View Single Post
  #92  
Old November 7th 15, 01:32 AM posted to rec.photo.digital
Eric Stevens
external usenet poster
 
Posts: 13,611
Default How to measure ISO

On Fri, 06 Nov 2015 15:43:28 -0900, (Floyd L.
Davidson) wrote:

Eric Stevens wrote:
On Fri, 06 Nov 2015 22:46:38 +0000, sid wrote:

Eric Stevens wrote:


As I have already described I have been
exposing to the left by deliberately underexposing by 1/2 stop as they
advised.

But if you are really serious/religious about avoiding clipped
highlights, you should underexpose by 2 or three stops. 1/2 a stop only
will usually leave you with lots of clipped pixels.

See
https://dl.dropboxusercontent.com/u/...-7500601-2.jpg
and
https://dl.dropboxusercontent.com/u/...R--7500957.jpg

Both of these were taken hand-held with the D750 set on -1/2 EV.

Without stating what your are metering from and what type of metering your
are using neither of your posts have any real meaning. Pointing your camera
at different parts of the scene you want to photograph will indicate
different exposure values as will different metering methods for example
spot metering as opposed to centre weighted or evaluative etc.


I can't say for certain but I was probably using matrix metering.
Irrespective of the mode of metering I would have been checking both
the captured image and the histogram as I went.


I downloaded the second of the two. Not bad on a
technical basis.

There are lots of "blown highlights" in the stained
glass area, but the significance is that it doesn't
matter! Who cares if there is visible detail or not,
the whole point is to have bright colors. That is
exactly what you got!

The shot was made using center weighted metering, with
EC set to -1. And Aperture Priority, so the shutter
speed of 1/15 was set automatically.

The only complaint I would have is the use of ISO 1250
and a wide open lens that isn't that good wide open.
ISO 1600 or even as high as 3200 would have been
virtually the same, and would have allowed stopping that
lens down to a sharper aperture. That is almost knit
picking though, and is only important if you are going
to make a large print. Of course if a large print is
the objective, then it is overexposed a bit too much too,
as that lost detail would be visible in a large print.


Yes - I only thought to check the EXIF after I had posted. When you
have bright lights which are vastly overblown, such as interior
lighting, the sun, bright clouds or (in this case) clear glass
admitting light from the outside, it doesn't matter if they are over
exposed. They will print as white anyway. The colors were what I was
after.

I might have done things slightly differently if I had had more time.
At the time of this photograph I had not long had the camera and I
found myself part of a tour party rushing madly from one scene to
another. There was almost no time to think or fiddle with camera
settings. Accordingly I had configured two alternative program
settings. U1 was for outside. U2 was for inside. For U2 I selected an
ISO of 1250. As far as the aperturewas concerned, from what I had
read, the lens is sharpest at f/2.8. As with all zoom lenses you can
qualify that but it definitely works best when nearly wide open.
--

Regards,

Eric Stevens