From: John McWilliams on
Wilba wrote:
> Paul Furman wrote:
>> Mike Russell wrote:
>>> Here's something similar that will work on most versions of PS
>>> http://www.anntorrence.com/blog/2008/02/multiply-and-screen-to-adjust.html
>> Wow, thanks! Interesting how he ...
>
> © Ann Torrence. :- )
>
>> ... duplicated the effect with curves and that looks a lot like a gamma
>> adjustment. Seems like it incorporates the linear-data conversion to
>> adjusted-for-human-perception idea (if that makes sense).
>
> Yeah, very interesting to see. Layer blending modes always seems like
> black-magic to me, since I don't have a clue what's really going on. I often
> feel like a monkey with a typewriter with a lot of Photoshop.
>
>> "Multiply/Screen"
>> -storing that away for future use...
>>
>> Fascinating to overlay those two files in PS & scroll through the layer
>> modes. Pick one from the drop-down then use the up/down arrow
>> keys to browse.
>
> Isn't it? I wish I'd thought of that. :- )

IIRC, there's an arrow and key combo that zips you through all the
blending modes if you're in the right tool.
There are video tuts just on blending modes, and it's pretty neat stuff.
They are grouped, the modes, and a tutorial may show what they have in
common.

--
john mcwilliams

From: Mike Russell on
On Sun, 11 Oct 2009 06:52:19 GMT, David J Taylor wrote:

> "Mike Russell" <groupsRE(a)MOVEcurvemeister.com> wrote in message
> news:1szocmugrm49z$.dlg(a)mike.curvemeister.com...
>>> Paul Furman wrote:
>> [re Image | Adjustments | Exposure... ]
>>>> Hmph, CS1 doesn't have that.
>>
>> Here's something similar that will work on most versions of PS
>> http://www.anntorrence.com/blog/2008/02/multiply-and-screen-to-adjust.html
>>
>> --
>> Mike Russell - http://www.curvemeister.com
>
> Seems like a grossly overcomplicated way of adjusting gamma. Can't that
> expensive program do that directly? My much cheaper program can.

It's not a gamma adjustment. This is just the tip of the iceberg - imagine
for example using a layer in luminance mode, and adjusting each of the RGB
channels for maximum contrast.

--
Mike Russell - http://www.curvemeister.com
From: David J Taylor on

"Mike Russell" <groupsRE(a)MOVEcurvemeister.com> wrote in message
news:1tzlxx791oaem$.dlg(a)mike.curvemeister.com...
> On Sun, 11 Oct 2009 06:52:19 GMT, David J Taylor wrote:
>
>> "Mike Russell" <groupsRE(a)MOVEcurvemeister.com> wrote in message
>> news:1szocmugrm49z$.dlg(a)mike.curvemeister.com...
>>>> Paul Furman wrote:
>>> [re Image | Adjustments | Exposure... ]
>>>>> Hmph, CS1 doesn't have that.
>>>
>>> Here's something similar that will work on most versions of PS
>>> http://www.anntorrence.com/blog/2008/02/multiply-and-screen-to-adjust.html
>>>
>>> --
>>> Mike Russell - http://www.curvemeister.com
>>
>> Seems like a grossly overcomplicated way of adjusting gamma. Can't
>> that
>> expensive program do that directly? My much cheaper program can.
>
> It's not a gamma adjustment. This is just the tip of the iceberg -
> imagine
> for example using a layer in luminance mode, and adjusting each of the
> RGB
> channels for maximum contrast.

But that's not what was shown in the blog.

Cheers,
David

From: Porte Rouge on
On Oct 10, 3:30 am, fl...(a)apaflo.com (Floyd L. Davidson) wrote:
> sigh... <s...(a)noaddress.com> wrote:
> >On Thu, 08 Oct 2009 18:36:37 -0800, fl...(a)apaflo.com (Floyd L. Davidson)
> >wrote:

> Setting White Balance will have a dramatic effect of the
> accuracy of both.

Could you tell me how you go about setting the white balance on your
camera? I read this page:

http://www.guillermoluijk.com/tutorial/uniwb/index_en.htm

The first method is over my head and I have a Nikon that doesn't use
saturated pixels for the calculation of white balance. My white
balance can be set to Auto(+ or - 3), by lighting(tungsten,
fluorescent, flash), outside light (direct sun, shade, cloudy), select
a temperature, and presets(measure a gray card or use WB from another
photo). Which would you recommend?

Porte

From: Floyd L. Davidson on
Porte Rouge <porterougeman(a)gmail.com> wrote:
>On Oct 10, 3:30�am, fl...(a)apaflo.com (Floyd L. Davidson) wrote:
>> sigh... <s...(a)noaddress.com> wrote:
>> >On Thu, 08 Oct 2009 18:36:37 -0800, fl...(a)apaflo.com (Floyd L. Davidson)
>> >wrote:
>
>> Setting White Balance will have a dramatic effect of the
>> accuracy of both.
>
>Could you tell me how you go about setting the white balance on your
>camera? I read this page:
>
>http://www.guillermoluijk.com/tutorial/uniwb/index_en.htm
>
>The first method is over my head and I have a Nikon that doesn't use
>saturated pixels for the calculation of white balance. My white
>balance can be set to Auto(+ or - 3), by lighting(tungsten,
>fluorescent, flash), outside light (direct sun, shade, cloudy), select
>a temperature, and presets(measure a gray card or use WB from another
>photo). Which would you recommend?

I use Nikon cameras and shoot RAW+JPEG virtually all of
the time, using the JPEG for preview only and process
RAW files for production. The JPEG is often enough
previewed using the camera's LCD (I show somebody a shot
that was just made, or even let them scroll through
several images), but is usually seen via a laptop LCD.
Hence a realistic JPEG is important, but does not need
to be perfect.

Also, I use the histogram as a rough indicator of
exposure, and the "highlight" display to actually decide
if an image was exposed correctly.

I always use "Auto" for White Balance. The significant
effect of doing so, for my workflow, is that the camera
records whatever multipliers were used. If a preset WB
setting is used, the preset multipliers are recorded,
but we already know what those values are so recording
them is not useful (they are also used as preset values by
the raw converter). By using "Auto" I get a record of
what the camera thought WB should be.

When I process the RAW file (I use UFRAW) I can use the
camera generated auto values or I can use what the
program's "auto" setting generates. Sometimes they are
very similar, sometimes not, but usually one or the
other is close enough to what I want to provide a
starting point for manual adjustment of the program's
White Balance.

I have tried the UniWB concept, and it certainly works
as described. It was nice to have a very accurate JPEG
in terms of the histogram's indication of exposure,
but... I can't live with the greenish looking JPEGs
produced. :-) I can live with knowing that to nail
exposure I need to allow at least some portion of the
image to blink in the highlight display. I just keep in
mind that getting more dynamic range means a little more
area blinking, and positively avoiding clipped
highlights means a little less. It's a judgement call
that depends on circumstances, and in any case is never
very critical.

A lot of folks say what they want is an histogram
generated directly from the camera's raw data.
Un-interpolated raw data histograms aren't very useful
though! What I would like to see is the camera do one
interpolation for the JPEG image and an entirely
separate interpolation for a histogram. For those who
shoot JPEG, the settings would be the same (or just use
the JPEG for the histogram as is now done). For those
who shoot RAW the interpolation could be set for using
1.0 multipliers, low saturation, no sharpening, and
a standard gamma, to get a very accurate histogram.

I wouldn't mind seeing an "expanded" histogram display
either, where the entire graph shows only the upper 2
fstops of the histogram and also truncates the vertical
range by 1/2. What a tool that would be!

--
Floyd L. Davidson <http://www.apaflo.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska) floyd(a)apaflo.com