AA filters are a double edged sword. they fix some problems but add their own
A comment was made re the kodak SLR/c and how they can be had for cheap$$ fact is they're hard to find and have relatively good resale.
who buys them? pros or advanced photographers and the ones who have learned to use them do not let them go easily and most are still in professional usage.
I have an SLR/c a 5D a 350d IR and have had a couple other canons.
With the kodak there is an absolute difference in detail and mircocontrast that is hard to miss, better color palatte but that is subjective and using a longer mode it has 2-3 stop of additional DR over the Canons. Not talking pixel peeping but prints.
I attribute it solely to the lack of an AA filter as the only cameras that give the same superior results are the leica (no AA) and the medium format backs, most of which come without or are user optional and can be used or not. It is worth noting that on the old gailbraith board where there were more pros with medium format backs relatively few used the AA filter as they could see side by side the negative impact.
i have had several of my clients go though a pile of prints 13x19 up to 24x30 and with the exception of portraiture they can easily sort them by 5D v Kodak SLR/c with 85% accuracy.
I WISH that were not the case as the Kodak is a bit quirky, but it is more like using a medium format or large format camera in terms of care in set up and usage and does not have stellar high ISO performance and is slow (of course it is a couple generations old now!).
I'll be selling my 5D as it doesn't cut the mustard for me. If Canon goes no AA i'll buy it. if not i'll wait until someone comes up with a camera that has the IQ of my Kodak.
fwiw i am looking at the leica but i think it is not an improvement over the kodak, just a bit more modern (and a lot more expensive!!)
I'm very interested to hear your comments on the SLR/c's noise and moire. I have a feeling that these two things are a bit exagerated by many and can be dealt with fairly well in software. How well does Neat Image or Noise Ninja clean up the images? What is the best solution you've found for moire?
I'm aiming for a 5D, but after reading your comments and looking at some sample photos over at pbase.com, I am willing to give the SLR/c serious consideration.
Sorry about the cheap price reference. I am on the other side of the isle and have the SLR/n. I have seen them sell for as low as $1500 and just assumed that the SLR/c were selling for the same.
I think you summed it up very well about the contrast and color palatte of the Kodak. Many people dismissed it for the poor ISO range and less than top of the line bodies they were put into, however as I said earlier, when used properly the camera is more than a stellar performer.
I don't know anything about the new Leica other than what I have read about it. I find it interesting that people are giving serious discussion to camera design and not having the AA filter. I do realize that the Leica is using the Kodak sensor vs the Fill Factory sensor of the /n or /c series of cameras. Whether or not people will be willing to live with the advantages of not having a AA filter vs the limitations is yet to be seen. The biggest hurdle will be learning how to use a camera that doesn't have an ISO rating 10 million.
losloslos wrote:
Re: gdeliz2/Alex's comments, they're both right (if added together). There is an analog to diffraction-in-audio: if an optic element was able to bend all frequencies of light to the same angle, it would be free of diffraction in a pathological sense. Amplifiers generally cannot amplify all frequencies exactly the same, hence distortion.
G.
Just for the sake of clarity... If an optic element is not able to bend all frequencies of light to the same angle it is called chromatic aberration (CA) which has nothing to do with diffraction. Diffraction happens even if we use monochramaric light like a laser beam.
BubbaJon wrote:
This is my guess why the Leica has no AA filter - the lenses are sharp and error free to the point that the resolved image is mostly correct - in their estimation anyway.
Leica is well known for engineering spherical aberrations into their lenses (or rather deliberately not correcting them), specifically because this type of lens error is what produces attractive bokeh. It comes at the expense of not being quite as sharp, especially wide open.
Moire, this problem is real and all cameras without an AA filter suffer from it and many with an AA filter can experience it.
It basically occurs near the nyquist frequency on repetitive elements e.g. fine patterns in textiles, a screen in a photo.
It rarely occurs in nature and there are tools for it. i rarely encounter it, in fact only a couple times in the past 2+ years, some shoot subjects more troublesome, on the DP kodak board there are processes covered for it.
Re Noise, also a real issue, primarily in low light situations. I do encounter this. there are two ways to deal with it, the built in longer exposure mode that manages noise, or jacking up the black point just a bit, and the newer noise programs work quite well e.g. noise ninja. except for my first couple low light nature shoots i have learned to expose properly (they tolerate over exposure more than other cameras) or utilize the longer mode (need a tripod)
Also fwiw my first kodak needed service and the original estimate was that it needed a sensor so my insurance company paid for a replacement ( i got the 5D ). kodak reassessed the situation and repaired it so i sold it hoping a full frame canon would do the trick ~ same pixel count, canon is a good company.
after spending a couple months with the 5D i think it is a fine camera incredibly flexible and user friendly but ultimately gives images only 80% the quality of the kodak
I am retired from professional photography and only do odd jobs now. were i full time i'd keep both as the canon will do things the kodak can't. However for my personal work it doesn't make it. I shoot mostly nature, landscape flowers and while i mostly print 16x20 or smaller the kodak images are better after 13x19 and bigger than 16x20 there is no comparison. I print to 24x30 with the Kodak.
Net i bought another kodak SLR/c.
Gary, based on my time on DP and the kodak board the SLR/c seemed to do a better job with some of the problems, can't say why but i never had many of the problems the nikon based folks had ...glass?? who knows.
SLR/c s seem to command better prices and they're all falling and are tied to shutter counts.
I figure unless i break it i'm good for 1-2 more generations to see what happens.
nma wrote " I suspect that in the best case without the aa filter, there is always some gritty haze that we may perceive as noise."
That is an iteresting observation. If this aliasing noise is additive with sensor noise it may explain the excessive noise at higher ISO's delivered by the DMR, Kodak SLR/c,and the various MF backs that have no AA filter.
re the post by gdeliz2 including text by nma i make a couple observations. first of all i am an ee and plasma physics major so in theory understand all said, however once you "blur" data it is lost and irrecoverable and the best you can ever get is an "assumption" of what the point of origin was.
i believe this is where the detail, microcontrast etc advantage comes from.
the difference between beef stew and the individuals, roast beef, taters etc.
the difference between the "plastic" look so often attributed to digital, blurring then software analysis equalizing the mix, and the results from non AA filtered cameras with beter separation and detail
jjlphoto wrote:
It is permanently bonded to the sensor.
That is what I have always thought. However I've come across three suggestions to the contrary:
1. That dust / cleaning fluid can get between the filter and sensor.
2. That Canon can clean said dust or fluid.
3. That cameras modified for IR photography have the AA filter removed.
The 560 and 520 (and twin brother Canon D2000) too. Its really an education side-by-side to remove the filter and see how it works. Literally I have been able to squeeze decent images for all purposes except large print from a 2MP camera, because I tested the AA first week I had it, and I tossed it.
BUT, I was doing focused photography 90%,...bands, in darkness, not much in the way of patterns that would trip it up.
Believe it or not, even a MF back will moire...it'll actually moire easier! I rented an H1 for the hell of it and was getting moire off of clothing...yeah, the lens (80) was pretty damned good. Nice cam, but, not for me at those prices.
I'm still cruising along with "junk" and loving it.
G.
p.s. Oh! I traded 10 hours of bowling with a local alley for some shots. I shot it with the 1D, and, didn't like it, whipped out that 50$ 630 and some film and shot it, and it came out beautiful, and wider !!!! Lotta megapixels in that there film.
TeamSK jay wrote:
That is what I have always thought. However I've come across three suggestions to the contrary:
1. That dust / cleaning fluid can get between the filter and sensor.
2. That Canon can clean said dust or fluid.
3. That cameras modified for IR photography have the AA filter removed.
OK- I wonder if I read that is was not able to be done by the user or something like that?
Often you can get away without AA. Like other things in life, you take the risk - no insurance, no condom etc. , you brag about your smartness until you run out of yout luck
I do see jagged diagonal lines from the images of the no-AA cameras. DPR reports on SD10 shows that Aliasing impact is much more severe than just Moire.
Usually the user group for the no-AA cameras are very passionate and loyal to their brand. They are extremely forgiving on the flaws of their equipments. On the other hand, Canon and Nikon cannot take the risk, with their huge user-base.
My interest would be to make the modification to a body dedicated to landscape photography. From my understanding the issue of aliasing and moire are more likely to appear when a man made pattern is being photographed. But in the end I'd gladly accept a few screwed up images for a significant increase in clarity across the board.
I just read the SD10 review at DPR again and found nothing to indicate a negative impact from the lack of an AA filter. To the contrary, I found only glowing accolades:
p.14 "Anyone who is used to viewing 'traditional' digital camera images at 100% will immediately see the improvement in sharpness and resolution delivered by the X3 sensor, it can take a little getting used to as there is far less aliasing between contrasting elements. Below are two examples of 'Single Pixel Resolution', very fine details which measure just a pixel across are rendered faithfully."
P.18 "(Thanks) to a lack of a color filter the X3 sensor doesn't suffer from color moire at the limits of resolution.
Because the X3 sensor doesn't employ an anti-alias (low pass / blur) filter it continues to deliver detail past Nyquist. Taking the vertical resolution bar as an example we can count all nine lines up to our 'absolute resolution' measurement of around 1550 LPH, after this point lines begin to merge and by 2000 LPH we can count five obviously combined lines. In a real image this could be the detail of leaves on a distant tree or bricks on a distant wall. At the time of my SD9 review there was much debate over the 'validity' of this extra detail. My opinion on this matter is that this detail is at least of photographic merit, it is the correct color and represents detail that the human eye although not able to distinguish perfectly would also see as a broken texture, it's certainly better than the blurred area we would get from the anti-alias filter of a Bayer sensor camera."
TeamSK,
Please, re-review page-23 of SD9 review and page-18 of the SD10 review.
Basically, they show the same thing. As the line spacings get denser, the lines SHOULD be merged into a just solid gray bar. Instead, the AA-less sensor creates FALSE images.
