edwardkaraa wrote:
Could very well be, but look at the ZM on the M240. It is excellent, small, and not expensive. The FE 35 is also very good, so it is doable IMHO.
Now if future sensors will feature the same glass thickness, and Zeiss were too lazy to make a better lens, I don't think they will sell a lot of these Loxia.
Ed: though I agree with your caveats on the Loxia line I do think Zeiss are fully aware of the design constraints. Even if they did _just_ tweak them a little to correct for thicker sensor glass I am sure it was a deliberate decision. Perhaps the tweak is all what is needed to shift the parameters enough to accommodate for a specific glass thickness interval which would mean that even if the tweak is "near perfect" on the current or some other thickness the correction would still yield significantly better results within a range of thicknesses.
The ZM lenses which we both enjoy and like are excellent lens designs but not designed with the current sensor technology in mind. This is all too apparent at the moment. The Loxia 50/2 seems to deliver in spades and I am sure they are working hard on a wide angle solution that will not disappoint.
I am usually the one playing the cynic cello when it comes to wishful thinking (and dare I say, often rightly so) but I think Zeiss are on the right track. Will it be perfect? Hmm, I dunno. I think they will hit a home run but they might not hit the ball out of the park.
Hi Henrik! Yes, of course, but in my point of view, the Loxia 35 has to perform on the A7 as well as the ZM 35 does on the M240. It doesn't really have to, as I'm sure many here will be happy with an incremental improvement over the ZM on A7, but I personally would find it disappointing from Zeiss. I still believe in the conspiracy theory that future A8/9 will have a thinner sensor stack though This said, I have to point out that he lens I'm really waiting for is the ZM 1.4/35
Not sure I believe that, as it will most likely impact the FE 16-35 performance on future cameras. I just can't see them making a huge change that impacts the current lenses. Unless you think they'll be unaffected by the change.
carstenw wrote:
I have read that before, although the Update section was new. I don't see anything which specifically addresses the sharpness of the Loxia 35 in the corners, unless those unmarked crops are corner crops. The ZM is clearly not good on the Sonys.
these are corner crops..also the article clearly states , that these lenses are calculated for a 2mm filter plate on top of sensor. or in other words - these lenses will no longer really work without 2mm filter plate.
tsdevine wrote:
Not sure I believe that, as it will most likely impact the FE 16-35 performance on future cameras. I just can't see them making a huge change that impacts the current lenses. Unless you think they'll be unaffected by the change.
-Tim
I don't think such lenses will be affected by a thinner glass cover. If anything, I would expect a better performance in the corners.
Murxster wrote:
these are corner crops..also the article clearly states , that these lenses are calculated for a 2mm filter plate on top of sensor. or in other words - these lenses will no longer really work without 2mm filter plate.
I don't believe Sony will go with a glass cover as thin as the M 0.8mm or the 1Ds2 1mm. But if they use just 1.5mm instead of 2mm, which sounds very reasonable to me, this will do miracles to a lot glass, including some of Sony's own FE.
edwardkaraa wrote:
I don't think such lenses will be affected by a thinner glass cover. If anything, I would expect a better performance in the corners.
It's a pity that the Loxia 35 can't be used ( without 'main surgery') on a Leica M. It would be very interesting to compare the performances of the lens on a Sony A7 vs the M240.
edwardkaraa wrote:
Hi Henrik! Yes, of course, but in my point of view, the Loxia 35 has to perform on the A7 as well as the ZM 35 does on the M240. It doesn't really have to, as I'm sure many here will be happy with an incremental improvement over the ZM on A7, but I personally would find it disappointing from Zeiss. I still believe in the conspiracy theory that future A8/9 will have a thinner sensor stack though This said, I have to point out that he lens I'm really waiting for is the ZM 1.4/35
I don't think it is really necessary for the Loxia to perform at the level of the ZM on the Leicas, but at least in the case of the 35 I am optimistic that it will, or at least that it will come close. Regardless, from the results so far it isn't a question of an incremental improvement, but of a substantial improvement...
Murxster wrote:
these are corner crops..also the article clearly states , that these lenses are calculated for a 2mm filter plate on top of sensor. or in other words - these lenses will no longer really work without 2mm filter plate.
I thought the filter plate was 2.5mm. If true, that would leave some room for a filter diet. In any case, I don't think that thinner filter stacks are detrimental in the way that thick ones are.
Artur, Carsten, I wish the reviewer actually could have taken the same shot with a ZM 35 on the M240. That would have been very telling. I am very confident the Loxia, while undeniably a major improvement, is still quite weaker in the corners than the ZM/M combination.
Murxster wrote:
these are corner crops..also the article clearly states , that these lenses are calculated for a 2mm filter plate on top of sensor. or in other words - these lenses will no longer really work without 2mm filter plate.
I don't think that is the case. Reducing the sensor stack thickness for a lens optimized for say a 2mm plate to 1mm should not result in deterioration. Roger Cicala touched on this in the responses that followed below.
"We can’t ‘see’ much of a difference between 1 and 2 mm when we test lenses, so I doubt that kind of reduction would have a major effect."
carstenw wrote:
In any case, I don't think that thinner filter stacks are detrimental in the way that thick ones are.
The filter stack is there for a good reason. Whether the advantages outweigh the disadvantages depends on use and taste. I would prefer a thick stack with dedicated (i.e., corrected for the filter glass) lenses to a very thin stack with classical lenses.
The depressing part of all this, is the (growing) diversity of stack thicknesses among camera manufacturers and the corresponding incompatibility of lenses. With the increasing number of dedicated lenses, there is also no way back for manufacturers to reduce the filter stack thickness in the future, unless they don't mind upsetting existing customers.
Tariq Gibran wrote:
I don't think that is the case. Reducing the sensor stack thickness for a lens optimized for say a 2mm plate to 1mm should not result in deterioration. Roger Cicala touched on this in the responses that followed below.
"We can’t ‘see’ much of a difference between 1 and 2 mm when we test lenses, so I doubt that kind of reduction would have a major effect."
I think that, especially for lenses where thick filter compatibility has been achieved by a rather more tele centric designs, there should be no discernible difference. But I'm sure there must be some degradation if said compatibility has been achieved by creating a negative astigmatism to neutralize the filter effect.
However from what I've seen from all mirrorless manufacturers, method 1 is used, and I can't say I know of any lens design that uses method 2.
Toothwalker wrote:
The filter stack is there for a good reason. Whether the advantages outweigh the disadvantages depends on use and taste. I would prefer a thick stack with dedicated (i.e., corrected for the filter glass) lenses to a very thin stack with classical lenses.
The depressing part of all this, is the (growing) diversity of stack thicknesses among camera manufacturers and the corresponding incompatibility of lenses. With the increasing number of dedicated lenses, there is also no way back for manufacturers to reduce the filter stack thickness in the future, unless they don't mind upsetting existing customers
Why, what happens to a lens when it is used on a too-thin filter stack? I would have thought nothing.
I am also no fan of too-thin filter stacks. I don't know if you recall, but I left the Leica M system when the M9 had traces of the same IR sensitivity as the M8. I also bought the D800 rather than the D800E. However, 2.5mm is really thicker than needed. I would think they could at least go to 2 or 1.5mm without ill effects.
edwardkaraa wrote:
Artur, Carsten, I wish the reviewer actually could have taken the same shot with a ZM 35 on the M240. That would have been very telling. I am very confident the Loxia, while undeniably a major improvement, is still quite weaker in the corners than the ZM/M combination.
Yes, me too, although there will surely be more detailed tests soon.
I've got a question, even if the corners are acceptable, is that not just the very first stage? We still need to see whether the rendering and drawing style make this lens worth that extra money over the numerous and legacy 35mm options do we not? As for the 50mm, it's not that hard to find a good 50mm, actually not that hard to find a faster one for that matter. I keep thinking that the corner sharpness is only just one tick in the questionnaire about these rather expensive lenses.
Yes, of course, you are right. However, the ZM35 generally had quite nice rendering, except for some tricky boke and the corners. The Loxia appears to improve both areas, so the hopes are high
My take on the whole issue is that lenses that are designed to work well with a 2.5mm filter stack /glass will not work worse on thinner top glass/filters though presumably they would not work as well on significantly thicker filters. I would guess that they would still work better than lenses that aren't tweaked. Where the line is drawn I am not competent to even guess.
I don't think there is a too obvious cut off point within tenths of millimeters, probably more of a sliding scale of deterioration with varying filter thicknesses and any lens that is designed to combat those adverse effects will "always" work better than a similar focal length that isn't designed with that in mind, i.e. film lenses or "older" digital lenses.
I would not be to surprised if a Sony A8/A9 comes out with thinner filters. I don't really trust Sony to stick with any sort of compliance throughout the incarnations of the FE mount cameras. I trust Zeiss can make the best of the situation but if Sony changes the premises all the time it will be hard to keep up and deliver a lens line that will perform consistently over time. That is something I would worry about in the long run.
PS
Hi Ed! Looong time no see. Might pop over to BKK this winter.
sebboh wrote:
i have seen similar SA weirdness in wide open m9 examples when looking at parts of the frame that are just a little out of focus.
Yes, and saying "not at all" is quite a bit misleading I would think. "Not as much" is more like it IMO. The difference is there, but it's definitely not all or nothing.
The same goes for the native Leica lens lineup, people often point to the wonderful corners and CA free rendering "when used on the original Leica cameras". The problem is that when you actually do the tests, the Leica bodies only show a bit less of the same, not NONE of the same.
The main problem with modern lens line-ups and cover-plate thickness(-es) isn't really that dire. The main problem (again IMO!) isn't the plates in themselves, but the poor quality of the execution.
Unfortunately Sony skimp quite badly on the quality of the plate materials, and the same goes for quite a lot of cameras today. I often see totally non-AR treated surfaces and badly implemented cementations/laminations. This is a larger problem than the variation that exist in thicknesses (for extremely large aperture lenses, or lenses with exit pupils close to the sensor).
Reflections in material boundaries increase with angle - light hitting a refraction index change (a material boundary) head on has a lot less reflection than light hitting it from an angle. Since angles at f/2 and f/1.4 are pretty steep, even in the center of the image, you lose quite a lot of contrast on the 3-5lp/mm levels and on up.
And looking at the ZM 35/2 corner on the 3D-kraft example shot isn't really that different from what I'm used to seeing from the ZM on other sensors, neither an M240 or a perfect test sensor (my 0.4mm plated 2µm BSI video resolution test sensors) give much better results in the corners with that lens. It's BETTER, but not good.
If the thickness of the layers had a severe impact on "normal" aberrations at f/2, the center would be totally unsharp too. I'd expect a large scale contrast loss and asymmetry in the corners if plate thickness mismatch was the main problem, not "pure" unsharpness.
