[nanok]

in other words (and perhaps this is the wrong thread)

the pentax offering today might be the first sign of something never before seen since digital has "replaced" film: it appears there's not much to shout about, after a few years "of research and development", the k-5 is what it should be, it's.. well, perfect. sure, small improvements here and there are possible, but in truth, the camera is so good, especially when it comes to image quality, ergonomics and build (which is basically all that matters (now i'm getting myself in trouble...) ), that really there's nothing much to work on, without stepping into a whole different class altogether (i mean, not even close price or feature-set/target wise).

perhaps we are getting ready to enter the new era: the era where one really needs to get out and shoot, as there's no longer any hint of a valid excuse: the camera is as good as it's going to get, time to stop watching for "the next one", and get busy shooting instead.

the internet forums shall go quiet, camera manufacturers will need to get imaginative to survive, to manage selling "more" (or any), difficult times are ahead, photography might become an actual artistic+technical hobby, instead of a consumerist pastime. the horror.

remember when cameras were good for decades, and all you had to worry about was go out and take pictures? that's what's coming

yeah, right.<gets off the wooden box>

[LeRolls]

Very nice review! If I do end up getting a Pentax DSLR sometime late next year it will most likely be the K30 which will hopefully drop to around $600 by then and I think I will be more than happy with that one.

[Class A]

Starting from ISO 6400, the K-5 IIs Siemens star images look very funny and as a matter of fact less sharp in the centre than the K-5 images (definitely at the highest ISO settings).

Seems like we are seeing a different JPG engine at work here, which also casts doubts, to an extent, on the other comparisons.

[Adam]

Starting from ISO 6400, the K-5 IIs Siemens star images look very funny and as a matter of fact less sharp in the centre than the K-5 images (definitely at the highest ISO settings). Seems like we are seeing a different JPG engine at work here, which also casts doubts, to an extent, on the other comparisons.

Perhaps moire is at play, assuming it influences the raw data that the sensor captures before the image is processed. This is worth investigating. May even have to shoot another chart at varying distances...

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com server and development costs are user-supported. You can help cover these costs by donating or purchasing one of our Pentax eBooks. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, KEH, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

[Class A]

The review (and other places of the forum) repeatedly claims that the K-5 IIs has better resolution.

How has this been measured?
What is the quantitative difference?
At which shooting parameters is this difference the highest and when is there no difference?

The Siemens stars in the "moiré" section nicely demonstrate that both cameras seem to resolve the same details (i.e., have the same resolution) and that the K-5 simply shows blur when the K-5 IIs is showing nonsense already. Just what you would expect from a well-engineered AA-filter.

It appears to me (and not only to me, according to some user requests) that what is heralded as "increased resolution" in many cases is just more micro-contrast and that this difference could easily be addressed by performing the required capture-sharpening on the K-5 images.

There will be situations where the K-5 IIs actually captures more detail, but you are not showing these. Instead, you are making incorrect comparisons by denying the K-5 images the capture-sharpening that comes as part and parcel with a Bayer AA-filter.

[Adam]

It appears to me (and not only to me, according to some user requests) that what is heralded as "increased resolution" in many cases is just more micro-contrast and that this difference could easily be addressed by performing the required capture-sharpening on the K-5 images. There will be situations where the K-5 IIs actually captures more detail, but you are not showing these. Instead, you are making incorrect comparisons by denying the K-5 images the capture-sharpening that comes as part and parcel with a Bayer AA-filter.

Review aside, all of your comments seem to be trying to prove that the K-5 IIs is no better than the K-5 when it comes to sharpness. I don't really get this. The K-5 Iis does have superior resolution, as can be evidenced through something as simple as a snapshot of two distant objects. Once can, however, say that removing the AA filter is a hack, because the increased sharpness isn't always going to be there (i.e. when diffraction is at play, etc.). If you are not convinced by the evidence that has been posted thus far, please stay tuned for the samples that we will be posting next Monday. The test charts can be misleading in a way, as you benefit more from the K-5 IIs's filterless design at long range than at short range.

[civiletti]

Review aside, all of your comments seem to be trying to prove that the K-5 IIs is no better than the K-5 when it comes to sharpness. I don't really get this. The K-5 Iis does have superior resolution, as can be evidenced through something as simple as a snapshot of two distant objects. Once can, however, say that removing the AA filter is a hack, because the increased sharpness isn't always going to be there (i.e. when diffraction is at play, etc.). If you are not convinced by the evidence that has been posted thus far, please stay tuned for the samples that we will be posting next Monday. The test charts can be misleading in a way, as you benefit more from the K-5 IIs's filterless design at long range than at short range.

As another commenter who questioned whether the K-5IIs has higher resolution and not just higher micro-contrast, let me clarify that I am not trying to prove that. I hope the AA filterless camera does resolve more detail, but I am not yet convinced that it does. Higher micro-contrast can make more detail visible under some viewing conditions, but it does not add more information to the image; which higher resolution does.

It's reasonable to wonder if sharpening and/or micro-contrast increasing in post production leaves the K-5IIs with an advantage.

[Class A]

Review aside, all of your comments seem to be trying to prove that the K-5 IIs is no better than the K-5 when it comes to sharpness.

I'm not trying to prove that. It would be a foolish attempt because the K-5 IIs can be superiour (as I've said before).

The reason why I'm posting "hold the horses" messages is because I

• don't like unfair comparisons, and
• don't want the K-5 IIs to cause disappointment. Even though moiré and other artefacts are rare, they can appear and cause frustration.

Not capture-sharpening images from a camera with an AA-filter is like not boosting the exposure for an image that has been captured with dynamic highlight protection. When the camera purposefully underexposes to protect highlights during capture, the post-processing must pull up the exposure to undo that effect.

Likewise, an AA-filter purposefully introduces blur to prevent detail from reaching the sensor during the capture that the sensor is not capable of recording. Accordingly, the post-processing must undo this blurring. Hence, the need for capture-sharpening.

The K-5 IIs images must not be capture-sharpened at the same level, because no prior blurring has occurred.

The K-5 Iis does have superior resolution, as can be evidenced through something as simple as a snapshot of two distant objects.

What you are seeing is

• better micro-contrast from the K-5 IIs because you don't capture-sharpen the K-5 images, and/or
• focusing issues with the K-5.

You cannot claim better resolution if you are only referring to sharper looking images.

If I sharpen a K-5 image, it will look much like a K-5 IIs image, but I won't have increased resolution, i.e., the ability to resolve fine detail.

If you can post an example where the K-5 IIs captures a detail that cannot be made visible by appropriate capture-sharpening of an equivalent K-5 image, then you have demonstrated a resolution advantage.

The test charts can be misleading in a way, as you benefit more from the K-5 IIs's filterless design at long range than at short range.

This statement suggests to me that your K-5/lens combination indeed has an issue with focus near infinity.

It would be really worthwhile looking into this and validate the micro-AF-adjustments.

Why would the absence of an AA-filter have a greater effect if the subject is further away? All that matters is the image that is projected on the sensor, not how far away the original scene is.

The only thing that I can image to change with subject distance are incident angles on the sensor. However,

1. I do not suspect the FA 85/1.4 to have an issue for the examples shown, and
2. I don't see how the absence/presence of an AA-filter would mean that different incident angles map to different levels of sharpness.

P.S.: Even when you provide a level playing field for the K-5 and K-5 IIs, you could still argue that the latter is superiour in many cases because the required capture-sharpening for the K-5 (that can be omitted for the K-5 IIs) can introduce additional noise. One may have to adjust the threshold parameter when using an unsharp mask, for instance, in order to make sure that the sharpening does not introduce grain where it would be visually unappealing.

Also, I'd love to be able to turn off sharpening in LR altogether, because LR performance is much better if you can switch off the detail panel completely.

[Adam]

When objects are far away, they're generally smaller with respect to the sensor size (i.e. they aren't magnified as much), and that's what I'm getting at. Here's how I look at it: a given subject takes up more of the frame, a sensor with lower resolution will only be marginally worse than one with higher resolution, because there's more detail to work with overall. On the other hand, if you shoot an object that's very far away, only a couple pixels may influence how that object ends up being rendered. If anti-aliasing then blurs all those pixels, the detail will not be recoverable through sharpening algorithms. You can use this same argument to say that in many situations, the image quality of APS-C sensor is "good enough" and you won't benefit much from FF.

I have a photo which I think can demonstrate this (it's the one with the hut). It wasn't used in the review because the WB changed after the sun showed itself, but I think if I start with the raw files, it shouldn't matter much if all we're looking at is resolution. I'll try to post this example in a few minutes.

[Class A]

On the other hand, if you shoot an object that's very far away, only a couple pixels may influence how that object ends up being rendered.

How difficult it is to accurately render a subject only depends on the subject detail. If you scaled down a hut, while you are approaching it, you would have to record the same detail close to it as you have to far away from it.

The Siemens stars show detail at close distance that neither camera can resolve. You don't need to increase subject distance to generate detail that is hard to record for a K-5.

If anti-aliasing then blurs all those pixels, the detail will not be recoverable through sharpening algorithms.

Here, you err.
Deconvolution sharpening can recover all the details.

It can only recover what is resolvable with a 16.3MP sensor, of course. But the K-5 IIs cannot resolve more. It just shows the unresolvable detail as artefacts, rather than blur.

Only when original image blur and AA-filter blur add up then the K-5 IIs has an actual advantage. This is obviously in situations when you do not have optimal conditions, such as lens blur, diffraction, camera shake, etc. If someone says "I have such conditions all the time, I'll go for a K-5 IIs", that's great. They should go for it. But to simply say the K-5 IIs resolves more detail no matter what (and equals a 24MP FF camera) does not make sense.

You can use this same argument to say that in many situations, the image quality of APS-C sensor is "good enough" and you won't benefit much from FF.

You can say that about the K-5 already. Or the K100D. What "good enough" is, is highly subjective.

There is no doubt, however, that FF is about more than just resolution. You have more DOF control, higher dynamic range, better low-light capabilities, and less expense for higher IQ.

I have a photo which I think can demonstrate this (it's the one with the hut). ... I'll try to post this example in a few minutes.

Thanks, but can you please do me (and many others) a BIG favour and address my concerns regarding your K-5 shots?

They seem unreasonably soft for an FA 85/1.4 @ f/5.6. AFAIC, you must make sure that there is no lens copy issue, AF-micro-mis-adjustment, or any other reason that would bias the comparison.

[wtlwdwgn]

Thanks for the review Adam. I've just added a nearly new K5 (from a forum member) to go along with my K20D. When the really new camera bodies come out in a couple of years I'll be ready for a K5 IIs. I think Pentax may have gotten itself into a model numbering quandary though. Will the next one be a K6, a K4, or K5 IIIs?

[Adam]

Thanks, but can you please do me (and many others) a BIG favour and address my concerns regarding your K-5 shots?

Which shots are we talking about now? The test charts or others?

Here, you err. Deconvolution sharpening can recover all the details.

Never say always and never say never As far as my understanding goes, there is no way for sharpening algorithms to know deterministically know what transformation, if you want to look at it that way, caused the blur. Thus, the sharpening can't perfectly undo it, even if it can do a good job.

Here's a link to a raw file from the K-5:
http://pfpho.com/k-5_2/IMGP9843.DNG

Why not try your magic on it wrt/ the pines? Sorry for the dust.

Here's what the K-5 IIs could do and my best attempt:

[JeffAHayes]

While I could definitely see a marked difference in fine resolution of leaves on the trees, etc., in the one image where you posted a full-sized crop for both the K-5 IIS and the original K-5, the fact that Pentax DIDN'T do things like improve its AF engine and the number of AF points just wouldn't justify the added expense of upgrading to the new, filterless camera for me (and MOST of my shots are outdoor... macros and landscapes).

In fact, the more I read, the more I think that if I DO decide to get a backup for my 15-month old K-5 (with the lens-change button missing for going on 2 months, now! -- have to use a toothpick or similarly small metal tool -- like the end of a very small Allen wrench -- to change lenses!), I'll go with a camera that on first look I wasn't as impressed with: the K-30.

The fact that the K-30 includes focus peaking AND an improved AF engine -- PLUS the fact that it sells for even less than the discontinued original K-5 -- makes it look increasingly attractive as a backup body (for now, my "backup body" is a Canon PS IS20, lol -- nice all-in-one megazoom camera, but NOT something on which I can mount my -- current count, 8 -- K-mount lenses on -- and that includes 2 of the six DA* lenses). It may "bite me" to continue waiting, though (but since I just got a new MB and 3770 Core i7 processor to upgrade my build I did in mid-2009, $$$ are tight for now, anyway. So I'll probably keep waiting until I'm in "crisis mode," then get whatever I get with the $3.99 overnight delivery option on my Amazon Prime account.

Adam, as you said Pentax is making claims of a new, higher-end APS-C body and a FF body sometime in the coming year, money being tight, I may wait to see what the new APS-C looks like (FF would be a burden, since some of my best lenses -- like the DA* -- are optimized for APS-C and will vignette with a FF body). As for those of you who question where and when Pentax said they'll bring out a FF camera (considering how long they've taken), my counter to that would be the fact that they started putting DISCLAIMERS beside all their best -- like DA* -- lenses on their website that they WON'T be fully compatible with FF. I have serious doubts Pentax would have added those disclaimers were they not in plans to bring out a FF camera -- not that I care because I think unless you're shooting for National Geographic, or blow your pictures up and put them on billboards, the quality available with the current K-5 is adequate for even most professionals -- and quite a few are using it, as well as some shooting pro with the K-30 and/or K-01 (mostly Europe and Asia, but they're doing it -- I see their posts on a different forum quite often). One shot one of the new "Bond girls" for one of the foreign editions of either Vogue or Harper's Bazaar with the K-5 and took video of the shoot with the K-1.

I just went up on the Blue Ridge Parkway a couple weeks ago to shoot fall color and I can't see how a Nikon D4 or Canon's equivalent could have given me any better images than I got with the K-5 (note that the $6,000 body-only D4 has the same size sensor -- 16MP as the K-5, as well -- of course there are dozens of other things it has the K-5 doesn't).

At any rate, Adam, I concur with your review that except in RARE circumstances it's not worth upgrading to the K-5 IIs, bang-for-the-buck-wise, for certain. As to its moire issues, on a different forum, a Pentax shooter who has more lenses than I can count (including things like a 500 F4.5, etc.), expressed concern about possible moire patterns on the repeating patterns on bird feathers. As you said one of the few advantages of the II/IIs would be shooting birds with long lenses in early morning light, it would be great if you could experiment on possible moire issues of this nature (wildlife is one of my favorites, too).

Thanks for the review!
Jeff

[fast50]

Here is the PentaxForums.com in-depth review of the Pentax K-5 II/IIs. Enjoy!

Pentax K-5 II / IIs Review - Introduction - PentaxForums.com



Other useful links for the old K-5:
Canon 7D vs. Pentax K-5 - Introduction - PentaxForums.com
Pentax K-5 Camera Review - Introduction - PentaxForums.com

Please remember when forming opinions of value that there is a whole world outside the USA. The UK list price for the IIs is £150 more than the II. That's about $240 or, to put it another way, not a bargain.

[JeffAHayes]

Excellent point, fast50. ALSO, since a camera with no AA filter but a Bayer pattern on it is bound to produce some moire, some of the time (as has been proven with Adam's test), although I have no plans on switching to Fuji, the way they fixed the issue with their newest camera (by having a "random" arrangement of RBG pixels on the sensor, similar to how color film was made, has eliminated the need for an AA filter AND produces razor-sharp images with no moire!

I've read just one review, and have no idea if Fuji has a patent on this process, or not, but if they don't, changing sensors from a Bayer pattern to random distribution should become the new standard for how everyone makes digital camera sensors, I would think. My guess is that they do have a patent on the process, but maybe they'll license it. After all, the sensor in both the K-5 AND the Nikon D7000 was made by Sony. Until I see where this is going (if it is going anywhere, and Fuji's solution doesn't become a "one-off" or completely proprietary technology patented from every possible angle which they refuse to license). To me, the random-distribution model beats Bayer patterns with no AA filter hands down (unless I'm missing something). Really keeping me in a "wait-and-see" mode.
Jeff