On 12/2/2012 2:44 PM, Chuck Norcutt wrote:
> Thanks for those links. I had started to read the first when you (?)
> posted it a week or so back but got interrupted and never finished it.
> I think I understand now but am still having trouble with his comments
> about large pixels and sensor saturation. He says:
> ----------------------------------------------------------------
> 1. Sensor ISO is based on the amount of light it takes to saturate the
> sensor to pure white. This is what DxOMark measures and reports.
> Sometimes that yields counterintuitive values. Double the collection
> efficiency of your sensor so it only takes half as much light to
> saturate the sensor, and you've doubled the ISO. That makes sense.
>
> Suppose, though, you just make larger pixels that can hold more
> photoelectrons. What happens then? Well, it takes more light to saturate
> the sensor, so the sensor ISO drops even though the sensor still has the
> same sensitivity to light.
> ---------------------------------------------------------------
> The first paragraph makes perfect sense. But in the second he says
> large pixels hold more photons so it takes more light to saturate the
> sensor. That doesn't make sense to me. It certainly takes more photons
> but the pixel size is an area measure. The pixel will collect more
> photons in a given unit of time just by virtue of its size. It's not
> evident to me that sensor ISO should drop. More photons were needed but
> more should have been received during the same exposure time. There's
> something here that's not stated or I just don't understand. Equated to
> film it sounds to me synonymous with saying that the ISO of 2-1/4 film
> should be less than that of 35mm film because the collection area is
> larger. Eh?
AHA, you read it all this time. ;-) Yes, I thought he was wrong when I first
read it. But many comments had already
piled up by then, and I didn't feel like posting a long, arcane disagreement.
AND, if you excise that second pp and the words "with their huge pixels and"
from the next sentence, it reads:
"This is one reason why many large format digital backs, with their huge
dynamic range (which means a huge exposure
range) come in with such low DxOMark ISO values."
It's not until you get to the first part of the next column, and its first
graph, that what he is saying becomes clear.
The title "RAW is not Raw" (rather than agreement with your argument about
capitalization) is saying that the designers
adjust/fudge the linear sensor response curve by setting the camera ISO. In the
case of the E-M5 , the effect is a
lowering of the sensitivity to light - but - only in the upper end of the
histogram.
As soon as "Raw" becomes non-linear in this way, the measurement of ISO by
saturation becomes less closely related to
photographic ISO. That it takes more light to push the E-M5 to saturation,
while it renders the same midpoint as other
cameras, is a good thing. As Ctein says about his last illustration, they have
devoted all the increased DR of the
sensor to the top end, for smoother, more film-like rendition of highlights,
with smoother roll-off and less clipping.
As he puts it in his review of the E-M5, "In Figure 1 the direct reflections of
the sun in the glass of the greenhouse
and the cars of the Ferris wheel are far beyond the exposure range of any
medium, silver or silicon, but they fade out
gracefully without looking artificially clipped."
Have I managed to turn it around so you might consider the difference between
'sensor ISO' and 'camera ISO' in DXO's
measurements to be a good thing? ;-)
Change-up Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|