dcmiller wrote:
But perhaps valid for Canon users, as husband has several adapters. As a Canon user, I would not expect to get better results than husbands.
Bruceconnelly has apparently already reported better results. Perhaps you might, too.
dcmiller wrote:
As Nikon will go full frame, If I were a Nikon user I would want to see a test on film.
"Will" is pretty strong. Do you really know this to be true, or is it speculation (regardless of how much logic helped form it)? I agree, however, that a test on film would be appropriate, interesting, and potentially revealing.
dcmiller wrote:
But I expect this lens is representative of what Zeiss is building at this price point. This lens was sent by Zeiss for review. Should I expect to get a better lens at retail?
If Zeiss actually selected that particular lens after testing it to be sure that it was representative, and the test adequately represents the performance of the lens they selected and sent, then all is lost. Everything we knew and trusted about Zeiss must have been thrown out the window. Since I don't believe that to be true, I suspect that your mileage might vary with a lens you get at retail.
The adapter placement issue could be rearing its ugly head again. If this lens is like the old CY28/2.8 where all the elements are locked together and move back and forth in unison to focus, adapter variation is a moot issue as the helicoid does all the work, but if their ZF25 relies on internal elements moving separately from other elements, then exact placement is critical to its performance. A lens mounting error of a mere .05mm either way could throw it out of proper performace.
The photozone review seem to suggest it does indeed use internal focusing mechanism. It says that the barrel does not change length when focusing. Although that might explain the sub-par corners when adapted to Canon, it does not explain the corners when used on a Nikon D200.
dcmiller wrote:
"Bruceconnelly has apparently already reported better results. Perhaps you might, too."
To quote Mr. Connelly,
I'd like to state very clearly that my Zeiss 24mm ZF on its Chineese adapter clearly outresolves my 28mm f2.8 Zeiss for Contax on the same brand of adapter and also every other lens I have now or have had (including the 24L) at the focal length, on full frame and at f8-f11.
The ZF I have is every bit as good as the older in every way and I have sent Mark a complete RAW file to show that. I also do not believe the adapter has anything to do with it given the design and that the adapter is flat."
This clearly suggests better results than obtained in the 16-9.net test.
The advantage of direct testing is that it's not dependent on existing belief systems. Hopefully Husband can find some crappy Canon primes so as to not upset the alternative apple cart too much.
dcmiller wrote:
The advantage of direct testing is that it's not dependent on existing belief systems. Hopefully Husband can find some crappy Canon primes so as to not upset the alternative apple cart too much.
I wonder who this "Husband" guy is, and if he's aware that there's a "hubsand" guy running around on this forum.
GEEZ!!! I'm tired of threads as this! Why can't you guys just go out, start taking photos and enjoy what you have!? Canon or Zeiss, they're both nice as long as there are no pixel peepers
It's the same analogy when comparing 2 great looking Cheerleaders ~ yes both of them looks breathtaking from a distance, but you guys wanted to look closer and check for pimples or blemishes ~ the whole campus can tell you varying reports about the Cheerleader's attitudes and approach to life BUT, my point is, the only way to get to know them is by asking them out and spend the whole year with one of them
shiwan wrote:
I wonder who this "Husband" guy is, and if he's aware that there's a "hubsand" guy running around on this forum.
Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.
As far as I know, no-one has suggested that Zeiss has fabricated their MTF data – so perhaps we can lay that to rest. Seeking explanations for the relatively disappointing 16:9 test, that leaves us with three options in the absence of a reply from Zeiss:
1. Sample variation.
Is this the same explanation we resort to in explanation of photodo's 'scientific' tests awarding the Olympus 28/2 a score of 2.6?
2. Duff test methodology.
Unlikely, given the consistency of the results, and the exact correspondence with Klaus Schroff's Photozone test.
3. Canon Adaptor problem.
Hard to rule out: suspicious CA. However, I'm not sure it's possible to make a Nikon adaptor thinner than the Chinese one I used, and I don't know of a thicker one than the Fotodiox – anyone? The same adaptor works fine with every other Nikon lens I've used.
Actually, four options: it may be that the ZF25 is a very fine lens that looks great until you compare it with a properly sorted Canon 24L. If I'd tested the Zeiss against a Nikon 24/2.8, it would have covered itself with glory.
hubsand wrote:
3. Canon Adaptor problem.
Hard to rule out: suspicious CA. However, I'm not sure it's possible to make a Nikon adaptor thinner than the Chinese one I used, and I don't know of a thicker one than the Fotodiox – anyone? The same adaptor works fine with every other Nikon lens I've used.
The adapter placement issue could be rearing its ugly head again. If this lens is like the old CY28/2.8 where all the elements are locked together and move back and forth in unison to focus, adapter variation is a moot issue as the helicoid does all the work, but if their ZF25 relies on internal elements moving separately from other elements, then exact placement is critical to its performance. A lens mounting error of a mere .05mm either way could throw it out of optimal performance.
Without testing all available adaptors, and investing time in modifying them to 0.05mm tolerances, it's impossible to rule out the adaptor as a factor in this instance. But if it's that tricky to match, it will obviously be less appealing to Canon users.
However, if we take the adaptor out of the picture altogether, and look at Photozone's test – on a Nikon body – we see centre frame resolution scraping 2200 lwph, but the ZF25 doesn't even reach 1950 lwph on the DX frame edges until f8. Contrast this with the Tamron 17-50/2.8 zoom which resolves 1950 lwph at the frame edge . . . at f4 . . . at 17mm!
I'm seeing exactly the same with either adaptor on a full frame Canon, with predictable deterioration in the outer image circle. How that matches with the MTF is open to debate, but in practice is irrelevant in the light of test results thus far.
hubsand wrote:
Without testing all available adaptors, and investing time in modifying them to 0.05mm tolerances, it's impossible to rule out the adaptor as a factor in this instance. But if it's that tricky to match, it will obviously be less appealing to Canon users.
Absolutely. And let us be clear, that the lens was not designed to be used on Canon cameras. In fact, Zeiss recommends using the ZS, M42 mount, version of the lens, if you going to try to adapt it to fit a Canon.
hubsand wrote:
However, if we take the adaptor out of the picture altogether, and look at Photozone's test – on a Nikon body – we see centre frame resolution scraping 2200 lwph, but the ZF25 doesn't even reach 1950 lwph on the DX frame edges until f8. Contrast this with the Tamron 17-50/2.8 zoom which resolves 1950 lwph at the frame edge . . . at f4 . . . at 17mm!
I'm seeing exactly the same with either adaptor on a full frame Canon, with predictable deterioration in the outer image circle. How that matches with the MTF is open to debate, but in practice is irrelevant in the light of test results thus far....Show more →
I disagree with your conclusion. While we can expect slight deviations in performance relative to what can reasonably be expected based on the MTF's, such LARGE differences are not "irrelevant". If we accept, as you've indicated, that Zeiss is not lying or misrepresenting the performance of these lenses, and if we trust that the these MTF curves were accurately created with state-of-art equipment and highly trained professionals with extensive experience using procedures developed and honed over decades, then we must look elsewhere for a problem. In practice, and regardless of what your "test" seemingly shows, there will simply not be such a wide disparity between observed performance and MTF's measured from actual lenses.
As suggested earlier, there are other much more plausible explanations. The testing methodology, which I consider to include the adapter used, is one possible explanation. Another is that a batch of sub-standard lenses got out of production mistakenly -- either by a lapse in Zeiss' rigorous, well-known QC, or by an error that resulted in rejected lenses finding their way to market instead of being re-worked. If it is the latter, then Zeiss has a problem on it's hands that is needs to rectify becuase it goes to the very heart of the quality that Zeiss stands for -- and damage has already been done. I think the Tamron zoom lens comparison you pointed out only underlines the potential that this was a production problem. The lens, correctly executed as designed, simply could not be that bad. This is where sample images and testing of other examples of this lens (preferably from a different production batch) would be useful and revealing.
Quite frankly, if the test results are to be trusted, then the only logical explanation is that there was a production/QC error and you got a duff lens. I believe that the preponderance of collective experience of lens use over the past several decades would strongly indicate that in no way are Zeiss' published MTF's "irrelevant" to the observed performance of a lens accurately produced to their specification.
This does not mean that Zeiss does not hae a problem on their hands -- it does. And I would consider a potential quality problem such as suggested here to be much more of an issue for Zeiss than a rare weak lens design popping up in their product line.
To recap: we currently have two, objective, independently conducted tests on low-serial samples that agree in all respects. Perhaps later tests on later samples (in ZS mounts) will move the goalposts.
With regard to what is relevant, I would contend that the only relevant factor is how the lens performs on our cameras.
With regard to Zeiss' reputation: past performance is no guarantee of future results.
hubsand wrote:
As far as I know, no-one has suggested that Zeiss has fabricated their MTF data – so perhaps we can lay that to rest. Seeking explanations for the relatively disappointing 16:9 test, that leaves us with three options in the absence of a reply from Zeiss:
1. Sample variation.
Is this the same explanation we resort to in explanation of photodo's 'scientific' tests awarding the Olympus 28/2 a score of 2.6?
2. Duff test methodology.
Unlikely, given the consistency of the results, and the exact correspondence with Klaus Schroff's Photozone test.
3. Canon Adaptor problem.
Hard to rule out: suspicious CA. However, I'm not sure it's possible to make a Nikon adaptor thinner than the Chinese one I used, and I don't know of a thicker one than the Fotodiox – anyone? The same adaptor works fine with every other Nikon lens I've used.
Actually, four options: it may be that the ZF25 is a very fine lens that looks great until you compare it with a properly sorted Canon 24L. If I'd tested the Zeiss against a Nikon 24/2.8, it would have covered itself with glory....Show more →
But the MTF data is theoretical and ideal in regards to the mechanical assembly of the glass.. It says nothing about tolerances.
dcmiller wrote
But the MTF data is theoretical and ideal in regards to the mechanical assembly of the glass.. It says nothing about tolerances.
Actually, that is not true in Zeiss' case. Zeiss' publish MTF's are based on tests from actual production lenses. Canon's MTF data is theoretical -- and there is a big difference between theoretical and actual MTF data.
hubsand wrote:
....past performance is no guarantee of future results.....
You sound like a Mutual Fund
After perusing all the posts and comments here, I am inclined to believe hubsand's testing methodology and controls are sound, and all points towards the lens. It is indeed stratling to come to that conclusion, but if these were blind tests with no prejudices towards either lens, and you needed a lens to shoot with today, one would simply buy the lens that performed better, IE the Canon.
Even Leica has issues. The DMR/R9 had film plane/focus screen misregistration issues. Add to that, look at the M8 fiasco.
