Lens variations

Started 3 months ago | Discussions thread
ipopov Regular Member • Posts: 229
Re: Lens variations
8

Honestly, there seem to be an awful lot of "bad samples" floating around out there. Is the lens manufacturing process really so loose that a high number of bad samples would be produced?

Judging number of bad samples from forums will definitely be overestimating. People who have unexpectedly bad results are much more likely to complain about it on a forum than people who are happy with their lenses.

Are bad samples just an aberration, or is there a range of sharpness (from great to poor) that affects all lenses?

There is definitely a range of performance. Something like a bell curve. Also, different optical designs have different sensitivity to manufacturing tolerances.

Do lens manufacturers test a lens before boxing it up, and is there some minimum level of performance that renders the lens "good enough?"

I don't think manufacturers share any info on that. I believe some high-end manufacturers may test all the copies (may be Leica, etc), but budget lenses are most likely only sample a few from a batch.

All the manufacturers have some performance thresholds there a copy passes as "within spec", but these thresholds are not public, and according to Roger Cicala, are very lax. Also, manufacturer's own testing equipment and methods may be not quite state of the art.

Finally, how can one know if a newly-purchased lens is a bad sample, other than by making a totally subjective assessment by comparing results to results claimed by others?

You can test for some obvious defects, such as tilt/decentering (see "clock tower test" or Jim Kasson's "acceptance test"). Other than that, the best thing is to compare to performance of other users or published reviews. Only places like aforementioned lensrentals have enough data to make an educated conclusion.

I have neither the time, skill, or desire to run lab tests on every lens I buy, and if a lens doesn't meet my expectations, I have no problem sending it back.

Roger reports that manufacturers quite often reject lenses sent for return or calibration as "within spec" even if they are obviously not.

That said, I see no reason to settle for "bad copy" when a better copy can be had for the same price.

No sample will be absolutely perfect. You'll have to stop somewhere and declare it "good enough".

In general, if you are interested in this topic I recommend reading an Lensrental's blog and Roger Cicala's articles on dpreview

https://www.dpreview.com/opinion/6856813208/roger-cicala-the-difference-between-sample-variation-and-bad-copies-part-1

https://www.dpreview.com/opinion/9596141908/roger-cicala-the-difference-between-sample-variation-and-bad-copies-part-2

https://www.dpreview.com/opinion/4042117089/roger-cicala-why-i-dont-use-an-mtf-bench-to-test-my-own-lenses

https://www.lensrentals.com/blog/2018/11/using-rapid-mtf-testing-how-we-test-monitor-our-lenses/

https://www.lensrentals.com/blog/2015/06/measuring-lens-variance/

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow