Re: [OM] TOP - DxO Gives OM-D High Marks

Subject: Re: [OM] TOP - DxO Gives OM-D High Marks
From: Joel Wilcox <jfwilcox@xxxxxxxxxxx>
Date: Tue, 02 Oct 2012 09:31:09 -0500
On Mon, Oct 1, 2012, at 04:32 PM, Ken Norton wrote:
> > Don't mention it. If I didn't look daft, I'd be invisible.
> Then I suppose that we must have looked like the daft twins photographing
> in the rain.

That goes without saying.

> The premise of my statement surrounds the nature of the digitizing
> process.
> No two types of sensors see light and color in exactly the same way.
> During
> the analog-to-digital process, there are offset gains applied to the
> sensel
> measurements to correct for this. The offset gains, are essentially done
> in
> the same manner as applying gain for higher ISO settings, but are applied
> to the individual color banks. This is done prior to the creation of the
> raw file as it is applied at the sensor level or in the chip used for the
> actual A-D conversion.

Not to jump too far ahead, but your previous note seemed to imply that
one can improve an ultimate raw conversion with some sort of
pre-raw-conversion tweak.  We all want to know what that might be, and
that's what I'm looking for in the verbiage below.

> This is a vital step 

For whom or what?  For me?  For the camera?

> because you want to have the color normalization
> characteristics applied during the digitization process (or even before)
> because once the raw file is created, the brightness levels and gamma
> characteristics are baked in. The typical A-D converter in our cameras is
> usually around 12 to 14-bits. If the image data wasn't normalized prior
> to
> the digitization, you could end up with many of those bits effectively
> thrown out the window and unusable.

How do I get to affect what gets baked in?

> For illustration sake, let's say that Panasonic and Sony both developed a
> 16MP sensor for the OM-D. (not far fetched, I'm sure. hehehe). What is
> the
> likelihood that both sensors had exactly the same sensitivity and linear
> response? Considering that the color filters themselves vary a bit and
> the
> pixel fill-factor is different, the chances that both are exactly the
> same
> are nil. Just for argument sake, let's make an assumption that the
> Panasonic sensor is 10% more sensitive to greens than the Sony sensor. To
> maintain same WB, we would have to dial in an adjustment on 100% of the
> images. What the manufacturers do instead is apply an amplification
> offset
> (gain adjustment) to the chip to give the sensor a standardized
> white-balanced so no further correction has to be done. Without this
> amplification offset, once the digital conversion happens, in order to
> maintain white-balance, you will lose up to 10% of the dynamic range.

So this is all about working to make WB settings non-destructive or less
destructive?  Is this connected by chance to your notion of using
daylight as the default WB setting?  Then if one wants to use AUTO in
raw conversion, there is more color information to work with?

Joel W.

http://www.fastmail.fm - Access all of your messages and folders
                          wherever you are

Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/

<Prev in Thread] Current Thread [Next in Thread>