I am a bit skeptical about this. Main reason being that I received 50/2 MP and it was a dud. Should not happen with "100% QA". But there are other reasons as well - for example, why are the cine lenses (and not only Compact Primes) so much more expensive?
I have read that (See below) they do test all the lenses they produce at Cosina. Look at six pictures near the top of the site. I don't know what the threshold for some of the lenses being not acceptable. But the author who has visited the factory states that the each of the lenses hand made is inspected and tested on the optical bench numerous times. Nuance might have been lost in the Google translation.
well if your desires for this lens end up being too great you can always leitax the ZF version for sony A-mount.... right?
edwardkaraa wrote:
Absolutely true.
I said it's getting tiring because we all know it's large and heavy. It's not like it's gonna get any smaller if we keep on repeating it ad nauseam. Now if we can find out why it is so big and heavy then each can judge for himself if the extra IQ is worth the price, size and weight, or not.
I too find it a bit oversized, but that wouldn't have stopped from buying it, if Sony would allow Zeiss to make it in the ZA mount that is
shogo73 wrote:
How do I know which post one may be replying to? I have not figured this out.
Use the "Quote" button at the bottom of the post that you want to reply to. With that there is no room for confusion. If you use the quick reply box, you're not replying to anyone in particular but just adding one more post to the thread.
shogo73 wrote:
Mirek Elsner worte:
Does it include Cosina?
I am a bit skeptical about this. Main reason being that I received 50/2 MP and it was a dud. Should not happen with "100% QA". But there are other reasons as well - for example, why are the cine lenses (and not only Compact Primes) so much more expensive?
I have read that (See below) they do test all the lenses they produce at Cosina. Look at six pictures near the top of the site. I don't know what the threshold for some of the lenses being not acceptable. But the author who has visited the factory states that the each of the lenses hand made is inspected and tested on the optical bench numerous times. Nuance might have been lost in the Google translation.
Edited to include the post I am replying to. How do I know which post one may be replying to? I have not figured this out. ...Show more →
I have to say, these images of Zeiss production at Cosina are very impressive. Reminds me of some of the Leica factory videos and, in fact, makes me want to go out and buy some more Zeiss lenses! Thanks for posting this.
denoir wrote:
That kind of sucks actually. It would have been much better to have one measurement at infinity, one at MFD and one at some typical application distance. There are some lenses like the 50/1.4 Planar that are excellent at larger distances but are completely overcome by spherical aberrations near MFD resulting in a drastic loss of contrast and resolution.
denoir wrote:
I don't see what you disagree about - that it would be good to have MTF charts for multiple distances and apertures?
Yes. Zeiss measures MTF's at infinity, and yes it would be interesting to see MTF as different focus differences. It would nice if some of those BS lens reviewers out there would actually measure MTF's of various lenses on a comparable basis and do it at different distances. Whether it be dpreview's testing or the large photo magazines (large audience, large revenues -- not the clueless, rinky-dink amateurs across the web) rather than inventing round-about, and often inadequate methods to measure and present lens performance info, why don't they just buy a Zeiss lens testing station and give us accurate, repeatable, testing using a common basis across lenses, measuring for different focus differences, measuring MTF's vs. focus distance, etc. I've always thought there is a real gap in lens reviewing/testing. Why doesn't someone just go out and buy Zeiss lens testing station and give consumers a more complete set of information about lens products that no one, including manufacturers provided. How expensive are these stations? Does anyone know? I wonder if I can lease one?
Lotusm50 wrote:
Yes. Zeiss measures MTF's at infinity, and yes it would be interesting to see MTF as different focus differences. It would nice if some of those BS lens reviewers out there would actually measure MTF's of various lenses on a comparable basis and do it at different distances. Whether it be dpreview's testing or the large photo magazines (large audience, large revenues -- not the clueless, rinky-dink amateurs across the web) rather than inventing round-about, and often inadequate methods to measure and present lens performance info, why don't they just buy a Zeiss lens testing station and give us accurate, repeatable, testing using a common basis across lenses, measuring for different focus differences, measuring MTF's vs. focus distance, etc. I've always thought there is a real gap in lens reviewing/testing. Why doesn't someone just go out and buy Zeiss lens testing station and give consumers a more complete set of information about lens products that no one, including manufacturers provided. How expensive are these stations? Does anyone know? I wonder if I can lease one? ...Show more →
Measured vs. simulated MTF curved:
Some time ago I compared many MTF data from Zeiss, Leica and Canon to the photodo measurements and my finding was that if three lenses from these three manufacturers would have the equivalent photodo measurement and rating, the Leica MTF would show about 7% higher values in the curves and the Canon MTF about 14% higher values compared the (always lowest and therefore IMO most realistic) published Zeiss MTF data.
So I always bear that finding in mind trying to compare and evalaute MTF data of these three lens manufacturers.
As for price, I have not been able to find it but I'm guessing it falls into the "If you have to ask then you can't afford it" category.
I do agree however that it should be within the reach of a major photo magazine or review site.
That's the one. I'm going to have to find out how much they cost. There are a couple of manufacturers of similar test benches, but I think Zeiss would be the way to go. ;-).
edwardkaraa wrote:
To be honest, I'm very happy with the ZS 35/2 as a fill gap until Sony replaces the not so well selling 35G with a ZA.
It's unfortunate that the ZS mount lenses are discontinued and that most focal lengths were never made available. David at Leitax is great but, unfortunately, I recently found shipping from Spain to the U.S. apparently is not reliable and never received my adapter ordered back in November from him even though I'm positive he of course sent it. That means I can't adapt my C/Y lenses to Sony, nor any of the ZF lenses. Sigh.
Tariq Gibran wrote:
It's unfortunate that the ZS mount lenses are discontinued and that most focal lengths were never made available. David at Leitax is great but, unfortunately, I recently found shipping from Spain to the U.S. apparently is not reliable and never received my adapter ordered back in November from him even though I'm positive he of course sent it. That means I can't adapt my C/Y lenses to Sony, nor any of the ZF lenses. Sigh.
I don't know about you Tariq, but I'm starting to find this manual aperture stop down thing very annoying. I'm using my ZS lenses only for landscapes or when I know I won't stop down the aperture smaller than f/2.8 as I can reliably focus at this aperture. My ZA lenses are seeing much more action.
edwardkaraa wrote:
I don't know about you Tariq, but I'm starting to find this manual aperture stop down thing very annoying. I'm using my ZS lenses only for landscapes or when I know I won't stop down the aperture smaller than f/2.8 as I can reliably focus at this aperture. My ZA lenses are seeing much more action.
edwardkaraa wrote:
With my humble understanding of MTF, I can already see that Zeiss achieved a great feat with this lens. Already from WO, the extreme corners are as sharp as the center, with very little to no veiling haze, and a slight drop in contrast in the corners. Of course some center sharpness was sacrificed to achieve that but the result is even sharpness across the frame. This a trend in the most recent Zeiss glass (especially the ZA line), probably caused by our obsession of examining extreme corners at 100% magnification :P
Usually f/1.4 lenses have high center sharpness but a dramatic drop as you move away from the center.
This also shows in the comparison with the f/2. The MTF being so close (with a 1 stop advantage) is extraordinary. One would expect an f/2 lens to outperform an f/1.4 at similar apertures but clearly this is not the case.
I do expect this lens to be an equally super performer at its entire aperture range, which will make it as good for low light shooting as for landscapes....Show more →
yes, the new zeiss certainly does extend the flat portions of the mtfs closer to the edges, as most would demand. what interests me is the choices made involving astigmatism correction (this is the part i was referring to with regard to similarity between the two Z* lenses versus the contax) and how that will effect the blur signature. it seems that in order to maintain the flat mtf the new zf shows greater divergence between saggital and tangential curves particularly near the edges. the contax appears to have been designed to have the two curves meet in the center and edge while the new design has been designed to have them meet in the center and at the rule of thirds points (should be better for narrow dof off center compositions) while diverging in the edges. i suspect this will lead to a different renderings of objects only slightly oof on the periphery of the frame. but, as this is only an mtf at infinity i may be reading too much into it.
There are other testbenches to try, CI-systems, TriOptics, Wells, and several German/Austrian manufacturers make manual systems that you can land for a fraction of the price of a complete automated K8 system. I don't have a quote for the K8 since it varies greatly with configuration, and is "on quote only". But yes, if you have to ask....
Grasmuc's findings (on average, about measured MTF vs official graphs) seem to fit quite well with mine.
You have to have a lot more experience and knowledge than me to accurately make predictions about an optical system from just the MTF graphs. You also have to know the base construction pretty well, different correction systems have different "symptoms" for different aberrations.
One thing that's almost impossible to tell from "just" an MTF is how the through-focus behaves, and this is maybe one of the most "characteristic" fingerprints of an optical system.
You can't tell if a certain set of MTF curves become what they are from longitudinal colour, SA or other higher order aberrations until you look at through-focus at separate wavelengths - and this is an extremely work-intensive labour. You need 20-30 different depths at three wavelengths to get a numerical representation of the in>out of focus behaviour for a lens. This is basically a four-dimensional graph, with every point in the graph-space represented by an FFT-intensity of the pointspread for that defocus/colour.
And this is just for the optical center - it says nothing about picture corners...
It's easier to just look at the pictures. The human brain is a wonder of stimuli analysis....
One thing you CAN tell from the MTF is if the bokeh will have a swirly look or not. If sag/mer lines deviate by a lot that usually means that you have different curvature of field for sag/mer detail - resulting in a pictorial rendition that as a whole is perceived as "assymetrical, centered around the optical axis".
And of course, how sharp the picture in a best-case scenario can get - in the plane of focus.
The sharpness is a result of many aberrations combined, but exactly what the ratios of the different aberrations are is almost impossible to say without more information about the optical system. But a better, flatter (over the field) sharpness means a smaller SUM OF ABERRATIONS, which in most cases makes a more distinct and homogenous picture.
theSuede wrote:
The absolute theoretical MTF limit at F/5.6 and 20lp/mm is ~0.90. This means having ZERO light-scatter internally in the lens elements (impossible unless you make all elements out of pure crystalline materials with perfect transmission), having ZERO surface reflection and random dispersion (equally impossible unless you have perfect nanocoatings on all surfaces and surface deltas approaching single molecules of crystal), and ZERO internal light scatter in the lens tubing, mirror housing, film or sensor, and front rim of the tubing.
I would add: and zero aberrations. I find ~0.93 by the way (@ 550 nm).
denoir wrote:
1) Unfortunately MTF curves don't say anything about CA or a host of other optical aberrations.
Au contraire. It's aberrations (plus flare and diffraction) that shape the MTF curves. At f/1.4, you can be sure that the departure of the MTF curves from ideal is mostly due to aberrations. The difficulty is to tell the aberrations apart, and I don't think there is anybody who can do that from a single MTF slice in object space.
Of course, there is always room for educated guesses. If the tangential MTF falls well below the sagital MTF towards the image corners at mid apertures, and if the lens design is asymmetrical, lateral chromatic aberration is the prime suspect.
Yes, that was my point as I clarified in a later post. You can't tell which one of a host of aberrations that cause a drop in contrast. I would however disagree that only aberrations are the only things that form the MTF curves - field curvature is a typical cause as well.
Toothwalker wrote:
Of course, there is always room for educated guesses. If the tangential MTF falls well below the sagital MTF towards the image corners at mid apertures, and if the lens design is asymmetrical, lateral chromatic aberration is the prime suspect.
If we have a series of MTF graphs for two or more f-stops, I believe we can take the educated guessing a bit further, since we know that lateral colour is the only aberration that isn't reduced by stopping down (except for distortion, which doesn't influence the MTF). If all aberrations except lateral CA are reduced by stopping down, but the tangential curves in the field don't improve or improve only slightly upon stopping down, then it's reasonable to assume that lateral chromatic aberration is the culprit.
