It will continue to do so until you switch it again. Were you looking through the menus at some point?

I would suspect you set bracketing inadvertanlty, so the exposures are normal, over and under exposed.

Or maybe HDR?

Interesting, thanks for the link.

By a whisker, and that can easily mean slight decentreing just as well as a slight focus shift. I would say that's close enough to mean even coverage.

As regards the second test, why believe that one and not another one?

If the tests have been done with the Paterson chart, someone would be taking issue with the film/developer combination used.

If the images used instead were the Wellington on the Thames, then someone would eventually notice that boats actually move up and down with the flow of the water.

We have to have something, but it's only a guide and there's a lot of useful information in there. You've been given the full detailed charts in my reviews, so there's plenty of scope there if you know what to look for.

Somehwere the suggestion came up that when people buy wide aperture lenses they do so to use them at wide apertures. Now looking around I noticed that in relation to the SMC Pentax 50mm f1.2 lens Photo Technique observed something like this (quoted from memory)

Of course apart from black cats in coal cellars there are not many situations where say even f2 is essential, but it does lead to brighter viewing and easier manual focusing.

So what do we think? Do lens purchasers buy wide aperture lenses for using wide open, or is there another reason as well, not necessarily the one postulated by the magazine?

I've removed that e-mail address and strongly advise against using it openly on the forum. You will attract every bot and spider in existence.

If the OP would care to PM you, then you will be able to reply.

Algi, you are misreading the information. So what is wrong with the 16-45mm showing excellent performance? It's an excellent lens. Gary doesn't give you the figures though, so you can't compare with later tests.

The kit lens results are also very good but nowhere do they suggest the edge is better than the centre.

The edges can be better than the centre at wide apertures though because of lens decentreing. This causes spherical aberration that reduces as the lens is stopped down.

Zeiss lenses can also perform poorly. Way back, I've just been looking at some wide aperture standard lens tests and the Noct-Nikkor wasn't as good in low light as the standard f1.2 Nikkor.

Throw preconceptions away, look at the facts. Those who doubt can always do their own testing and see for themselves, but bear in mind that samples of lenses vary as well. Lenses can be as much as 20% decentred and still give a good account of themselves, leaving us to wonder what they might achieve if they were put together right....

Oh my, if you want open aperture there's an example in the 77mm review.

We need to realise that this is a review, not a lab test and there's a limit to what is done in the available time. There is also a practical limit in terms of space because if the reviews were all twice as long who is going to pay for all the time it would take in preparation? People are working to get these things on the web.

Different readers are interested in different things, so we should be thankful we have several reviews to look at and to aid us in our lens selection.

I'm aware that there are limitations to everything, but we have to start somewhere. I chose this image as there were lots of colours, shapes, shadows, bright areas and as the DOF has been limited also the gradation from sharp to the out of focus areas.

http://www.ephotozine.com/articles/x-rite-colorchecker-passport-review-16829/ima...

It could be argued that another image would be better, but then where do we stop? It's up to the individual to take the point further, do his ir her own assessment and reach a conclusion based on what their own work and output is. This was a talk of ideas and limited experimentation, setting the scene if you will.

It's probably a copyright issue as the file is a video file.