Thanks for that information, Chuck.
I've just taken a series of shots (RAW) with my K-5 to see what was usable. I
then tried to upload the results so that you could see. Then the upload took a
lunchtime to send one image (!) and I realised that I had taken the shots with
-0.7 compensation set. I could re-compensate in Aperture, but I wondered then
if that made a difference compared with no compensation in the first place.
So, I took the shots again, 400, 3200, 6400 and 12800; below 3200 is all pretty
good, above 12800 is usable but messy. And the results are better with no
original compensation, obviously, I suppose.
My conclusion is that up to 3200 and the noise-speckling is fine and I can
ignore it. At 6400 the noise remains almost as fine, and at 12800 it's obvious
at normal viewing resolutions.
Subjectively I should say that my K-5 is marginally better than your 5D, but
that's not particularly scientific a test :-)
Chris
On 7 Jan 2012, at 11:43, Chuck Norcutt wrote:
> The 5D has a normal ISO range of 100-1600 selectable within 1/3 stop
> increments. It will also do 50 or 3200 but they are not labeled as
> such. They are labeled "L" and "H". To be able to use them at all they
> must be deliberately enabled by a custom function setting and there are
> no 1/3 stop increments between 50 and 100 or 1600 and 3200. This
> deliberate custom function setting requirement says to me something
> like: "We really don't want you to go there but, if you insist, we'll
> allow it and you agree to accept the risk."
>
> I've never seen high ISO images from a 5D Mk II but the Mk II increased
> the definition of "normal" ISO range by 2 stops from 100-1600 to
> 100-6400. It also has an expanded range that must be selected via
> custom function. It has an L = 50 setting but adds 2 high settings, H1
> = 12,800 and H2 = 25,600. I have assumed that the 5D Mk II's 6400 is as
> good as the 5D's 1600 but do not know that for sure. A 2 stop
> improvement was not enough to make me jump but a 3 stop improvement
> (that I believed) probably would. For the few times I've been forced to
> use 3200 it's usually OK but subject motion can be a problem. I've
> concluded that a 3 stop improvement would allow a fast enough shutter
> speed to avoid the motion blur. It would also help compensate for my
> lack of IS lenses.
>
> There is another clue to noise levels in the Canon full auto modes. The
> 5D's full auto mode automatically selects ISO but limits the range to
> 100-400. The 5D Mk II's auto ISO range has been increased to 100-3200.
> Since only JPEGs are allowed in full auto mode I'm not sure that's an
> improved sensor development or just better noise reduction firmware.
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|