The Canon 200 f/1.8 is on there twice? Where did these come from, out of curiosity? Alf Beharie wrote:
Grade: 4.8 35mm/AF Canon EF 200/1,8L USM
....
Grade: 4.8 35mm/AF Canon EF 200/1,8L USM
One copy of a cheaper Zeiss lens against one copy of a Canon lens does not prove anything to me about brand superiority. I have an FL 55/1.2 converted to EF mount. It is a great lens, no doubt. But it does not make me want to try out more old Canon lenses nor ignore all the new Zeiss lenses coming out. How old are those Photodo ratings anyways? How many magnificant lenses have been made since that list was produced?
gasrocks wrote:
One copy of a cheaper Zeiss lens against one copy of a Canon lens does not prove anything to me about brand superiority. I have an FL 55/1.2 converted to EF mount. It is a great lens, no doubt. But it does not make me want to try out more old Canon lenses nor ignore all the new Zeiss lenses coming out. How old are those Photodo ratings anyways? How many magnificant lenses have been made since that list was produced?
Lotusm50 wrote:
You lost me with the word "exact". While these 3 Zeiss lenses share a similar design, they are not exact. The 50/1.7 being the most "different" of the 3, not to mention the cheapest. The ZF50 is clearly descended from the c/y 50, but they are not exact. The ZF is an update of the older design, and the differences in the MTF's confirm that. There are almost certainly differences in glass formula as well as small tweaks to the geometry of the elements. That said, I think that Zeiss should have put more effort in the the ZF 50/1.4 to more effectively distinguish it. While the c/y 50/1.4 might have been a category leading lens 25 years ago, the rest of the industry is closer today. I think there are a lot of people that would have preferred to see a step-out new design for the ZF50/1.4 rather than a minor updating tweak to the old c/y design.
Finally, on this comparison, I'm not so sure that a top-of-the line, exotic, very expensive (in its day) Canon lens is the likely competitor for this entry-level, cut-rate, corner-cutting, inexpensive, basic Zeiss 50/1.7 lens. I would have expect the Canon to be better. Hell, it cost a lot more. Further, these one number average weighted scores for lenses are highly suspect and can be misleading depending on how the lenses are actually used. Quite clearly, the Canon could be sharper in the area of the frame you examined while garnering a lower overall score when the rest of the frame is considered.
Alf Beharie wrote:
But the more expensive Zeiss 50mm f1.4 Planar T* has a lower MTF rating than the "cheap" 50mm f1.7 Planar T*.
According to this MTF list of lenses with MTF ratings of 4.1 or higher the Zeiss 50mm f1.7 Planar is one of the sharpest 50mm lens in the whole list, equal only to the Leica M Summicron-M 50mm f2.0, Pentax AF SMC-F 50mm f2.8 macro and the Pentax AF SMC-F 50mm f1.4.
Grade: 4.8 35mm/AF Canon EF 200/1,8L USM
Grade: 4.7 35mm/AF Contax G Planar 45/2,0 ... {snip}
I would be supprised to find my FL 55mm f1.2 has a higher rating than 4.6 but its certainly possible going by my comparison pics....Show more →
First, taking these old, amateur measurements as gospel is probably a mistake.
Second, the difference between a 4.6 for the 50/1.7 and 4.5 for the 50/1.4 is not significant given a reasonable standard error of measurement (and I would assume that the standard error here is quite large). Further, you need to look at what specific numbers went into the calculation for each and whether it was an apples-to-apples comparison ( for example, on these 2 Zeiss lenses the averages are made using MTF's at different apertures for each lens).
Third, there is no way to confirm that the sample used is accurately representative (i.e., to spec, not damaged or out of alignment in some way, etc.)
Fourth, the weighted averaging used to create a single number for each lens is arbitrary and potentially biased, and may not represent the value of aspects of performance that you would assign. For example, is a lens that is great in the center and lousy in the corner better than a lens that is average across the frame? Further, what about the shape of the MTF curve between the 2 measured points is important. You could have 2 lenses with equal scores in the center and corner averaging to the same score, yet in the space between these 2 points one is significantly superior to the other. This rating "system" completely misses this. No lens MTF is completely linear from center to corner yet this is what is implicitly assumed in these "averages".
I put no faith in these rankings to discern fine or small differences between lenses. Surely, there was a large effort put in to assembling a list like this, but ultimately it is only of very modest real value.
Alf Beharie wrote:
But the more expensive Zeiss 50mm f1.4 Planar T* has a lower MTF rating than the "cheap" 50mm f1.7 Planar T*.
According to this MTF list of lenses with MTF ratings of 4.1 or higher the Zeiss 50mm f1.7 Planar is one of the sharpest 50mm lens in the whole list, equal only to the Leica M Summicron-M 50mm f2.0, Pentax AF SMC-F 50mm f2.8 macro and the Pentax AF SMC-F 50mm f1.4.
MTF is measured in graphs not absolute ratings. Zeiss MTF show a clear advantage of the 50/1.4 in the center at f/5.6. Photodo averaging system tends to favour lenses with smaller maximum apertures.
I purchased a Contax 50 1.7 because of ratings and have always been unhappy. My cheapo Canon 50 1.8 mk11 slates it would you believe!
I did a brick wall test Canon 50 1.8 mk2, Contax 50 1.7, Voigt 40 Ultron, Nikon 55 2.8 micro and Leica 50 f2 Sumi (latest) and the Leica beat them all centre and corners.
wolfloid wrote:
Contax Planar T* 85/1,4 is on there 3x and many others are doubled up - any reason?
These looks like the Photodo ratings.
Whoops, that was caused by some dodgy copying and pasting...I have edited the list so it should be ok now.
The list was sourced from this site I've known about for years:
Lotusm50 wrote:
First, taking these old, amateur measurements as gospel is probably a mistake.
Second, the difference between a 4.6 for the 50/1.7 and 4.5 for the 50/1.4 is not significant given a reasonable standard error of measurement (and I would assume that the standard error here is quite large). Further, you need to look at what specific numbers went into the calculation for each and whether it was an apples-to-apples comparison ( for example, on these 2 Zeiss lenses the averages are made using MTF's at different apertures for each lens).
Third, there is no way to confirm that the sample used is accurately representative (i.e., to spec, not damaged or out of alignment in some way, etc.)
Fourth, the weighted averaging used to create a single number for each lens is arbitrary and potentially biased, and may not represent the value of aspects of performance that you would assign. For example, is a lens that is great in the center and lousy in the corner better than a lens that is average across the frame? Further, what about the shape of the MTF curve between the 2 measured points is important. You could have 2 lenses with equal scores in the center and corner averaging to the same score, yet in the space between these 2 points one is significantly superior to the other. This rating "system" completely misses this. No lens MTF is completely linear from center to corner yet this is what is implicitly assumed in these "averages".
I put no faith in these rankings to discern fine or small differences between lenses. Surely, there was a large effort put in to assembling a list like this, but ultimately it is only of very modest real value.
Hmm. Somebody must be wrong. Tests are all over the place. Just confirms how unreliable and suspect they truly are.
I noticed several other conflicting ratings but thats understandable as the various lists on that site are compiled from many different sources and each tester will come to his or her own conclusions due to how they actually tested the lenses...As each tester seems to use their own test standards instead of a standard universal testing methodology then lists like these should really always be treated as a rough guide and nothing more.
Alf Beharie wrote:
To ensure the focus was spot on with each lens I drew a letter A on the back of fridge with a fine permanent marker and focused on that.
Sorry but that doesn't "ensure" anything.
Also your sample sizes for both lenses is too small to draw a meaningful conclusion for any lenses other than the ones you own. You may have the world's best example of the Canon, or worst example of the Zeiss.
I don't see why you are being so hard on the man. The thread title is a bit sensational, but the content of his posts seem to say that the old Canon is sharper for him in the center than a 'highly rated' Zeiss lens. Whether or not those ratings are super trustworthy is a separate matter. And of course it's his two lenses. When was the last time any of these tests used 20 lenses of each make and model to ensure proper statistics? It's always about the lenses the person tests, and usually a good helping of the biases of that tester too.
I think the take home message is that some lenses are really sharp in the center but might be dogs in the corners. Others might be a hair less sharp in the center but retain that sharpness out to the edge. Hell, if we measured a lensbaby by a center crop, it would be a bit misleading...
That one copy of one of the cheapest lenses Zeiss ever made a long time ago is less sharp in the center than one copy of a professional-grade lens Canon made also a long time ago is fine by me. But the OP's title is so sensational, based on such limited evidence, as to make a British tabloid proud.
philber wrote:
That one copy of one of the cheapest lenses Zeiss ever made a long time ago is less sharp in the center than one copy of a professional-grade lens Canon made also a long time ago is fine by me. But the OP's title is so sensational, based on such limited evidence, as to make a British tabloid proud.
You may have failed to notice that I placed a question mark at the end of the title, making it far less sensational as its more a question than a statement.
Anyway, I did'nt say which crop was which and noone got it right so now I'll give you all the answer:
crop 1 is the Zeiss 50/1.7 at f2.8, crop 2 is the Zeiss 50/1.7 at f5.6, crop 3 is the FL 55mm/1.2 at f2.8 and crop 4 is FL 55mm/1.2 at f5.6.
Here are the original FL crops (The 1st is 6 pixels wider and the 2nd is 11 pixels wider):
tmessenger wrote:
I converted a FL 55/1.2 and it is a great lens no doubt, the copy I had even at f2 was vary sharp.
tm
I'd like to compare it against the Rokkor 58mm f1.2 as its the lens with the most similar performance to the FL 55mm f1.2...Rokkors just cost too much now though so that'll probably never happen.
