I've never heard of any reviews or tests that report this. But, even if your friend is right, how much inaccuracy are we talking about? Would the actual aperture be, say f/3.0 instead of f/2.8? Would that be significant? This sounds like something a diehard film fanatic might come up with - and I used to be one of those!

Why don't you see if you can get some more info from your friend about his source, test report, reviews, whatever, and bring it back here? I'll bet you a beer that he got it off the internet......

Thanks for the response.

From his point of view up to 2 stops. Way off base. Even I could concede a couple 10ths of stop but no more.

Yep, I think a two-stop error would be obvious even to a casual user shooting in automatic mode. Another urban legend bites the dust......

He finally got back to me on how he's "testing" these lenses. This was his response:

Quote:
I have tested this and gotten f-stops that I know the lens can not do; one of the easiest ways to detect this is to go to an aftermarket lens and watch it do the same thing - give you readings that are impossible

That is how I am determining that the "setting" and actual aperture do not agree, it is giving you the closest calculation (i.e. 1/3 stops) when the aperture leafs only have settings for 1/2 stops; etc. so, say the aperture only does f 11 and f 16 it will still give you f 13, and f 14 even though the actual aperture has no such mechanical setting available to it

How he is doing this is anyone's guess. But, out of curiosity I watched the aperture leafs on my lenses (hold dof preview while adjusting aperture) change for every 1/3 stop the camera can set. Every lens I have, Canon, Tamron and Tokina I can visually see an aperture change for the full range of settings in 1/3 stops. Which leaves to believe that he is either using older mechanical aperture lenses or just doesn't know what he's talking about. And at this point further discussion would appear useless, he is never going to be convinced he is incorrect.

Thanks for the feedback.

There was an interesting discussion in another forum in which the same kind of question came up, how accurate are ISO's and the answer was much more surprising than I anticipated. The problem with your question and your friends test is that it not only is testing f stops but the camera's other features - shutter speed and ISO.

Historically it turns out, ISO's (ASA's) just weren't that accurate and there are different ways to say an ISO 100 is an ISO 100. I tested my Canon 10s (film) with my 300D and 450D, and none of them agreed when I presented the same scene to them, but the deviation wasn't that dramatic. Amazingly my 300D agreed more with my film camera than my 450D.

So these f stops, shutter speeds, and ISO's are sealed in cement.