someone else, and it causes me to regret all over again my decision to sell my Pentax lenses (screw and K) about 5 years ago, because Pentax was not yet on the digital train. I had primes from 24 mm to 400 mm, flashes, bellows, 2x converters, you name it, all bought 2nd-hand for not really that much.

Mike, please stop.

[Adam Zolkover]

Good stuff. I've still only had a chance to take a few pictures with the K20D -- mine had a memory-writing problem, and I had to have it exchanged -- but my impressions match up with yours. Not having done any testing, in fact, I might be even more enthusiastic about high ISO performance. I was never happy with the K10D at ISO400, while the K20D produces almost no noise at all.

Also, I'm wondering about your impressions of black-and-white conversion with the K20D. I've used Lightroom to convert a couple of files, and am very pleased with what I've seen -- the smoothness and extent of tonal range seems better than with the K10D, and I feel like I'm getting more detail at the top and bottom of the tonal curve, even in high-contrast situations. This might be a function of using the extended dynamic range feature -- I haven't tried without it yet -- but I've found the difference very noticeable.

[Christopher Lane]

Dear Carl,

Thanks so much for this series. This is the kind of objective reviewing/reporting we need more of. It makes the decision-making process a whole lot easier.

Chris

[Carl Weese]

Andreas, are you saying that if you convert the PEFs to DNG with Adobe's DNG Converter utility and then take them into Lightroom (which uses ACR to handle RAW files)you get different/better noise results than reading the PEFs directly? Have you tried simply shooting DNG in-camera?

[MarcinB]

Very curious findings, totally the opposite of conclusions made at www.optyczne.pl (a professional Polish camera & lens review site). The reviewers tested 2 samples of K20D and found the sensor disappointing: poor dynamic range (7.5 EV at ISO100), visible noise at 1600 and 3200 ISO, lots of noise at long exposures (coupled with NR that cannot be switched off), lots of hotpixels in RAW that cannot be mapped (shift from image to image). They say the effective resolution of K20D equals that of K10D.
Other criticism from that site: unreliable autofocus, slow operation when reviewing/magnifying images (sometimes you even have to wait more than a dozen seconds to see them!). Obviously they gave an average grade to the K20D (4.3/6), recommending K10 instead.
So, who do you believe? :-)

[Shadzee]

Carl,

With my K10D, I shoot in PEF and use Lightroom to convert to DNG. With a few tests I did last year, shooting in PEF and converting to DNG in Pentax's software did not make any difference in outcome.

However, I've read reports that shooting in PEF AND then converting to TIFF you get better results! But, for me this is not an efficient work flow.

[Carl Weese]

Andreas, I'm interested in differences in RAW interpretation, so I looked for the effect you describe. I did some snaps at 1600 both as DNG and PEF. Then I ran copies to DNG with the converter. In ACR from Bridge I see no difference in noise. I don't use Lightroom but happen to have a trial version in date on my computer, so I looked at the DNG-from-PEF in LR and saw no difference from the view of all the files in ACR. The converter did reduce the file size drastically to 'just' 14 megs. But I don't see any improvement--or deterioration--from running a PEF through the converter.

[Gary Sutto]

Thank you Carl, What I should have said is that I'd like to use dng right off, as I assume that since my older programs read/open the dng format that I will be able to use the software I have without an upgrade. Am I missing something here? Is there a quality issue or difference using dng or the Pentax raw file? I'm looking forward to your next post on this.
Best.

[Carl Weese]

Gary,

Look up last year's post on the K10D called something like 'all dng is not equal'--older programs can access a dng from a new camera, but that's not the same thing as fully supporting it. WB readings are apt to be flakey, and handling of color in general may be inferior to a program that supports the new model.

MarcinB,

Some of these complaints sound like analysis of JPEG files, which I seldom use. The K20D has a plethora of JPEG options--which I'll talk about in the next post--that can muddy the waters when comparing to JPEGs from another camera. The problem of hot pixels in RAW has not been evident in the thousand or so exposures I've made so far. I have not yet done any long exposures so I will be on the lookout for that potential problem. The resolution difference leaps off the screen when viewing RAW files. This can be masked by comparing JPEG files created with different levels of in-camera sharpening, among other factors.

[taran]

I would also like to thank you Carl for this in depth and decidedly subjective review. Certainly a breath of fresh air from pixel peeping and other, more ponderous generic reviews that wallow in their own cold, quantifiable, statistical analysis.

I do have a favor though... could you comment on your daytime on-board fill flash experience with, say, portraiture subjects? I hear the 180th/second speed limitation of the onboard strobe cannot be easily employed during normal daylight conditions because its just not fast enough.

Down here in florida, although you wouldn't think it, we have to use on board flash a lot because of the extreme harshness of our light... its on of the things I enjoy most about my Nikon DSLR's.

[Ctein]

Dear Carl,

Regarding the total dynamic range of the camera, you gave us the range above middle grey (+3 stops and change) but not the range below middle grey. What's your evaluation of that?

For me, total range is more important the highlight clipping, as I can always dial in exposure compensation if I don't like how the range is split between highs and lows.

pax / Ctein

[Andreas]

Carl, I have not tried to shoot DNG directly because of the larger file-size. And I believe that PEF is native ... is better, but this is irrational.

However, the effect I saw might be subtle or even invisible in most files. I had one example where the PEF read with LR showed ugly magenta blotches in the dark areas, where the PEF-DNG from Pentax Browser was much cleaner and still retaining all detail. Unscientific as I am, I deleted the ugly version...

I was just lucky to have such a file at the moment that I tested Pentax Browser. The file is ISO 3200, f2, 1/45s, converted with Pentax Browser from PEF to DNG and then exported with LR as full jpg in srgb. As mentioned I applied the usual LR workflow, i.e. exposure, brightness, contrast, whitebalance... and standard nr of color 25.
The result is here: http://www.flamelingo.net/pics/_IMG8439.jpg

In the PEF version read directly from LR there were these ugly blotches of color which could not be removed, neither with LR nor with Noise Ninja.

What I want to say is that you need such a file which is prone to ugly chroma noise in order to see the difference.

best always
Andreas

[Carl Weese]

Andreas, I hadn't noted yet that the PEF files have some compression. Since ACR seems to support them fully, I may switch for the storage advantage. I haven't even unwrapped the Pentax software for the K20D. If the Pentax Browser allows conversion of PEF to DNG I have to put that right on the list of things to investigate.

Ctein, I'll do a quick test to see how low the exposure can go compared to the K10D, but I don't find that information--the absolute bottom limit before merger to black--terribly useful. When I have to underexpose the middle values to save the highlights, the important issue to me is how well those underexposed middle values can be lifted back where they belong, and that doesn't show well in simple gray card tests. I need real world pictures to get a feel for that, and I'm already forming an opinion based on the shooting I've done.

OK, using a smooth surface target, at the bottom the two cameras are absolutely identical, tracking levels to the point from -3 all the way down to -5.7 stops where the level is down to 4. However, around -2 stops the K20 reads a few points higher. That might mean picture information at that level will come up to mid-tone a little better.

Taran, I don't think I've even popped up the on-board flash, and likely never will. But your question can be answered by the numbers. If the sync speed is limited to 180th, even at ISO 100 in full sunlight the aperture will have to be closed down past f/11. That's flirting with diffraction, it makes selective focus impossible, and it means the little flash is going to have to throw at least f/5.6 worth of light to provide useful fill, which will limit the distance you can shoot from. Not promising.

[David Bostedo]

Andreas - Did you try doing both exports out of LR as uncompressed 16-bit TIFF or PSD files with a different color space (Adobe RGB or Prophoto RGB?) to avoid any problems caused by conversion to JPG?

Just curious as it might help track down if it's a real difference in the raw formats, or if the problem lies in LR's conversion of them to JPG, since it has to clip the color space and bit depth, and compresses the file.

[Andreas]

Carl, from your writing I assume that you don't use Lightroom? I hope that we still can compare our results...

However, luckily I found the original PEF of the picture I had linked in my previous reply on some memory card ;-)

Now, things go as follows:
I have one version, the original PEF. This is being imported in Lightroom (1.4 or so) and the usual adjustments are made in order to get "good" color and tones. Then the whole thing is exported to DNG, which reduces size to the usual 15 megs.

The other version is the same PEF converted to DNG with Photo Browser. Note that this version takes some unbelievable 30 megs, so I also import this in LR and export as DNG again. This compresses the size to the usual amount (around 15 megs).

Now we have 2 DNGs of the same source image, about the same size, both finally catalogued in LR. When these are viewed against each other, you will see a difference in color and noise characteristics. Honestly I am not able to match both colorwise.

I have put both dngs on my webspace for you to see what I mean. Note that these DNGs incorporate my settings, but can be reset to the orignial state in LR.

The upload section can be accessed here:
ftp://www.flamelingo.net
user: web328f2
pass: banane

I would be very interested in your opinion.

best always
Andreas

[Ian Rivlin]

My experiences have been very similar to yours, In my view, it's better in every way. - some aspects are subtle improvements and some are dramatic. I use - almost exclusively - the Pentax 18-250mm lens. I used to have a K10D with the Sigma 18-200mm lens. The optical performance of the Pentax lens is light years ahead of the Sigma. - People ask me why I don't use a prime lens.... It's because the difference in performance between the prime lenses I have and the Pentax lens is very small. The extreme convenience of 18-250mm makes almost all pictures immediately available. - Sure I have other lenses but most of the time, they stay in their cartons. If I got the K20D stolen or damaged, would I get another one? - Yes, definitely. It's a superb camera. Anyone who can't get a great photograph using the K20D should seriously re-assess their photographic skills. The camera would not be the inhibiting factor....

[Adnan EROL]

Do you have any test images comparing old lenses on K10D and K20D?

http://www.pentax.jp/english/imaging/digital/slr/k20d/feature.html

..."Including the top-of-the-line the DA Star series, all smc PENTAX interchangeable lenses exclusively developed for digital photography feature an advanced optical design that leads the light at a proper angle to each pixel, even at its peripheral areas. On the other hand, conventional lenses designed for film photography lack this advantage. The K20D’s new CMOS image sensor is designed to receive the incoming light at optimum efficiency and accuracy even with these film-format lenses."...