Toothwalker wrote:
The dominant aberration is astigmatism, not field curvature. A first-order correction may be feasible in the lens design, but I don't see how the designer can achieve a high level of correction at all apertures and subject distances.
It's probably the case that Brian was just being extremely thorough and doing whatever correction was possible, however small it might be.
Toothwalker wrote:
But if there are lenses that are designed to counterpoise the astigmatism of the filter stack, is it then assumed that the optical properties of the filter stack are standardized? Or can we expect backwards incompatibility between today's digital lenses and tomorrow's sensors?
Filter stack thickness is probably very similar amongst digital cameras (aside from those made intentially very thin (Leica)), and whilst the refractive index will also differ, it will be a certain ballpark and is always greater than air!
Toothwalker wrote:
The telecentric route seems much safer to me.
For performance definitely, but there's always a tradeoff, in this case size.
philip_pj wrote:
'The reason for an overly pessimistic view is the misconception that only the
resolution limit of the system determines the image quality and that it is identical to the resolution of the weakest link of this chain. This is not the case, though, since the curves are multiplied...' Hubert Nasse.
A x B = C
sensor Mp x lens resolution = final image quality
A rising tide floats all boats and a rising number of pixels 'floats' all lenses in terms of final image quality. This phenomenon is apparent in images from old (lesser) lenses used on the a7r.
A better way to express it is to say that while the best results come from the best lenses and the best (highest MP) sensors (all else being equal) - increasing both A and B; but that higher Mp counts (the 'resolution advantage') increase final imaqe quality with ALL lenses, increasing A but B stays constant. These are two different statements and issues, stemming from the same optical science - the multiplication modality of the transfer function....Show more →
Yes, MTF can be multiplied. However, if A, B, and C stand for resolution, then a more sensible rule of thumb is
1/A + 1/B = 1/C.
This reflects the fact that a further increase of B yields only a small increase in system resolution when B is already much higher than A.
Of course, there is much more to image quality than resolution.
_julian_ wrote:
Alternatively try to substitute that first OLPF layer which is bonded to the sensor carrier with the IR absorption glass which is a far more delicate and non-reversible action. theSuade says it can be removed using a sharp knife and experience!
There was a thread on an astro-photo site about removing CFAs where several sensors were written off during attempted removal of the coverglass - I'd wait till someone else does it first, or send it to lifepixel or MaxMax !
Jeff Kott wrote:
What could possibly be wrong with RF lenses at 50 or above on the A7r? For example, performance of the ZM 50 planar or 50 Summicron should be great and I'd like to see what Zeiss comes up with that will beat a 90 Summicron on the A7r.
Well, Ron has mentioned that the 50 Summilux performs better at the periphery on the M9 compared to the A7, but, even if the full potential of those lenses is realized, the FE 55 looks to be the next step up, nearer to the 50/2 AA or OTUS.
douglasf13 wrote:
Well, Ron has mentioned that the 50 Summilux performs better at the periphery on the m240 compared to the A7, but, even if the full potential of those lenses is realized, the FE 55 looks to be the next step up, nearer to the 50/2 AA or OTUS.
I expect the 50 Cron to perform better than the 50 Lux wide open for obvious reasons and I think it's unlikely we'll see 1.4 aperture native FE lenses anyway.
You were commenting on the best compromise between size and performance. My point is that, although performance of wider RF lenses may be questionable, rangefinder lenses 50 mm and over may be the best combination of size and performance especially the F2 and smaller aperture RF lenses. Note, I'm not saying the absolute performance will necessarily be better, but the RF lenses 50mm and over may be a better size/performance compromise (of course this assumes one doesn't want AF). And again, I think we'll be waiting a long time for the native FE lens that will outperform the 75 or 90 Crons on the A7r (pure speculation on my part).
Edit: I realize in rereading my post that most of us are looking at the results on these cameras and thinking of our own lens collections or lenses we are likely to buy. Personally, I have both a 50 planar and 50 Cron as well as a 90 Cron that I'm itching to get on an A7r and I do think that those lenses may provide the best combination of quality and compactness as compared to native FE lenses.
sebboh wrote:
increasing mp resolution is different from interpolating up an image! that's why more mp make better images than just interpolating smaller images. lenses that exceed the resolution capabilities of 36mp have been made for more than a 100 years. the issue is more a matter of how much contrast the lenses have at spatial frequencies near the limit of the sensor (and the lower contrast is at high frequencies the more you need additional mega pixels to see it). nearly any lens made in the last 50 years will easily outresolve the a7r's sensor over 80% of the frame (statistic i just made up but would be shocked to see disproven), and even the one that can't will still look better in 36"x24" inch prints than they would from a 24mp sensor.
I simply disagree with you that nearly any lens from the last 50 years will out resolve FF 36MP. Certainly there are some but not "nearly any". This is more-so the case at the wide end and FF. If you restrict your 80% claim for frame coverage to maybe 50%, than perhaps. I saw this with wides and FF 24MP's already. There are obvious gains to be had going with higher resolving lenses (if resolution is the goal), particularly on the wide end.
Tariq Gibran wrote:
I simply disagree with you that nearly any lens from the last 50 years will out resolve FF 36MP. Certainly there are some but not "nearly any". This is more-so the case at the wide end and FF. If you restrict your 80% claim for frame coverage to maybe 50%, than perhaps. I saw this with wides and FF 24MP's already. There are obvious gains to be had going with higher resolving lenses (if resolution is the goal), particularly on the wide end.
i suspect your right about wide angles being worse, don't really pay much attention to lenses wider than 35mm. still, having the center 50% look noticeably better and the whole thing upscale smoother sounds worth it to me. there are lots of people who pretty much never put a corner in focus anyway (steve huff? ).
I simply disagree with you that nearly any lens from the last 50 years will out resolve FF 36MP. Certainly there are some but not "nearly any". This is more-so the case at the wide end and FF. If you restrict your 80% claim for frame coverage to maybe 50%, than perhaps. I saw this with wides and FF 24MP's already. There are obvious gains to be had going with higher resolving lenses (if resolution is the goal), particularly on the wide end.
I do not see it that way.
Better is to think that no FF 36mp chip has the capacity to render the image coming from any lens (even a bad one in terms of absolute resolution)
I think if anything, these cameras from Sony have a much greater influence on Canon relative to the Alt community here, than Leica. For years Canon EF was the most adaptable platform for various SLR lines. That Canon has been lagging in sensor development, in terms of features valued in this community (dynamic range, lack of shadow noise, color quality - rather than video or high-ISO performance), has left a user group eagerly looking forward to anything offering improved performance.
If you go to the Canon board, there is a 47+ page post about Canon EF lenses on the Sony cameras. It's not just the 'traditional' Alt community that has taken note. As someone in both camps, as a Leica M and Canon EF shooter, the Sonys interest me in both respects. It might be worth getting the a7R even if only to use with my TS-E collection as the client images I make with these tend to place the greatest technical demands on image/sensor quality. Added bonus if it will work well with at least some of my RF collection.
I'm not sure what edge Leica has lost. If they really wanted to compete in this CSC market, they would have had to release a non-RF camera half the price of the M240. According to their executives at the time of the M240 launch, the camera was first and foremost made for Leica's rangefinder users. A few years ago I too would have thought 'so what,' but now in the system, I realize it's pretty much THE reason for its existence. No matter how quaint or antiquated it might seem to the 'non-believers' it actually works really well in real-world applications if it suits your style of work.
Leica's lenses, particularly their most current ones, work great on their cameras. Should this be any surprise? Fuji's XF lenses work best on their cameras (compared to adapting many RF lenses). As much as we won't like it, it's certainly in Sony's interest to have their and their partner's lenses deliver the best technical results on their cameras. In some respects, they're following in Fuji's footsteps, touting their system's adaptability with other lens systems. While the lenses physically fit, there is no guarantee the optical results will always be satisfactory, as previously seen with RFs on the Fuji platform, which prior to these Sony cameras were touted by some as Leica-killers.... ...Show more →
Ron, I wish to thank you for taking the time to test the various lenses, and making the comparisons with the A7/M9.
Excellent commentary! I feel most M9/M240 users do very much hope that the A7/A7r will be successful, and it would be great to use along a M9/M240. There is a place for both when using Leica M lenses. In the last few years, I have sold all my old Canon L and Zeiss ZE lenses, to opt for M mount lenses as a lightweight system. My intention is to use whatever ergonomic FF platform to use these lenses, whether it is the Monochrome/M9/M240 or A7/A7r.
sebboh wrote:
i suspect your right about wide angles being worse, don't really pay much attention to lenses wider than 35mm. still, having the center 50% look noticeably better and the whole thing upscale smoother sounds worth it to me. there are lots of people who pretty much never put a corner in focus anyway (steve huff? ).
Or those whose perfect corners are attached to unremarkable images,
This whole ethos of: OMG the corner is smeared, forget it! --- is junior high school.
It will be interesting to look at all the contenders and the intricacies, hopefully sans macho blather. OK that's too much to hope for.
I wonder how many sold off larger, very competent, gear to buy into M lenses but not necessarily M digital cameras? Around here, it seems like quite a few if not approaching most. Outside of alt-glass forums are the percentages the same?
In my case I kept and bought new M lenses not for film any more but as small high quality lenses for size and as a hedge against digital camera vendor lock in. Since then I changed my mind about the latter about a year ago when the RX1 showed up.
I got to thinking that sensors getting denser and larger (full frame, figuring on the A7 showing up inevitably) problems with some wide angles seen on APS-C sensors would only be exacerbated. Actually it was the pre-delivery uncertainty about the NEX-7 which first got me thinking about the happy accident nature of M lens support by non-Leica digital cameras. Right or wrong, I concluded that one of my main reasons for holding on to my existing buying M mount lenses was no longer valid and in fact I was more likely to be locked in keeping them than not. I do think that'll be true for some designs and focal lengths, absolutely.
For example, can the value of Zeiss Biogons hold up over the long run, absent some new organic sensor? There aren't many full frame cameras where these can be successfully used today. Lock-in!
I still want a light, small, high IQ system of course; I'm ok if that means adopting new lenses, given I understand why it is that some old friends might not be as useful going forward.
Sony's real edge on Leica is they can deliver the small high IQ system without catering to legacies of the past, bring out new lenses better suited for the high resolution large sensor reality of today and tomorrow, and they can price the lot at a commodity albeit premium commodity levels.
When the dust settles and there's a good understanding of what RF lenses work well, and the native lens line up is more complete, those that once could only turn to Leica for their high IQ full frame drugs will have another dealer to consider.
Given both 50mm and 35mm - classic rangefinder focal lengths - are going to be available on the A7/r cameras out of the gate, and that one looks fantastic and the other small and very good, can there be any doubt that Sony is appealing to that segment of rangefinder buyers who are in it more for the size/IQ ratio than the rangefinder experience itself?
Hmmmm... the note on Flickr says that is how the 25mm CV looks like on film. I think the CV looked quite great on film, wide open, nothing like that shot.
As an example, look at the upper right corner of this photo, the edges of the machine: (wide open on Tri-X)
(The vignetting is post-process, sorry about that- don't have the un-edited handy)
michaelwatkins wrote:
Sony in Australia is gifting their A7/r buyers a coupon in the box redeemable for a free Metabones Nikon, Canon, or M mount adapter. I wonder which adapters go out the door the most?
The one going out the door to me will be the Metabones Canon to NEX adapter, even though I only own two EF mount lenses: the Canon 40/2.8 and the Voigtlander 40/2.
genji wrote:
The one going out the door to me will be the Metabones Canon to NEX adapter, even though I only own two EF mount lenses: the Canon 40/2.8 and the Voigtlander 40/2.
Maybe Sony could add a 24x24 crop mode. Square format that would cater to using RF glass ... kind of an RF meets MF @ FF. Classic RF glass on a classic (reduced, but still square) format.
).
