I've gone back and re-read Carlos's post (and also checked the specs on
80A filters). The filter factor for the various 80A brands sold by B&H
varies from 1 to 1.25 stops. Not the 2 stops Carlos mentions but,
nevertheless, still a significant boost to be applied in post...
possibly affecting both noise and even posterization.
So, there is clearly a good point but there are confounding variables.
Compensating for the filter requires at least doubling the exposure time
or opening up a stop or doubling the ISO. That's the negative half of
the pie I mentioned. If it can be done by doubling the exposure time or
opening up a stop then I think there's a gain. But, that's not always
possible. If it's necessary to double the ISO then the effect of a 1
stop noise increase has just been applied to all channels.
My own solutions would favor shooting to the right whenever possible ans
adjusting the red and green channels down rather than adjusting the blue
channel up. I would also tend not to do a complete white balance, I
would still leave the image with a bit of red/yellow cast. But even
more likely, I would fill the room with electronic sunlight which also
helps to alleviate the mixed lighting problem which is everywhere these
days.
Dr. Flash
On 10/12/2011 10:42 PM, wayne.harridge wrote:
> I think Carlos' approach using an 80A filter has some vaidity.
>
> The exposure is properly balanced over the RBG channels so less
> chance of saturating the Red in giving adequate exposure to the Blue.
> This means that each channel is operating in the same exposure region
> of it's "characteristic curve" - compare this with film. The end
> result being that the noise should be more equally distributed across
> RGB channels and I expect this has some benefit in reducing noise
> reduction artifacts.
>
> ...Wayne
>
>
>> -------Original Message------- From: Chuck
>> Norcutt<chucknorcutt@xxxxxxxxxxxxxxxx> To: Olympus Camera
>> Discussion<olympus@xxxxxxxxxxxxxxxxx> Subject: Re: [OM] ETTR with
>> the Sekonic L-508 Sent: 13 Oct '11 03:00
>>
>> I think you're only considering the favorable half of the pie and
>> ignoring the unfavorable half. It's an interesting concept that,
>> by using a blue filter, you have avoided having to increase
>> exposure on the blue channel in post processing. You think you
>> have avoided noise in the process and so you have. However, the
>> act of using the filter has also increased the total exposure time
>> and has thus introduced additional noise across all channels. It's
>> not clear to me that you have gained anything of significance.
>>
>> Chuck Norcutt
>>
>>
>> On 10/12/2011 9:57 AM, Carlos J. Santisteban wrote:
>>> Hi Chuck and all,
>>>
>>>
>>> From: Chuck Norcutt<chucknorcutt@xxxxxxxxxxxxxxxx>
>>>> Sorry, I didn't follow your logic about decreasing effective
>>>> ISO. Can you first define "effective ISO"?
>>>
>>>
>>> OK, I think I'm mixing two related, but not identical concepts
>>> here. By 'effective ISO' I understand the achieved exposure value
>>> for a certain lighting conditions. _Overexposing_ with the intent
>>> of recovering the proper tonal scale in post (be it on digital or
>>> film) means going to exposure values 'just right' for a lower
>>> ISO, thus _decreasing_ that 'effective ISO'.
>>>
>>> On the other hand, assuming that most sensors' native WB is close
>>> to daylight (5500 K) the Tungsten setting is achieved by
>>> _cranking up_ the blue channel gain, and possible decreasing the
>>> red gain... The blue channel with its 'boosted ISO' is going to
>>> be the major contributor to noise in Tungsten-WB shots -- it is
>>> in my current GF1 and was that way in the EOS-300D.
>>>
>>> Using a 80A filter instead of the WB setting means getting the
>>> blue channel back to a reasonable gain, increasing picture
>>> quality as per my experiment. But the exposure correction for the
>>> filter factor (x4) implies exposure settings of a _quarter_ of
>>> the set ISO (e.g. 200 instead of 800).
>>>
>>> It's likely that shooting at that reduced ISO without filter but
>>> on Tungsten setting won't get the blue channel gain any higher
>>> than with the filter on (and high ISO + daylight WB) but haven't
>>> actually tried.
>>>
>>> Cheers,
>> --
>> _________________________________________________________________
>> Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
>> Archives: http://lists.thomasclausen.net/mailman/private/olympus/
>> Themed Olympus Photo Exhibition: http://www.tope.nl/
>>
>>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|