I'm very curious if you have changed or softened your position. Among other points, these seem to jump out (if my understanding of the points are correct):
Stopping down a lens to induce diffraction is not an ideal solution to dealing with moire (beyond the loss of dof/ light control) because it's effect is not as accurate as having a proper AA filter do the same:
"That's why you use an AA filter, and not diffraction (when you can). Thank you for the comparison pictures re apertures btw.
AA-filters are really beamsplitters, with a very controlled spread ratio - and very close to zero spread outside the limit the designer choses. You can specify that you want a 6µm spread, and that will result in an octagon-shaped PSF with a very strictly defined outer rim."
The reason why attempting to minimize moire via PP is not ideal:
"Colour aliasing issues can never be "fixed" though, and the destructive power that effect wields is present at any magnification. The original information in the reality in front of the lens has been destroyed - forever. You would then have to resort to manual chroma in-painting and severe signal decomposition filtering techniques (which is what most "de-moire"-filters use). That's like using a smart sledgehammer on details. It doesn't really matter how smart the sledgehammer is, it's still a sledgehammer."
The reasons why Leica and some MF manufacturers don't use AA filters (which, I'm guessing are extremely expensive at very large sizes and/or low quantities):
"The reasons why Leica don't incorporate AA filters into their designs are quite simple:
1) they cannot afford it
2) since they're already using the cheapest, cheesiest materials available for the filter package, an AA filter would add to much thickness, and make the edge colour vignette even worse.
The reasons why MF manufacturers don't include AA filters:
1) they cannot afford it
2) they expect their users to be aware of the effect and do what they can when shooting to avoid it."
rico wrote:
That the 800 and 800E have the same resolution may sound like a truism, but is also likely true in fact. The caveat is testing both sensors with the same input that is free of frequencies higher than the sensor-site pitch can properly sample. As mentioned in prior FM threads, it is sufficient to stop down to reach that threshold of blur. Alternatively, use a conforming test pattern. If chroma/luminance moire can be induced, the image is invalid and resolution measurements from it are invalid. Such a determination is distinct from personal preferences for "crispy" images. I like crispy, too, but only in my cereal or fish-n-chips....Show more →
"likely true" means you yourself are not sure. I don't know how many time over the past few years, I've heard people say that the AA filter reduces effective resolution -- and seen examples that suggest it. Sean Reid has shown in his tests that the 12mp (Bayer) Ricoh GXR without an AA filter "outresolves" (his word) the 16mp (Bayer) Sony NEX-5 with an AA filter. We can throw theory around all over the place, but for me the proof will be in the eating. Show me. I've got a 14mp Kodak SLR/n without an AA filter sitting on the shelf here, and it is just as detailed as my 21mp Canon 5Dll (if I have the energy and patience today I can run a test with the same lens). And while I haven't used the Kodak SLR/n much, I've never had an issue with moire or "spurious resolution". Right now, I'm with Luka on this issue until proven wrong with real images.
(as an aside, when looking at a real image -- not a test pattern, how do you know the resolution you are seeing is "spurious", and what does it matter?)
Do you have a link? All I've seen is the comparison of one image. I'd like to see a comprehensive test covering a range of image types, a range of situations and a range in apertures, to know with more certainty just what the difference in using the 2 cameras might be in practice, what the pluses and minus might be and how that might figure in with how you work.
No, my position is fairly constant, and I've had it since I started using AA-less cameras quite some time ago. There are good points, and there are bad points. For the general, average photographer the bad points are worse than what the good points can balance out.
On this part: -"Stopping down a lens to induce diffraction is not an ideal solution to dealing with moire (beyond the loss of dof/ light control) because it's effect is not as accurate as having a proper AA filter do the same"
-I have backed off, since I actually didn't realize how much of a problem angles are for a RF shooter, or in a mirrorless construction. I was WRONG in those two cases - the AA-filter gives a better controlled spread in long exit pupil distances (DSLR cameras, DMF cameras), but the diffraction is a better tool in RF or mirrorless applications when you look outside the image center. Unfortunately diffraction is only usable as an AA filter at F8-9 and up in this camera (DSLR, D800, just below 5µm pixels), and 11-13 and up in a MF / M9 camera. This limits F2.8-5.6 shooting considerably.
There are however some situations where being without an AA filter gives you more than it takes away - but none of those does generally apply "to me". They usually don't actually apply to anyone lots of sharp work outside of outside those parameters either. When going for very deep DoF is one, when you usually only use WA lenses is one other. Other than this - nope, can't think of anything.
The concept is really quite counterproductive... Look at this chain of reasoning:
1) I need / want the sharpest possible image (note wording, do not use "resolution" here!).
2) I will get a camera without an AA filter
3) I will then make sure that none of my shots are sharp enough to create aliasing / moire
() - huh?
The only two scenarios where this chain of reasoning actually makes common sense, and increases image quality available to the user is if you belong to either of the two categories mentioned above (WA or always stopped down).
And since the trouble you CAN get into is way above most general users PP skill to even try and "smooth over", I can very easily see why manufacturers that CAN include the AA filter in their constructions do so. They have no wish in alienating their customers. One really noisy, dissatisfied customer is often gives the same amount of publicity as 100 satisfied users.
theSuede wrote:
The reasons why MF manufacturers don't include AA filters:
1) they cannot afford it
2) they expect their users to be aware of the effect and do what they can when shooting to avoid it."
It does not seem reasonable to assert that MF manufacturers don't include AA filters becuase they can't afford it. How much more can it cost? A MF back costs $30 - 40,000. Even if it another $1000 (and clearly the AA filter in a FF DSLR doesn't cost anywhere near that), that would be hardly noticeable in the scheme of things and hardly dissuade anyone from buying one.
Lotusm50 wrote: "likely true" means you yourself are not sure. I don't know how many time over the past few years, I've heard people say that the AA filter reduces effective resolution -- and seen examples that suggest it. Sean Reid has shown in his tests that the 12mp (Bayer) Ricoh GXR without an AA filter "outresolves" (his word) the 16mp (Bayer) Sony NEX-5 with an AA filter. We can throw theory around all over the place, but for me the proof will be in the eating. Show me. I've got a 14mp Kodak SLR/n without an AA filter sitting on the shelf here, and it is just as detailed as my 21mp Canon 5Dll (if I have the energy and patience today I can run a test with the same lens). And while I haven't used the Kodak SLR/n much, I've never had an issue with moire or "spurious resolution". Right now, I'm with Luka on this issue until proven wrong with real images.
(as an aside, when looking at a real image -- not a test pattern, how do you know the resolution you are seeing is "spurious", and what does it matter?)
I don't think we will get to an agreement on this even after a comparison between the D800 and the D800E. The reason is simply that it depends on the spatial spectrum of the image projected onto the sensor, and the criterion for resolution.
In the case where there is no aliasing, the non AA filter sensor will outresolve the AA filter sensor, using the "line pairs/image heigth at a certain MTF" criterion.
In the case where there is aliasing, the position is reversed. The disadvantage of destroyed detail is much more severe.
In the case of critical frequencies, there will be disagreement on the criterion used for resolution, and about the detail being true or false. Even after comparison with a higher resolution reference image, we might not be able to agree on what image that best represents the reference.
Lotusm50 wrote:
Do you have a link? All I've seen is the comparison of one image. I'd like to see a comprehensive test covering a range of image types, a range of situations and a range in apertures, to know with more certainty just what the difference in using the 2 cameras might be in practice, what the pluses and minus might be and how that might figure in with how you work.
Oh, sorry - misunderstanding.
We've ORDERED both versions, and are on the NPS priority list. That's what is "already done".
I've not DONE the comparison yet.
theSuede wrote:
No, my position is fairly constant, and I've had it since I started using AA-less cameras quite some time ago. There are good points, and there are bad points. For the general, average photographer the bad points are worse than what the good points can balance out.
On this part: -"Stopping down a lens to induce diffraction is not an ideal solution to dealing with moire (beyond the loss of dof/ light control) because it's effect is not as accurate as having a proper AA filter do the same"
-I have backed off, since I actually didn't realize how much of a problem angles are for a RF shooter, or in a mirrorless construction. I was WRONG in those two cases - the AA-filter gives a better controlled spread in long exit pupil distances (DSLR cameras, DMF cameras), but the diffraction is a better tool in RF or mirrorless applications when you look outside the image center. Unfortunately diffraction is only usable as an AA filter at F8-9 and up in this camera (DSLR, D800, just below 5µm pixels), and 11-13 and up in a MF / M9 camera. This limits F2.8-5.6 shooting considerably.
There are however some situations where being without an AA filter gives you more than it takes away - but none of those does generally apply "to me". They usually don't actually apply to anyone lots of sharp work outside of outside those parameters either. When going for very deep DoF is one, when you usually only use WA lenses is one other. Other than this - nope, can't think of anything.
The concept is really quite counterproductive... Look at this chain of reasoning:
1) I need / want the sharpest possible image (note wording, do not use "resolution" here!).
2) I will get a camera without an AA filter
3) I will then make sure that none of my shots are sharp enough to create aliasing / moire
() - huh?
The only two scenarios where this chain of reasoning actually makes common sense, and increases image quality available to the user is if you belong to either of the two categories mentioned above (WA or always stopped down).
And since the trouble you CAN get into is way above most general users PP skill to even try and "smooth over", I can very easily see why manufacturers that CAN include the AA filter in their constructions do so. They have no wish in alienating their customers. One really noisy, dissatisfied customer is often gives the same amount of publicity as 100 satisfied users....Show more →
Thank you very much for the explanation, very helpful.
Lotusm50 wrote:
Sean Reid has shown in his tests that the 12mp (Bayer) Ricoh GXR without an AA filter "outresolves" (his word) the 16mp (Bayer) Sony NEX-5 with an AA filter.
At least in Sean's recent article on LL, "The Naked Sensor", he does not claim the Ricoh 12MP non AA out-resolves the NEX-5 with AA when both are compared at the same size. His point seems to be that just at 100% view, the Ricoh has better per pixel sharpness/ detail. Here is what he says:
"But what I also discovered (in the course of doing careful, controlled, focus-bracketed studio tests of resolution) was that - with every tested lens - the Ricoh out-resolved the Sony on center. The key difference between them, in this respect of course, is that the Sony uses an AA filter and the Ricoh does not. Of course, if one has to up-res. the 12 MP Ricoh file to match the dimensions of the 16-MP Sony file the former's resolution advantage is lost."
Tariq Gibran wrote:
At least in Sean's recent article on LL, "The Naked Sensor", he does not claim the Ricoh 12MP non AA out-resolves the NEX-5 with AA when both are compared at the same size. His point seems to be that just at 100% view, the Ricoh has better per pixel sharpness/ detail. Here is what he says:
"But what I also discovered (in the course of doing careful, controlled, focus-bracketed studio tests of resolution) was that - with every tested lens - the Ricoh out-resolved the Sony on center. The key difference between them, in this respect of course, is that the Sony uses an AA filter and the Ricoh does not. Of course, if one has to up-res. the 12 MP Ricoh file to match the dimensions of the 16-MP Sony file the former's resolution advantage is lost."
alundeb wrote:
I don't think we will get to an agreement on this even after a comparison between the D800 and the D800E. The reason is simply that it depends on the spatial spectrum of the image projected onto the sensor, and the criterion for resolution.
In the case where there is no aliasing, the non AA filter sensor will outresolve the AA filter sensor, using the "line pairs/image heigth at a certain MTF" criterion.
In the case where there is aliasing, the position is reversed. The disadvantage of destroyed detail is much more severe.
In the case of critical frequencies, there will be disagreement on the criterion used for resolution, and about the detail being true or false. Even after comparison with a higher resolution reference image, we might not be able to agree on what image that best represents the reference. ...Show more →
If this is correct in practice, in what portion of what you use the camera for would you experience aliasing? I suppose if most of what you do is photographing textiles, then I suspect your decision is made. If the experience of users of cameras without AA filters is to be taken, then we get the impression that this portion of use that experiences aliasing is very small -- and when it does appear most of the time it can be corrected in post.
theSuede wrote:
No, my position is fairly constant, and I've had it since I started using AA-less cameras quite some time ago. There are good points, and there are bad points. For the general, average photographer the bad points are worse than what the good points can balance out.
+100
The concept is really quite counterproductive... Look at this chain of reasoning:
1) I need / want the sharpest possible image (note wording, do not use "resolution" here!).
2) I will get a camera without an AA filter
3) I will then make sure that none of my shots are sharp enough to create aliasing / moire
Bingo !!!
The only two scenarios where this chain of reasoning actually makes common sense, and increases image quality available to the user is if you belong to either of the two categories mentioned above (WA or always stopped down).
Or Both
And since the trouble you CAN get into is way above most general users PP skill to even try and "smooth over", I can very easily see why manufacturers that CAN include the AA filter in their constructions do so. They have no wish in alienating their customers. One really noisy, dissatisfied customer is often gives the same amount of publicity as 100 satisfied users.
Spot On
Tariq Gibran wrote:
The reason why attempting to minimize moire via PP is not ideal:
"Colour aliasing issues ... it's still a sledgehammer."
And it generally sucks to use a sledgehammer when you're striving for detail ... see above @ the Suede's 'counter-productive' reasoning. The best antidote that I've found for color aliasing is B&W conversion. If you have that luxury/discretion with your work it's a nice 'work around' for when you get bit.
If Canon (plays nicer than Nikon with all my alts) comes out with a high res non-AA body ... I'll likely upgrade from my SLR/c. But ... I still strongly warn anyone who is thinking about using the 800E for anything production related where you don't want to contend with the pains of moire ... which always come at the WRONG time (i.e. is there ever a good time).
If you look at my profile, you'll see my lenses are heavily weighted toward the WA side of things, and I'm constantly aware of my subject matter (mostly organic). I'll have to try and find the pic (pretty sure I trashed it though) of my first outing with my SLR/C ... took a routine/mundane picture of a golfer at the driving range with tele glass ... bingo MOIRE in his shirt and golf bag right out of the gate. More recently, I was shooting a rusty old truck with my 24L TS-E stopped down and still got bit by the moire bug.
I won't even bother to try and speak to the theory or technicals ... (we all know how that usually turns out ) ... but I can simply tell you from a routine and regular user of a non-AA filter system perspective ... moire is real and you must be A) vigilant to it's potential, B) ready to tackle its perils and C) able to tolerate its consequences.
If you can deal with the ABC's of moire' ... GO FOR IT !!!
But if you are doing weddings, portraits or any other kind of production work ... not a snowball's chance in Hell that I'd recommend it as my primary body (studio work that you've got "dialed in", maybe ok). Working your butt off to develop your skills and investing in excellent equimpent, etc. to get your keeper rate high ... only to have moire lower it ... not a good thing, imo.
Non-AA works for me and my approach/style for most of what I like to do. I've got other bodies (5D, 1D II, 1D IIN, D70s) for other applications, and while none of what I have is close to the likes of the D800/D800E ... I'm not trying to prove anyone or anything right or wrong about it ... I simply offer my experience (i.e. not theory) with non-AA to my fellow FM members. You guys have helped me immensely over the years in my decision making, I'm just trying to reciprocate.
Lotusm50 wrote:
OK, but it seems to me the point about the AA filter is still made.
The point per pixel is there but I guess it comes down to exactly how much increase in resolution one might see vs. the risk of shooting without the AA filter and dealing with moire and other artifacts. If we were to say that, for instance, a 12MP non AA sensor equals the actual resolution of a 16MP AA sensor as above (and I don't know if that's the case but just being conservative), then how much does that actually get you? Is it even 5%? So, is this the extent of the gain? I don't know, I'm just curious myself.
Thorsten wrote:
There is only one version of these cameras, without AA filter, so no point in debating it. In the rare case when moire hits, you just deal with it. If Nikon had released only the D800E, I don't think anybody would bother to discuss the AA filter or lack thereof. It would just be another small item in the 100 spec items that are new with this camera.
Judging by this conversation and the amount of passionate users who are so strongly anti No-AA I'd disagree. I agree more people would just be like "well we'll deal with it" but it'd still have people frothing with anger.
Thats besides the point of my question though, if it was such a common and massive issue on either of those cameras wouldnt people not have purchased them or have brought it up as a problem more often?
As I was first working with the X100 I was taken with how crisp the images were. They reminded me of these from the Leica M9. At first I simply thought that this was due to their functional similarly, but that didn't make sense. There's nothing inherent about an optical viewfinder camera that makes it sharper.
It wasn't long before I realized that it was likely that, like the M9, the X100 didn't have an AA filter (Anti-Aliasing Filter, AKA blurring filter, AKA moire remover). I figured that it shouldn't be too difficult to find an example to determine this one way or another. Right? Wrong!
The major camera makers would have you believe that the world is full of lurking examples of subjects that will jump out at you and ruin your photographs with moire. The truth of the matter is, it's very hard to find such examples when you're looking for them, let alone in day-to-day photography.
If you're a fashion photographer this may not be the case. Your world is full of man-made fabrics with very fine patterns that can "beat" with the grid frequency of your camera's sensor. For you, a Nikon, or Canon, or other DSLR with a strong AA filter is what's needed (or, lots of time in post-processing). For the rest of us? Not so much....Show more →
If its such an issue for fashion photography why do they still often opt for AA filterless MFDB's?
I can tell you that with my experience with the X100 I had for 6 months I didn't run into moire issues once. I wish more M9 users could join in the conversation and offer their input, one or two don't seem to be enough to sway any opinions. Maybe some pixel peepers can scan through these photos and find us a lot of examples of ruined shots.
I have pre-ordered a D800E but I'm still entirely open to the D800. When it comes to this camera we're all still working off speculation on the frequency of its occurrence but I've based my decision on how other non AA sensor camera users are reacting and I'm just not seeing it as the great negative some others are. Hopefully production cameras get in the right testers hands soon so we can all find out if we made the right choices for our style or not. Hell I'd be happy to find out the D800 is better suited for me so I could use that $300 and get a 85 1.8
Wonger wrote:
Judging by this conversation and the amount of passionate users who are so strongly anti No-AA I'd disagree. I agree more people would just be like "well we'll deal with it" but it'd still have people frothing with anger.
Thats besides the point of my question though, if it was such a common and massive issue on either of those cameras wouldnt people not have purchased them or have brought it up as a problem more often?
If its such an issue for fashion photography why do they still often opt for AA filterless MFDB's?
I can tell you that with my experience with the X100 I had for 6 months I didn't run into moire issues once. I wish more M9 users could join in the conversation and offer their input, one or two don't seem to be enough to sway any opinions. Maybe some pixel peepers can scan through these photos and find us a lot of examples of ruined shots.
I have pre-ordered a D800E but I'm still entirely open to the D800. When it comes to this camera we're all still working off speculation on the frequency of its occurrence but I've based my decision on how other non AA sensor camera users are reacting and I'm just not seeing it as the great negative some others are. Hopefully production cameras get in the right testers hands soon so we can all find out if we made the right choices for our style or not. Hell I'd be happy to find out the D800 is better suited for me so I could use that $300 and get a 85 1.8 ...Show more →
This is a rational and reasonable perspective based on what we've seen to date from a variety of manufacturers and products. It is pretty similar to my view. All these people using non-AA filter camera don't seen to be complaining or consistently producing fatally-flawed images due to their missing AA filter. I too am open the the regular D800 if it can be demonstrated to be better for me. Right now, the sample images, excluding the kimono image, just look better from the D800E.
theSuede, reading your post is a good learning experience.
If I am wedding photographer (depend on how I am mostly shooting), or professional frequently deal with production shot. I will choose D800.
D800 and D800E has the exact same resolution, but different contrast because AA filter. so image will look sharper generally with D800E, assuming there is no Alias happen.
(I am not so sure about below statement, I am listening.) Without AA, The sharper image is not come from Spurious resolution or alias or whatever its called. It is from no smudging the high frequency contrast by AA filter.
It is controversy, at one hand, you want image to be as sharp as possible, at other hand you don't want it is too sharp to capture a lot high freq stuff alias to your image. BUT, BUT, this is only happened at:
A. You have extreme sharp lens out-solve sensor.
B. You have a case have many high freq pattern in your image. For random stuff like most landscape shot, I doubt there is any frequency spike out there, other than high freq noise floor way lower than your Camera DR can capture, even there is alias, this is not a issue. Again I could be wrong, I am listening.
So far, most moire example, I see is either fabric, or chart at lab to exaggerate the issue. Theoretically, without AA filter, you will have alias. Real world, you have AA filter in front of your sensor as listed: lens, camera vibration (Not exact sure how this will affect things.) and diffraction AND having fixed high freq pattern in your image AND will it affect your image? AND can you fix it in PP?
All in all, chances you get screwed are low.
Experience can tell you past, but there is no such high pixels density AA free camera exist before. How does it apply here?
KaaX wrote:
If your lens does not outresolve your sensor, your gain from having an AA-less sensor is precisely zero.
See, you didn't get the point, your lens may not out solve the sensor, but it out solve the frequency at AA filter kick in. of course, you gained higher contrast here.
BTW ... it should be noted that I shoot with M645 glass for my longer stuff on the SLR/c. I should run a test shot or two with SLR glass vs. MF glass to see if I can induce moire differently between them.
I like crispy,
) - huh?
) ... but I can simply tell you from a routine and regular user of a non-AA filter system perspective ... moire is real and you must be A) vigilant to it's potential, B) ready to tackle its perils and C) able to tolerate its consequences.
