It is time to remind people (at least those who don' t understand this) that comparisons among 100% magnification crop samples from cameras with different pixel dimensions (e.g. — number of photo sites if the sensor dimensions are the same) can be misleading.
(The two sample photos at the page linked from the original post illustrate the point.)
- - -
A moment later after a quick look at the articles... For those desiring/needing greater system resolution from a full frame DSLR, the results look quite good. I'm still on track to get one. Leaning perhaps 60:40 toward the "R" model at this point...
My biggest concern with the non-AA cameras is just how hard moire'/color aliasing is to remove when present, and just how intrusive it can be, particularly with lenses that out-resolve the sensor (as many a Nikon user has found!).
That said, given that most of your subjects tend to be 'organic' Dan, I'd be leaning toward the 'R' model if I were in your shoes too.
johnctharp wrote:
My biggest concern with the non-AA cameras is just how hard moire'/color aliasing is to remove when present, and just how intrusive it can be, particularly with lenses that out-resolve the sensor (as many a Nikon user has found!).
That said, given that most of your subjects tend to be 'organic' Dan, I'd be leaning toward the 'R' model if I were in your shoes too.
I already see some moire from the 21/22 MP sensors with AA filters, especially on the birds/hairy mamals which are organic as well. I'd really like to see how the R does in that case.
EB-1 wrote: johnctharp wroteMy biggest concern with the non-AA cameras is just how hard moire'/color aliasing is to remove when present, and just how intrusive it can be, particularly with lenses that out-resolve the sensor (as many a Nikon user has found!).
That said, given that most of your subjects tend to be 'organic' Dan, I'd be leaning toward the 'R' model if I were in your shoes too.I already see some moire from the 21/22 MP sensors with AA filters, especially on the birds/hairy mamals which are organic as well. I'd really like to see how the R does in that case.
EBH
No AA filter, but significantly higher resolution. The situations where you see it will change quite a bit, in many cases, others may not, as we know that it's all related to the frequency of the patterns in the subject versus the resolution of the sensor.
The 36MP sensor in the D810 is pretty close when the 5Dsr is downsampled. The Canon's pixel noise at 6400 is pretty competitive though; more than I would have suspected.
The Canon looks soft next to the 645z. (Lens?) This was certainly a surprise to me, given the similarity in resolution.
For still work, the Olympus looks surprisingly good.
The 5D3 isn't bad by any means. I wonder how it would compare sans AA filter.
Maybe I missed it, but they didn't mention the lenses or apertures used for the tests.
All in all, I can see the allure in upgrading to the 5DS or R if you shoot a Canon body for low ISO work, but I can't see much of a reason to do so if you are shooting an 810 or A7R. However, if you are shooting Canon glass on a A7R, the choice becomes a little more difficult. The 645z is an awesome performer but I'm not sure it's really worth the extra money or weight/lens system.
A 50MP MF (even mini MF) body is always going to be capable of higher system resolution than a 50MP FF body, even with lenses of equal optical quality as measured in the usual lp/mm fashion.
Why doesn't everyone just use digital MF then? We all probably recognize at least some of the reasons why it isn't for everyone despite its demonstrably higher IQ potential:
1. Price
2. When both formats are excellent, sometimes being more excellent won't make a difference that is visible.
3. The larger format also requires larger lenses for the same coverage, producing a very large and heavy system. If you use a lens or two and they aren't too long this may not be an issue. If you rely on long lenses or use a larger number of lenses, it may be.
4. Although strides have been made recently, there certain subjects and kinds of shooting for which MF is not really suitable: wildlife, sports, etc. (I'm not saying it isn't possible, just not ideal for most people.)
MF is great for some people and some kinds of photography, but not for all.
Regarding the comparisons among the FF options (here including the 36MP Sony/Nikon sensor cameras and the 50MP Canon bodies, but also the Olympus option in the article), the image quality differences are likely quickly get into the "how many angels on the head of a pin" territory. I know that some are going to disagree, but the differences between the photographs produced on these cameras are likely to be pretty small — so small that they probably won't be visible in most cases and so small that other functional/operational differences may seem more important to a lot of folks.
Dan, I agree with pretty much everything you've said. I have seen discussions of sensors in terms of "information content captured" as a way of rolling up the sensor area and detail/resolution capability. Differences in total information content will only become apparent when the output medium is no longer limiting - hence the visibility of these pixel count increases will only become apparent when the print is large and can "handle" more information than is being thrown at it. An interesting way of conceptualizing the size of the pinhead
gdanmitchell wrote:
A 50MP MF (even mini MF) body is always going to be capable of higher system resolution than a 50MP FF body, even with lenses of equal optical quality as measured in the usual lp/mm fashion.
Why doesn't everyone just use digital MF then? We all probably recognize at least some of the reasons why it isn't for everyone despite its demonstrably higher IQ potential:
1. Price
2. When both formats are excellent, sometimes being more excellent won't make a difference that is visible.
3. The larger format also requires larger lenses for the same coverage, producing a very large and heavy system. If you use a lens or two and they aren't too long this may not be an issue. If you rely on long lenses or use a larger number of lenses, it may be.
4. Although strides have been made recently, there certain subjects and kinds of shooting for which MF is not really suitable: wildlife, sports, etc. (I'm not saying it isn't possible, just not ideal for most people.)
MF is great for some people and some kinds of photography, but not for all.
Regarding the comparisons among the FF options (here including the 36MP Sony/Nikon sensor cameras and the 50MP Canon bodies, but also the Olympus option in the article), the image quality differences are likely quickly get into the "how many angels on the head of a pin" territory. I know that some are going to disagree, but the differences between the photographs produced on these cameras are likely to be pretty small — so small that they probably won't be visible in most cases and so small that other functional/operational differences may seem more important to a lot of folks.
Oh yeah, totally agree; I was just pointing out the differences that I saw. Though the overall takeaway message for me is that any of these cameras can really give great results. Though people may sway either way if they are into certain niches and have the capital. The law of diminishing returns is as present as ever.
sb in ak wrote: gdanmitchell wroteA 50MP MF (even mini MF) body is always going to be capable of higher system resolution than a 50MP FF body, even with lenses of equal optical quality as measured in the usual lp/mm fashion.
Why doesn't everyone just use digital MF then? We all probably recognize at least some of the reasons why it isn't for everyone despite its demonstrably higher IQ potential:
1. Price
2. When both formats are excellent, sometimes being more excellent won't make a difference that is visible.
3. The larger format also requires larger lenses for the same coverage, producing a very large and heavy system. If you use a lens or two and they aren't too long this may not be an issue. If you rely on long lenses or use a larger number of lenses, it may be.
4. Although strides have been made recently, there certain subjects and kinds of shooting for which MF is not really suitable: wildlife, sports, etc. (I'm not saying it isn't possible, just not ideal for most people.)
MF is great for some people and some kinds of photography, but not for all.
Regarding the comparisons among the FF options (here including the 36MP Sony/Nikon sensor cameras and the 50MP Canon bodies, but also the Olympus option in the article), the image quality differences are likely quickly get into the "how many angels on the head of a pin" territory. I know that some are going to disagree, but the differences between the photographs produced on these cameras are likely to be pretty small — so small that they probably won't be visible in most cases and so small that other functional/operational differences may seem more important to a lot of folks.
DanOh yeah, totally agree; I was just pointing out the differences that I saw. Though the overall takeaway message for me is that any of these cameras can really give great results. Though people may sway either way if they are into certain niches and have the capital. The law of diminishing returns is as present as ever.
By the way, I'm very intrigued by the 645z and I haven't ruled out getting one for my photography. It seems like a really fine camera and a great value for a camera that does what it does. The question marks for me involve the relative functional advantages of the smaller system (which can produce really fine results) versus the somewhat better IQ potential of the 645z in very large prints.
For some time I have considered MF digital —in fact I gave some though the 645d predecessor of the 645z. At one point I had a conversation with a friend who at that time was using only the Mamiya 645 body with the full size MF 80MP Phase One back, in which I mentioned my interest. Keeping in mind that he is a person who used to shoot only LF film and then downsized to MF digital, I was a bit surprised (until I thought about it) when he pointed out that despite the IQ potential of the larger system there would be things that I regularly do in my photography that I really couldn't do with MF, and that these functional issues might make the MF systems less useful to me than I imagined.
He was, of course, correct. Some of my photography could be done nicely with the MF system, but some of it literally could not be done — some of the wildlife photography for example, or some of the photography that I do in the backcountry while traveling on foot. A major issue would be, for me at least, the fact that I often use longer focal lengths.
(Interestingly, he started using the D800e and then the D810 for some of his work during the past two years, and on a recent two-week Antarctica trip, where he photographed and instructed, he took only the D810 system.)