Julian Loke wrote:
>
> >> Julian Loke wrote:
> >> ... ignoring "bright segments" or "dark segments"? ...
>
> Hi Hugo, Thomas, and List,
>
> Could the quote also be describing a comparative weighting system
> (e.g. 60% primary, 30% secondary, 10% peripheral)?
> And could a simple a weighting system e.g. 70/20/10/0% still be
> consistent with Canon's own explanation of Evaluative metering?
>
> As a example, a specular highlight could cause a SINGLE segment
> to be 3 stops brighter than all the other segments. It should
> "intelligently" be ignored.
>
> So, if a right-sided AF sensor was used, would a single specular
> highlight on the left be ignored.
>
> But this would also occur in a simple weighting system, just
> because the left side is "peripheral" and receives 0% weight.
>
No, it's not as easy ;-) I finally found the source of my
"knowledge". It was in the Magic Lantern guide for my EOS 5.
My copy is in German, so I don't know if it's in the English
version as well. It's also unclear IMHO, if the author had his
information from directly Canon or from own experiments. Anyway,
he states, that a bright part of the scene outside the primary
zone would get less weight (as you described above), but that
*very* bright parts (like specular highlights) would be completely
ignored even in the secondary zone.
> So what would happen if a single specular highlight fell on the
> right?
>
> Should it be "intelligently" ignored?
> Or will it contribute a fixed weighting (e.g. 20%) to metering?
If it's in the primary zone, it would likely lead to underexposure.
If it's anywhere else, it should be ignored.
>
> Cheers
> Julian Loke
> P.S. I am having trouble understanding Philip Chong's article
> "Using the Evaluative metering of your 35mm EOS cameras"
> http://www.camera.canon.com.my/photography/art/art14/02.htm
> Can anyone explain to me the point of his article?
No, I can't ;-) I have to admit I read only the first page. I stopped
there having read the remark about the "perfect" exposure of the biker
despite the bright background. To my eyes, the exposure is correct for
the bright background but not for the biker, who is clearly underexposed
IMHO. Anyway, during my own experimenting with the evaluative metering
I came to the conclusion, that the evaluative metering usually tries to
squeeze the brightness levels of the whole scene into the (about) 5 stops
that (slide) film or prints can reproduce. With backlit scenes, this
still gives an underexposed main subject, even though the weighting of the
background may be smaller than that of the main subject. So I think, that
it really is some kind of average metering in those cases. It really
*ignores* specular highlights, though. It had no problem with pictures
that included the sun in the frame.
Thomas Bantel
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
