EF Lenses - Bad Copies?

[Gazer]

Greetings. I was wondering if someone may be able to shed some light on HOW a lens can leave its point of manufacture and end-up in a clients hands as a "bad copy" as I've read over and over again in a multitude of forums. Aren't these 1k+ lenses thoroughly tested before they are placed in a box for distribution?


Once again, thanks for any input.

[RichardTaylor]

Is this just EF lenses or all lenses?

EF lenses are checked & tested - See 4 minutes into part 3 of the virtual lens plant.
Here:
http://www.canon.com/camera-museum/t...ant/index.html

[Sime]

Not all EF lenses are 1k+

Where did you read this? be interested to know more. (But don't have time to Google right now)

Sime

[OsmosisStudios]

Originally Posted by Gazer

Aren't these 1k+ lenses thoroughly tested before they are placed in a box for distribution?
.

Every single one? No.

Quality control means that you test one out of every x number.

Assuming Canon (or any manufacturer, for that matter), tests 1% of it's lenses, the tester could try one lens, find it to be absolutely perfect, and yet miss the one right next to it that is severely out of alignment. Remember, these are mass-produced items, so theres a fair bit of sample differentiation.

Now, Canon, Nilkon, Zeiss, Leica, Olympus, etc all test a rather large number of lenses, and their tolerances are rather tight. Sigma, Tamron, Tokina have a larger sample variation because their QC rules are considerably more relaxed (less copies tested, greater tolerances, etc).

The lens would actually leave the factor with the defect, not gain it along the way.

[Gazer]

First off, thanks Osmo, your explaination is what I was looking for.

I emphasized EF lenses because they are the only ones I've looked into (and the only ones I'm interested in discussing). I mentioned 1k+ lenses for the same reason.

I actually thought that the more expensive lenses, say those with an 'L' designation, were individually tested before release - naive or what? Anyhow thanks for clearing that up.

[OsmosisStudios]

Originally Posted by Gazer

I actually thought that the more expensive lenses, say those with an 'L' designation, were individually tested before release - naive or what? Anyhow thanks for clearing that up.

It would be totally impractical to test each and every one. Hell, Canon makes several hundred (if not thousand) 24-70L and 70-200L lenses each week: testing each one is just prohibitive.

[inkista]

The other issue, when it comes to autofocus, is that the camera body may be out of alignment, too. When cameras/lenses are tested, they're tested for an acceptable range. As in, if it's ±2 from "ideal" (I'm just pulling these numbers out of thin air, btw), and the camera's -2, and the lens is +2, then you've got a total variance of 4, which is why someone might complain.

And lots of people have no idea how to properly use the autofocus microadjust feature, so they blame the gear, especially if they paid for an L and expected it to be like the second coming and up their overall image quality/keeper rate all on its own.

Mostly, when I hear stories of bad copies of a lens, I think "user error" and "pixel peeper". That's not to say there aren't any bad copies. I just think they're rarer than you'd think from reading sites like dpreview, where everybody's a measurabator . Remember, folks with perfectly good copies are probably out shooting with them, rather than complaining about them on messageboards.

[Gazer]

Thank you Inkista. These are very valid points.

[4is2tooMany]

Note that many newer, high end camera bodies have "micro focus adjustment" ability built in.

So if you buy a lens that is not focusing exactly correct on your camera, you can have the camera compensate for it. The camera then remembers that adjustment for that lens. And it can remember many different lenses. So once adjusted you dont ever have to repeat the process once your have set up all your lenses.