Moose, I think you are right, it is highly possible A/D problem, I have
checked with some low ISO images, there are large size "noise" at the shadow
area even without boost up, what 14 bit A/D seems not doing better than the
E-1, all are BS, even the 40D images look better.
There has been a long time curious for me why some OOF background didn't
look as smooth as they should be, it was not enough bit deep. Now there is
one more thing to look for when checking out a new camera.
C.H.Ling
----- Original Message -----
From: "C.H.Ling"
> Yes, there is similar problem even at full resolution:
>
> http://www.accura.com.hk/temp/IMG_6442.CR2 (23MB)
>
> I really have no idea how this happen as a totally dark low ISO image is
> very clean even boost up, at the mean time high ISO image is clean except
> the shadow, just like problem with low ISO image.
>
> C.H.Ling
>
> ----- Original Message -----
> From: "Moose"
>
>> C.H.Ling wrote:
>>> ...
>>>
>>> Just found another phenomenon, when I shot a dark frame at the same ISO
>>> and shutter speed (ISO125, 1/8s) of the problem image. The dark frame
>>> came out totally dark even I push it two stops so I can't use the dark
>>> frame subtraction method to cancel the pattern.
>>>
>>> The pattern noise only appear on the area that is not totally dark so it
>>> doesn't like a hardware (circuit noise) problem, may be a firmware
>>> update
>>> can really solve it?
>>>
>>
>> As I think about this, I wonder if the effect is the same regardless of
>> the resolution at which the shot is taken. A repeating pixel level
>> pattern could result from the way adjacent pixels are merged (I assume
>> they are merged?) for the smaller sizes, like the 5.2 MP version you
>> posted. If so, a firmware solution may be possible.
>>
>> I didn't expect dark frame subtraction to help. That's a different sort
>> of issue without any pattern.
>>
>> I think it was AG, earlier in this thread who suggested it could be a
>> demosaicing artifact. I could see how that could be the case, although I
>> don't see any meaningful difference using the three algorithms available
>> in RawTherapee.
>>
>> Down at the bottom of the brightness range, the nature of linear digital
>> coding means that there are very few different numerical values
>> available to represent shades of brightness.
>>
>> Whatever the analog ability of the sensor to react to tiny differences
>> in light, the analog voltages have to be encoded digitally. Perhaps this
>> is some sort of quantization effect where tiny diffrences in voltage are
>> forced into "buckets" with average values that create too much contrast
>> between pixels. This could magnify any subtle pattern of sensor site
>> sensitivity differences resulting from physical sensor design.
>>
>> If that's the source, it may require a hardware changes, perhaps to the
>> Ato D converters.
>>
>> (Now that I think of it, at what point do the sensor sites start showing
>> the effects of the number of individual light quanta that they sense?)
>>
>> Moose
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|