No - it's slightly better in one region of the 40lppmm test and by an amount I doubt could be seen with a 5D sensor. To describe it as 'clearly superior' is a gross exageration of the facts.
It is clearly superior in terms of the MTF, that is not an exaggeration. One can immediately see a superior difference in the MTF's. Whether you can notice a difference on a 5D might depend on the subject matter, but looking at the entire frame I noticed a difference. That combined with a closer minimum focus, better build quality and half stop of speed made the c/y/ f1.4 lens worth at additional cost to me. If these things aren't important to you and you don't perceive a difference with your camera then perhaps the f1.7 is a better value for you. However, I've seen people in this forum obsess about finer differences than what I observed between these 2 lenses. For me, after comparing them, the f1.4 was the clear and unambiguous choice.
My decision how is whether to keep the c/y 50/1.4 as I now have the ZF 50/1.4 and ZF 50/2.0 Makro. So far, I've noticed less of a difference between the c/y 1.4 and the ZF 1.4 than between the c/y 1.4 and c/y 1.7.
I see such small differences in the 40lppmm graph as inconsequential in practical terms but if you interpret it as 'clearly superior' it just goes to show how subjective lens evaluation can be.
Your mis-interpret or mis-read what I wrote. I said, "clearly superior in terms of the MTF". "Clearly" means "easily seen", and looking at the MTF's the f1.4 is "easily seen" as better. Whether it is a consequential difference, or whether you see that superiority in any particular image with a particular camera, is another story. Overall, I noticed a difference in favor of the 1.4 lens (you'll just have to accept that) and combined with the other advantages of the lens, made it my choice. Subjective my evaluation may be. But the MTF's are objective, and they show the f1.4 to be better (and again, whether you will perceive that difference in any given situation is another matter).
The little f1.7 is fine lens (which I have repeatedly said), I don't mean to insult it. It is not, however, in conceivable, that another lens might be better (especially when the manufacturer of the lenses thinks so).
Well when I look at the MTF charts, what I see at f/5.6 is that the two lenses have nearly identical lines for 10 lppmm and 20 lppmm and the f/1.4 lens has a line that is slightly higher (1 to 5 percent across the frame) at 40 lppmm. Now by any interpretation this is remarkably similar MTF charts. It is true that the f/1.4 lens has a little better numbers, but are those numbers even reliable and if they are reliable are they meaningful. Just because you can see a difference in a chart doesn't mean it is a real difference. I could take a poll in Ohio tomorrow and either Obama or McCain is likely to be ahead. I could draw a graph with the results and it would be easy to see that the one candidate would be ahead, but is that difference reliable--probably not. Is it meaningful this far from the election--almost certainly not. The are essentially in a dead heat. To me these graphs point to a similar situation between these two lenses. Are the differences shown in the figures statistically reliable--I doubt the test have that much precision, but even if they do are they meaningful in terms of performance? To me it seems unlikely that such small MTF differences would translate into meaningful performance differences. Are these differences still there or bigger or reversed at f/8, f/11, etc. We just don't know or at least I haven't seen the data. So with such small differences, tests at only one aperture, and no inferential statistics, I don't think it makes much sense to make a big deal about the small differences in this chart, but others are certainly free to make up their own minds.
Yes, we are making too much of this. And at the risk of carrying this forward, the MTF's for 10 and 20 lppmm of the f1.4 lens are actually actually higher than for the f1.7 lens -- by at 1-5% (or more) across the frame. The 40 lppmm line is actually close to 10% higher in the center (>65% vs <60%) and retains most of that difference across most of the frame (they equalize in the corner). I do believe this difference could be visible.
But the truth is that most standard 50mm lenses perform similarly. The differences in MTF performance across all major manufacturers of 50mm lenses is small -- whether they be Leica, Zeiss, Canon, Nikon, Voigtlander or Pentax. There is a strong reason for this. Nearly all of them are based on the original Zeiss Planar design. Some are a little better wide open, some better around the middle apertures, others retain their sharpness a little better at the smallest apertures, due to slight variations in design and glass types, etc. They are all competent. The differences between the c/y 1.4 and c/y 1.7 are on the scale of the differences between ANY 50mm lens. If you don't think that these differences are visible then you'll be hard pressed to see any MTF differences between ANY two 50mm lenses. They are small differences to be sure and the subject matter might dictate whether you will see it or not.
Quite frankly, standard 50mm lenses are more likely to be differentiated by other performance factors -- flare and ghosting resistance, color rendition, the transition from in-focus to out of focus areas, bokeh, the presentation of depth (3D-ness), etc. Many (not all) Canon lenses suck (IMHO) not because they are not sharp, but because of these other factors.
Returning to the Zeiss c/y 50/1.7, what really makes it a bargain is that it brings typical Zeiss quality in all these areas at a price in the current market that is very low. The Zeiss c/y 50/1.4 is incrementally better (as well as having a better construction, closer minimum focus and being faster), but you might not always see the MTF differences (depends on the camera, subject matter, technique, etc.) but they are there and potentially can be evident in an image. I suspect that they might be more evident on a 1DsIII. Each buyer will decide whether the incremental benefits of the 50/1.4 are worth paying twice or more than the price of the 50/1.7. I decided on the f1.4 lens (the extra $100 or so was insignificant), others decided that the f1.7 met their needs and saved them some money. Again, we are talking about small differences here, and whether one considers them to be significant or not is a subjective and personal assessment.
I think this is really nice comparative test ... check it out http://www.geocities.com/ilprode/TestZ.htm. It seems that 50 f/1.7 performs not only as good as f/1.4 version but the center and corner sharpness at f/2.8, f/4 and f/5.6 is a bit better.
hubsand wrote:
None of the 50mm lenses are are sharp wide open in the f1.0-f1.4 range as the best 85mm lenses, of course.
Why is this I wonder. And, should this be? I mean, is there some design/physics reason for it? (I'm thinking of the gulf of performance between 50 f1.2L vs 85L. I'll be much interested in the wide open part of your testing involving the 50 f1.2L.)
I think you'd need aspherical elements to get a wide aperture 50mm lens to be sharp
wide open, and the traditional double-Gaussian 50mm lens design has no
aspherical elements.
trumpet_guy wrote:
I think you'd need aspherical elements to get a wide aperture 50mm lens to be sharp
wide open, and the traditional double-Gaussian 50mm lens design has no
aspherical elements.
I'm not sure about that. The Leica Summilux-R 50/1.4 E60 is "sharp" wide open, but there are no ASPH elements involved. With "sharp" I mean a nice combo of resolution and contrast.
But... how "sharp" do we need, for different applications?
Agreed. The 50/1.7 is the better lens. Less sample variation (50/1.4 was apparently CZ kit lens for a time and produced like cookies) , sharp as tack, smooth focus, nice bokeh.
Oh - I also have a Mint/Mint- one for sale on B&S (all bias aside) ;>
Thanks Tim, your test confirms most of the sources I've found on the net regarding the CZ 1.7 vs 1.4 (even Zeiss's own database), unless focus was missed in the 1.4's shots, the 1.7 pretty much smokes its f1.4 brother for sharpness, from wide open to f4. And it's a few times cheaper too.
Based on posted images and owner comments, the Leica Summilux-M 50 ASPH appears to be the best fast, standard prime for 135 ever made. That accolade applies to its illumination, sharpness, focus range, color rendition, and bokeh. There are complaints about build quality and price but, hey, that's Leica AG in 2008. I think the army-issue 50/1.4 can be greatly improved, but price will reflect the small market and the exacting manufacture.
I'm happy to sacrifice speed for other qualities (lower cost, lower weight, better linearity). Lenses like the Elmar-M 50/2.8 or CZ MP60C have a distinct appeal.
Some M8 owners may disagree now that they've seen the summilux on digital. Digital has a way of forcing us to re-evaluate what we once took for granted.
While the new ZF 50mm Makro-Planar and the CZ 50/1.4 are two very different price points, the ZF wins that challenge easily. With the CZ stopped down to F2, the ZF wins that challenge easily. And the ZF doesn't have CA issues.
I like the CZ for its size and normal aperture ring, but the ZF is an 800 pound gorilla ready and willing to the kick the snot out of CZ when it comes to sharpness and contrast. In terms of bokeh - that's a different question. Arguing one lens has better bokeh than another has proven futile lately, so all I'll say is - I like the ZF better.
Conner999 wrote:
Agreed. The 50/1.7 is the better lens. Less sample variation (50/1.4 was apparently CZ kit lens for a time and produced like cookies) , sharp as tack, smooth focus, nice bokeh.
Oh - I also have a Mint/Mint- one for sale on B&S (all bias aside) ;>
Thanks Tim, your test confirms most of the sources I've found on the net regarding the CZ 1.7 vs 1.4 (even Zeiss's own database), unless focus was missed in the 1.4's shots, the 1.7 pretty much smokes its f1.4 brother for sharpness, from wide open to f4. And it's a few times cheaper too.
Well, I had both and kept the 1.4. To say the 1.7 smokes the 1.4 is a little bit harsh. I found them to be very similar in performance in the center at wide open and I liked the 1.4 more. Sample variation seems to play a heavy role with the 1.4 as does the thinner DOF ot the 1.4. And prices are also not that far apart between the two, so in my opinion it's a matter of try and error to choose between them.
BTW, according to those charts, there is not much to choose between the 85mm 1.2 and 1.4 for center sharpness which is the main thing for portrait lens, while the 100/2 shows very even performance across the frame that makes it the better landscape lens out of the bunch.
'Smokes' would be harsh - chalk it up, as you say to SV, on the 1.4
Also, as you state, I can't think of any 85mm (a FL typically reserved for portrait lenses) that would make a good landscape lens. Center sharpness, thin DoF and Bokeh
are what you want to judge an 80 or 85 by.
The Contax CZ 85/2.8 Sonnar works fine as a landscape lens.
Of course, it's not your typical wider aperture, bokeh-optimized,
portrait glass either. But I still like it for portrait and landscape, both.
I think the army-issue 50/1.4 can be greatly improved, but price will reflect the small market and the exacting manufacture.