Do you think they will make a d610 with no low pass filter

kiekie

Just think as I will be up grading to the d610 from the d5100. but if the bring out a lower end FX camera with no low pass I would get that as I love landscape and street photography.

spraynpray

Won't be anytime soon if they do.

Pistnbroke

Why dont you go for a D7100 ..My tests taking the same subject with a D800 and D7100 show almost no difference and its a tiney almost...So I dont think D6100 will show any difference at all. The advantage to FX is that if you are going wide angle then a 15mm on FX will be better than a 10mm on DX. The low light argument does not work because you need to stop down more on FX to get the same depth of field as DX

heartyfisher

One of the unpublicized features is that the low pas filter on the D610 is weaker than the D600 so when used with very sharp glass you get about 10% more resolution. Love it ! just get the D610 .. I think its great value! I have one !

donaldejose

Yes, I think no AA filter is the future of all Nikon DSLRs. It is just a matter of time.

TaoTeJared

Yes, I think no AA filter is the future of all Nikon DSLRs. It is just a matter of time.

I agree with this not sure if I like it though. I hate Moiré in my images and I have never been good at removing it in post. I can easily bring up the sharpness and clarity in post to make up having the filter though.

Just think as I will be up grading to the d610 from the d5100. but if the bring out a lower end FX camera with no low pass I would get that as I love landscape and street photography.

There is some bloggers that get real idiotically ridiculous about Anti-aliasing (low pass) filters. Leica used strong filters so those shooters saw a big improvement. Everyone else had been lowering the strength with each body release for the last 8 years to where it barely makes any difference.
The improvement on images is almost nothing. Cropped at 1600% maybe, depending on the subject you can see a bit more definition. I have never seen any resolution numbers like heartyfisher claims - the most I have seen is about 2-3% or about 50-200 lines. That is basically nothing. If you edit at all, especially add noise reduction, your resolution drops a good deal and you can easily get the added sharpness in post. That just negates any minuscule advantage and makes the difference between a body with and a body without the AA filter zill. My point being is that it really shouldn't be a decision point at all. If you shoot street type stuff, you want the AA filter as the backgrounds are continuously full of repetitive lines that cause Moire and distracts from the subject.

spraynpray

I agree with the above comments on moire and am surprised @Pistnbroke hasn't commented on that too because as somebody who shoots weddings he would have seen a heck of a lot of moire on suits and dresses and in a lot of cases it isn't correctable. If the gain is as small as TTJ says, I am surprised that they sell them without AA filters at all. Possible explanation is marketing triumphing over knowledge.

heartyfisher

There are no "number" published for the difference between the D600 and D610 in fact Nikon doesnt even mention it. However, when I was researching the D610 vs the D600. I at first found comments from bloggers and reviewers that the D610 seems sharper. The numbers I stated with the 10% are my own reckoning from data on DXO tests. Maybe i exaggerated it a bit, but if you look at the the 70-200 F4 and see the "sharpness" results the value is 18 vs 20 for D600 and D610 respectively. so approximately 10% for this one lense. I am sure there are rounding issues as the 85mm has 20 and 21. Making it about 5%. So averaging to about 7.5% .. yes its hardly noticeable. but its nice to see the improvement.

PS: another interesting thing that I got out of my wondering around DXO data is that the 18-200 on DX mode on my D610 has about the same PMpix as the 28-300 in FX mode. however since I can use a larger surface area in FX mode with the DX lense (ie about 1.35 crop mode) it should actually be better to use that DX lens on my D610 !

haroldp

Aliasing happens mostly when the lens outresolves a beyer sensor. Therefore, the finer the pixel pitch (more pixels in a given area) , the less likely it is to happen, as fewer lenses at aperture/ focus / depth of field are likely to encounter an appropriately scaled pattern.

This is the reason that Nikon gave up AA filters on 36mp FX sensors and 24mp DX (whose pp is that of a 54mp FX sensor).

Is 24mp FX fine enough ?, probably. my 18mp Leica M9 very rarely gets any artifacts, ( I would not use it at a wedding, bridal veils and herringbone suits are scary), but my D800e has never shown me one (maybe I do not look hard enough) and I do use it at weddings, with very good lenses.

.... H

donaldejose

"24mp DX (whose pp is that of a 54mp FX sensor)." Nikon has produced many 24 mp DX sensor which work very well so there shouldn't be any technical problem with them producing a 54mp FX sensor which works about one stop better because it collects more light on its surface. An approximately 50mp FX sensor should be appearing soon, I would think. Nikon has all the experience with it they need. Just have to have a supplier be able to produce them at a reasonable price and they should appear in a D4x or D810x within two years. However, Nikon may also need to produce an Expeed 5 processor to handle the data crunching.

kanuck

I am fairly certain they will not and keep this in their D800 line only for FX unfortunately. D800 prices will drop now though! However, you might feel very comfortable with the D600 line. U1 and U2 options, 24mp comfort zone, which I fully understand B-)

donaldejose

I love my D600 (once I got the sensor cleaned and got it to stay clean)! But I do find the smaller focus area coverage a pain at times. Unfortunately, the D610 didn't fix that issue! Maybe the next iteration will. Nikon's 24mp FX sensor is great but it could benefit from better camera components such as the focusing module.

Fred_B

Adding higher pixel density only reduces the size of the repetitive pattern that will have aliasing issues. When we had low resolution sensors a filter was a must because a simple striped shirt would have issues. With the density on the D800 it takes a very small repetitive pattern to show a problem (often you have to zoom way in to find it).

haroldp

Adding higher pixel density only reduces the size of the repetitive pattern that will have aliasing issues. When we had low resolution sensors a filter was a must because a simple striped shirt would have issues. With the density on the D800 it takes a very small repetitive pattern to show a problem (often you have to zoom way in to find it).

Aliasing happens when edge acuity (lens resolution/ contrast) crosses a beyer photosite pattern in particular ways, which only happens when the lens is capable of such resolution. The smaller the photosites, the higher the lens resolution needed to get there.

An AA filter lowers the resolution of the lens below the nyquist frequency of the sensor thereby 'blurring' edges so that edge acuity at needed contrast does not occur.

If the lens resolution stays constant, a finer pp sensor will be less likely to suffer artifacts.

cellphone and compact camera sensors have discarded AA filters for years now because with their tiny photosites (1-2 micron, D800 photosites are over 4.88 microns), their simple lenses have no chance of such resolution so the lens is itself an AA filter.

While 4.88 microns is sometimes achievable by Nikons best lenses, It requires absolute focus point and technique to get there, to which one would have to add the appropriately scaled and shaped repetitive pattern, which is why no AA filter is an acceptable risk for some of us.

If lens resolution was perfect (absolute contrast at edges across infinitely small areas, then your statement would be true.

Such lenses are not scheduled for 2014.

Ironheart

What you are saying is that diffraction is starting to set in:
http://forum.nikonrumors.com/discussion/2918/diffraction-at-small-apertures-a-limiting-factor-in-sharpness

If you have a 1/2" 12MP sensor (and cell phones have even smaller than this) you are diffraction limited at f/3! No need to blur the image any further...

haroldp

With current FX 36 mp sensors, all but the best lenses will be aberration limited right up to the point of diffraction limiting and never be sharp enough to cause aliasings, Even the best lenses will require perfect focusing and technique to come close.

,,, H

Fred_B

I agree to disagree. It's not hard to find many examples of aliasing with the D800E. I've seen it with my D800 but you have to try to make it happen.

spraynpray

I agree to disagree. It's not hard to find many examples of aliasing with the D800E. I've seen it with my D800 but you have to try to make it happen.

Can you explain how one tries to make aliasing happen? I would expect to have to try to not make it happen with my D7100 on fine patterned materials.

haroldp

I agree to disagree. It's not hard to find many examples of aliasing with the D800E. I've seen it with my D800 but you have to try to make it happen.

No one (certainly not me) said it can't happen, only that in the course of a lot of my photography where I was trying to make photos, not moire, I have not seen it.

.... H

spraynpray

OK then, perhaps @haroldp you could explain how you 'try to make moire'? In my experience it is a result of things that are out of ones control.

haroldp

Of course it can happen randomly, it just has not happened to me with the D800e.

If I wanted to cause it, I would pick a high contrast regularly spaced pattern like a grid, herringbone also works well, and I would shoot with direct light (for contrast) at varying distances trying to find the exact magnification that would interact with the pixel pitch of the sensor.

There is no guarantee because conditions would have to be perfect, but after enough effort, it would probably work.

I have no intention of ever actually doing this.

I have had occasional moire with my Leica M9 (3 frames out of 25,000), with 18mp FF sensor no AA filter and extremely sharp lenses.

So far, none with D800e.

Regards ... H

Ade

While 4.88 microns is sometimes achievable by Nikons best lenses, It requires absolute focus point and technique to get there, to which one would have to add the appropriately scaled and shaped repetitive pattern, which is why no AA filter is an acceptable risk for some of us.

Unfortunately that's a bit simplistic. 4.88 microns is achievable by practically any lens. It is only a question, for a given lens+sensor pair, at what MTF this resolution is achieved? At MTF 50? Probably not. At MTF 10? Maybe yes. Can a signal near the Nyquist frequency captured by the lens+sensor at MTF10 still cause aliasing? Probably yes.

So in reality aliasing can still happen without the best lenses and/or the best technique.

TaoTeJared

There are no "number" published for the difference between the D600 and D610 in fact Nikon doesnt even mention it. However, when I was researching the D610 vs the D600. I at first found comments from bloggers and reviewers that the D610 seems sharper. The numbers I stated with the 10% are my own reckoning from data on DXO tests..cont.

Bad numbers to use for so many reasons but beginning with they are "end" number results. Just the Raw resolution numbers are good enough to tell.

The other thing one has to remember is that every manufacture has been "cooking" their internal software to get the "Best" results out of the big testing outlets so small changes could just be software. Perceivable differences in printed or normal viewed work is another aspect that is continually lost - mostly since there is none.


A couple of things to note on Moiré:
•For moiré to appear, you have to be at the right angles. Some of it can be avoided by simply moving yourself or your subject just a bit to a different angle (if you have that ability.)
•Nikon has been adding software "fixes" to reduce moiré and I have noticed with the recent firmware updates moiré has reduced. (I always catch it in ties and some women's scarves.) The issue here is that just like Noise reduction, "smudging" happens and detail is lost. So really with, or without the AA filter, sharpness is lost.
• There is an argument that exists which shows that pixel sizes are reaching a point "in-between" wavelengths. I have read a couple of articles that have said that a pixel pitch a little above 4 microns will be in between wave lengths. Nikon has eluded to this in various statements as well. This argument along with what lenses can resolve reaches into the differences of measuring analog vs digital.
•One must also remember Pixels on a sensor never equal pixels on an image - it is always a calculation of multiple pixels. That calculation can be tweaked and also has it's own inherent issues.

Ade

For photography (visible light) we are no where close to having pixel sizes reaching "in-between" wavelengths.

E.g., the wavelength of the dominant green light is around 550 nm (0.55 microns), an order of magnitude smaller the pixel density of the D800.

At 4 microns, what's happening is we're getting near the diffraction limit. E.g., for green light at 550 nm, a numerical aperture of 2.8, the diffraction limit is 3.7 microns. We can't resolve anything smaller than this limit (which has implications for optical microscopes, etc.)

A pixel smaller than this size (SDL: sub-diffraction limit) is still much larger than the wavelength of visible light.

However, even supposing we define visible light as starting at 800nm (long edge of red), an FX sensor with a pixel pitch of this wavelength would have a resolution of 1350 megapixels (1.35 gigapixels).

An FX sensor with "deep SDL" pixels -- those with a pixel pitch matching 550nm green -- would have a resolution of 2.8 gigapixels.

We're long ways off from having these kinds of pixel densities in our sensors. I think it will be some years before mainstream FX sensors will cross even the 100 megapixel boundary.

Ade

@PitchBlack

A Foveon-type sensor can't "beat" these limits and is susceptible to aliasing & moire.

But, I think you're intuition is correct, in the sense that our conventional "Bayer" sensor is even more susceptible to aliasing and moire than a Foveon sensor. Let's look at the following diagram from Wikipedia to see why:


Bayer R, G, and B filter locations on the sensor (Wikipedia)

We can see from the diagram that we are capturing reds, greens and blues at slightly offset locations. The reds are shifted from the greens, and the blues are shifted from both of them.

This shifted arrangement makes our Bayer arrays more prone to moire because all the colors are out-of-phase with each other. I.e., when we try to interpolate the original colors, the phase differences may cause a false color pattern to appear in the resulting image. (A good RAW converter can try to minimize this artifact.)

In a Foveon sensor, all the colors are captured in-phase with each other, so the sensor is immune from this type of color moire. However, it is still subject to other kinds of aliasing -- in particular to the same fundamental "Nyquist limit" when fine patterns on a subject exceeds the sampling capabilities of the sensor.

kanuck

Just passed the 100,000 shutter mark on my D800e this weekend, and I have had moire issues just 3 times. Not an issue as far as I am concerned, I simply snicker when I read about it being so.