Comparing sharpness of 16-80 to other lenses

Started 9 months ago | Discussions thread
boogisha Senior Member • Posts: 1,160
Re: Comparing sharpness of 16-80 to other lenses

Iuvenis wrote:

I'm not talking about 'comparing them meaningfully'. I don't know what the point would be.

Hmm, the point being for the comparison to have meaning? Otherwise, any conclusion taken out of a meaningless comparison is, well... meaningless.

How can you compare one and the same lens with itself? Any difference would have to be down to the testing.

No, because it`s not "one and the same lens with itself", but physical lens compared with its theoretical model. Two already (sanely) expected to differ, it`s up to testing methods to stay the same as much as possible in order to showcase differences between the two.

Testing method used for simulation needs to account for all (as many as possible) caveats found in testing of the physical lens itself - or vice versa, physical testing needs to try to stay true to simulation as closely as possible.

This is where I think you're confusing things. The lens didn't perform at all for the Nikon mtf chart - it's just a simulation.

No, as that is my point as well. The important part is that we don`t know all the parameters of the simulation, so comparing it against a physical test outcome and saying the physical one is better is meaningless, as two test methods (simulated vs physical, outputting two MTF charts) might differ a lot, thus direct MTF comparison in purpose of lens performance evaluation has no point.

, it might just be a measuring method difference (which we are not aware of).

No, it must be a measuring method difference, since the lens is the same.

Nope, as we agreed above, in case of simulated MTF there is no lens at all. So in order to be able to compare simulated vs real life lens performance, simulated vs real testing methods needs to be the same - which is not in case of Nikon (Canon, Fuji...) vs Lens Rentals you mentioned, as manufacturers don`t disclose their simulation method parameters.

But, I`m a bit confused now - first, you said that comparing Lens Rentals` (physical) MTF against Nikon`s (theoretical) MTF can be used to show that lens real-life MTF can be, in fact, better than expected (simulated), but now you claim that any difference between simulated vs physically measured MTFs would come down to measuring method difference...? Could you pick one, please?

If your _only_ point is that physically measured MTF can look better than simulated MTF, even if only due to different methods used for measuring (discrepancy between physical setup and simulated setup), then I agree, but that tells us nothing about simulated vs real-life _lens_ performance - which is what we are all interested in here instead.

And for that, we`d need simulated and measured MTFs to be produced using the same (simulated vs physical) testing setup and measuring methods. In all the other cases, MFT producing approaches being different, this:

luvenis wrote:

You can compare Lens Rentals' mtf chart of the Nikon 70-200 and Nikon's theoretical mtf chart for the 70-200 and say that the Lens Rentals chart is better.

Given they are for the same lens, you can then say that theoretically modelled mtf, which is what most manufacturers use, is not necessarily higher than real-world mtf. It could be lower (and in the case I gave, it is).

... does not stand in terms of comparing/evaluating/expecting lens performance.

 boogisha's gear list:boogisha's gear list
Canon PowerShot A75 Canon ELPH 300 HS Canon PowerShot S120 Canon PowerShot G7 X Canon EOS 5D Mark IV +11 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow