[ironlionzion]

True! I never thought of bird feathers. Come to think of it, a lot of animals have repeating patterns. Hmm...

[crewl1]

This is a moire eel. If you take this picture with a K-IIs the eel will look like it has little diamond shapes on it.



And the camera will stop working soon after (because you immersed it in seawater.)

[RyanW]

Ya, no one would want this version - i mean who would want higher resolution/sharper images? Who would actually want to improve a soft 2.8 wide open lens/shot into a sharp 2.8 wide open lens/shot?

Mushy is mushy. The glass has to be up to the task first. Removing the AA filters isn't magically going to present you with detail the glass can't pick up.

So if the glass is not the bottleneck, yes it would be an improvement assuming moire isn't an issue.

[RockvilleBob]

Thanks for the explanation and the excellent Nikon D88 vs. D800E explanation. Since I am not looking at 3 foot wide prints, except for panorama shots where even this size would be a limitation, I can see no reason to get the IIs.

[goubejp]

True! I never thought of bird feathers. Come to think of it, a lot of animals have repeating patterns. Hmm...

I wonder if it will be really an issue; most of the times with these long tele lenses the 16 Mpix sensor outresolve the lens; anyway as I have 2 K5, I think I'll get one K5-IIs to shoot with long lenses, and keep a K5 for use at short distances with the DA*200.

[goubejp]

And I've seen feather detail in birds taken with the D800E which showed clear moire and compared that to the D800 which had none. It is a bit of a dice roll though.

May be you are referring this test :

A bit unrealistic for a birder; usually when you shoot a whole bird with a long lens you can't reach this level of magnification in feather details IMO

[Frogfish]

May be you are referring this test :

A bit unrealistic for a birder; usually when you shoot a whole bird with a long lens you can't reach this level of magnification in feather details IMO

Of course you can and not unrealistic at all. I could easily have sharpened the feather detail here and magnified the crop (and I notice the forum software has softened these - they are much sharper on my screen). These were at ISO 2,000 so imagine the detail at ISO100 - 400. This though is largely irrelevant since you don't need to show such detail to see moire in bird feathers.



Feather detail - no sharpening or NR applied.

[Class A]

Many people have posted shots from the D800E that show repeating patterns without moire.

There are many reasons why moiré does not always rear its ugly head:

1. The spatial frequencies in the scene are too low.
2. The lens is wide open and acts as a low-pass filter.
3. The f-ratio is high enough to cause sufficient diffraction.
4. The lens is slightly defocused.
5. There is motion blur from camera shake or subject movement.

In all these cases it would be better to remove the cause of the problem rather than slightly mitigating its effect by not using an AA filter.

Conversely, if you manage to make all stars align so that image sharpness is optimal, capture sharpening can completely restore any loss in contrast when using an AA filter, while not using an AA filter just makest moiré likely to occur.

Contrary to what is often stated, moiré cannot easily be removed:

• It is always a manual process, requiring detailed attention. There is no way to "batch process" images, repairing moiré in an automated manner.
• Information has irrevocably been lost. Some invented information has to replace the ugly moiré patterns and this patching may or may not look good.

In contrast, the capture sharpening required to remove the little loss in contrast caused by AA filters, can be applied globally to an image and thus be made part of a default import setting (e.g. in Lightroom).

Once an appropriate level of capture sharpening has been applied to an image recorded through an AA filter, there is no significant difference to an AA filter-less version. This is a 100% true for images with no blur. Only when images already contain some blur, an AA-filter may actually destroy some resolution.

BTW, "AA filter" is a misnomer for the birefringent plates that are put before imaging sensors. Watch the following demonstration of birefringence.

You Tube

You Tube

Bear in mind that a sensor using a Bayer-filter matrix absolutely needs an approach like this (or equivalent) unless you are talking about images with (relatively) low spatial frequencies only.

No doubt sooner or later someone will publish comparative lw/ph resolution figures showing how much more the K-5 IIs resolves compared to the K-5 II.

I'm just hoping that

1. These figures will be obtained using appropriate levels of capture sharpening respectively, and
2. the figures won't be based on B&W patterns and an approach that ignores the false colour that will be produced by an AA-filter less sensor.

The main purpose of an AA filter in front of a Bayer-matrix is to prevent colour-moiré. This slightly impacts on the sensor's ability to resolve the finest B&W patterns (in a situation where you know that the source is B&W and you then can faithfully reconstruct the B&W pattern using RAW data), but in the real world we are typically facing colourful subjects.

A more useful representation of a sensor's resolution capabilities would be measuring colour resolution. With the current B&W patterns, AA filter-less sensors appear to be resolving more detail (partly due to clever demosaicing algorithms) than they actually do, if you look at true colour resolution. In other words, do not believe that any lw/ph figure advantage the K-5 IIs may demonstrate, translates 1:1 into more detail in colour images.

One way of looking at an AA filter-less sensor is to observe that each part of a scene that corresponds to the size of a sensel on the sensor is only seen with one colour. The sensel is colour blind to other colours and hence only records one colour independently of the real spectrum the particular portion of the scene has. The AA filter, however, manages to expose the same portion of the scene to four sensels (which in combination see all colours), making sure that its full colour can be captured.

There is a reason why manufacturers have been equipping DSLR with the cost-adding, precision element called "AA filter".

Falk (falconeye) wrote this:

Personally, I consider a sub-100MP camera w/o Bayer-AA filter technically flawed which is and must be sold with a disclaimer.

He also has some good things to say about removing AA filters; his bottom line is "The ommission of the Bayer-AA slightly improves many images and massively deteriorates a few. The exact balance depends on the #MP."

IMHO, there are only very few people -- who have full control over their subject matter -- that can truly benefit from an AA filter less camera.

The vast majority is much better served with a complete image forming system.

[goubejp]

Of course you can and not unrealistic at all. I could easily have sharpened the feather detail here and magnified the crop (and I notice the forum software has softened these - they are much sharper on my screen). These were at ISO 2,000 so imagine the detail at ISO100 - 400. This though is largely irrelevant since you don't need to show such detail to see moire in bird feathers.



Feather detail - no sharpening or NR applied.

Nice shot ! I don't think this kind of shot will produce very visible moiré effect; the feathers are small and each have a slightly different direction;

[Frogfish]

Nice shot ! I don't think this kind of shot will produce very visible moiré effect; the feathers are small and each have a slightly different direction;

Thank you !

Well that's just one bird. I have seen examples of moire on birds with the D800E (on FM) which I have not seen with my K5 or D800. Therefore it seems that cameras sans an AA filter may well be inclined to elicit moire - it makes me a little dubious about replacing my K5 with the K5IIs

Maybe Google would find you some D800E bird moire examples ?

[goubejp]

Thank you !

Well that's just one bird. I have seen examples of moire on birds with the D800E (on FM) which I have not seen with my K5 or D800. Therefore it seems that cameras sans an AA filter may well be inclined to elicit moire - it makes me a little dubious about replacing my K5 with the K5IIs

Maybe Google would find you some D800E bird moire examples ?

I think I'll experiment by myself; I currently have 2 K5; I'll keep one and exchange the other for a K5IIs to use with the FA*600 which I own; the K5 with either the DA*200 or DA*300 for short distances, as I usually stop down at F5.6 with these lenses, the K5IIs with the FA*600 which I usually stop down to F7.1 or F8 to have some depth of field; at F8 the slightly blur effect caused by diffraction should be enough, according to D800 E test samples. That's the way I work typically when I'm in a hide.
When I google bird D800 moiré the most responses I got are from... the D800, people experienced moiré issues caused by the weak AA filter.
An example of shot which should be moiré-free cause the feather are randomly oriented... and the feathers on the tail are blurry

[Frogfish]

I think I'll experiment by myself; I currently have 2 K5; I'll keep one and exchange the other for a K5IIs to use with the FA*600 which I own; the K5 with either the DA*200 or DA*300 for short distances, as I usually stop down at F5.6 with these lenses, the K5IIs with the FA*600 which I usually stop down to F7.1 or F8 to have some depth of field; at F8 the slightly blur effect caused by diffraction should be enough, according to D800 E test samples. That's the way I work typically when I'm in a hide.
When I google bird D800 moiré the most responses I got are from... the D800, people experienced moiré issues caused by the weak AA filter.
An example of shot which should be moiré-free cause the feather are randomly oriented... and the feathers on the tail are blurry

Nice shot Goubejp.

That is a similar set-up to my current one (K5+DA*300 and D800 + Sigma 500mm). The Sigma I'd usually shoot at f8 unless the light necessitates opening it up, then even at f4.5 it is still really sharp. I have never noticed any moire at all with my D800 - though over on FM, as a lot of nature shooters are using the D800E, there has been some complaints re. moire, though not often.

[Winder]

Can someone explain why Pentax would offer a variant without the AA filter? What type of photog would want this particular feature?

My guess is they are testing the market for a future model. If they are going to a 24MP APS-C sensor then there is no reason to have the AA filter and they want to put a K-5 super model on the market to see how it performs and how customers react.

We wont have AA filters in the future and this is a way to transition people over and get people use to the idea.

The new 24MP APS-C sensor without the AA filter would easily outperform the K-5 in terms of noise, color, & detail.

[kh1234567890]

We wont have AA filters in the future

No, we'll just oversample with our 100 megapixel sensors and sort it all out in software. Hardware solutions are so passe ...

[miles]

Moire: no big deal. Surely it's even easier to remove from a raw file. Software will soon be better at identifying it, since going AA-less looks like where we're heading (see recent Pentax interview).

I just edited this pic posted here in under one minute using the gimp... With other examples i've seen, C1 pro makes it disappear with its moire tool (2 sliders).

/

original / no moire