Recommended monitor luminance levels?

In article ,
says...
-In article , Ilya Zakharevich says...
-
- Show a white image on your monitor. Switch your (d)SLR to spot
- metering, point it to the white area, and read the exposure it is
- suggesting. (Mine reads 1/80sec f/8 with ISO100.)
-
- Given your number, we poor candelameter-less people may start to
- appreciate what you are talking about (for a photographer, what I did
- is a cheap brightness-meter with precision about 20%; but it is not
- calibrated without your report).
-
-But the measurement obviously depends on the distance between the camera
-and the screen. You have to specify the distance.


obviously wrong.

the exposure of your car at 6 feet
is exactly the same as at 100 feet.

-

--
Bob Crownfield

Ilya Zakharevich wrote:
[A complimentary Cc of this posting was sent to
Kulvinder Singh Matharu
], who wrote in article :
Well, my eyes have now got used to the lower luminance...in fact,
I've now gone down to 90 cd/m2. It makes all the difference for
monitor/printer matching. So I'm now happy :-)

Just for reference, could you please spot meter the white
(RGB=255/255/255) area on your display at 100ISO?

Thanks,
Ilya

Boy, that's gotta be helluva bright. My LCD screen, plenty bright for
me reads 1/13 at f/8.

Colin D.

--
Posted via a free Usenet account from http://www.teranews.com

Ilya Zakharevich wrote:
[A complimentary Cc of this posting was sent to
John McWilliams
], who wrote in article :
Well, my eyes have now got used to the lower luminance...in fact,
I've now gone down to 90 cd/m2. It makes all the difference for
monitor/printer matching. So I'm now happy :-)

Just for reference, could you please spot meter the white
(RGB=255/255/255) area on your display at 100ISO?

Could you explain a bit how one does this, and what expected results
might be?

Show a white image on your monitor. Switch your (d)SLR to spot
metering, point it to the white area, and read the exposure it is
suggesting. (Mine reads 1/80sec f/8 with ISO100.)

Given your number, we poor candelameter-less people may start to
appreciate what you are talking about (for a photographer, what I did
is a cheap brightness-meter with precision about 20%; but it is not
calibrated without your report).

Thanks,
Ilya

P.S. If you do not know how to create a white image, I put one on
ilyaz.org/photo/tmp/wh400.png
(created with `convert -size 400x400 xc:#FFF wh.png').

Thanks. I'm installing a new video card today, then will recalibrate.

Complimentary copies, however kind in origin, aren't appreciated by
many, or end up in a Spam trap. They are also counter to NG concepts of
sharing everything in public. Kindly stop!

--
john mcwilliams

On Wed, 5 Mar 2008 22:22:03 +0000 (UTC), Ilya Zakharevich
wrote:

[snip]
Given your number, we poor candelameter-less people may start to
appreciate what you are talking about (for a photographer, what I did
is a cheap brightness-meter with precision about 20%; but it is not
calibrated without your report).

OK. Spyder says 90 cd/m2.

At ISO200 in P-mode and spot-metering (Canon D40 with 28-135mm
f3.5-5.6 IS USM):

ISO200
f5.0
1/50 second

I don't think that there were any reflections from the screen to
affect the results.

Would have got back to you earlier but been busy getting some travel
gear!

--
Kulvinder Singh Matharu

Website : www.MetalVortex.com
Contact : www.MetalVortex.com/contact

Blog : www.MetalVortex.com/blog
Experimental : www.NinjaTrek.com

Brain! Brain! What is brain?!

On Wed, 05 Mar 2008 09:12:47 -0800, John McWilliams
wrote:

[snip]
Are you printing from PS? I generally print from LR and it's difficult
to use third party profiles, on a Mac, at least. I got the 3800 for
Christmas, and now I am confident that when a print turns out poorly....
it's my fault. Although I sometimes can blame the gremlins that get into
the print driver settings.

I process in CS3 and then use Qimage for the actual printing. Qimage
is really good and I've forgiven its quirky interface, much in the
same way that I forgave Photoshop's interface once I started learning
where everything was and how much power I had there!

Thanks for revisiting this subject! (The luminance one, that is) I hope
we don't eventually have to process images separately for print vs. web,
though some already do that (not just size and color space, but luminance)

Yes, I've been thinking about this for a while now, and I think that
getting the new printer and trying to match screen and print got me
of my behind and do a bit more active research on this!

The way I see most people's screens these days and the super-bright
default settings on new LCDs I'm afraid we may just have to have
different settings :-(

--
Kulvinder Singh Matharu

Website : www.MetalVortex.com
Contact : www.MetalVortex.com/contact

Blog : www.MetalVortex.com/blog
Experimental : www.NinjaTrek.com

Brain! Brain! What is brain?!

Kulvinder Singh Matharu wrote:
I've just been calibrating my new printer but during the process I
found recommendations that monitor luminance level should be between
90 cd/m2 and 100 cd/m2.

I checked my (calibrated) monitor and it was something like 190
cd/m2. So I've now adjusted my monitor to 100 cd/m2, gamma 2.2 and
colour temp of 6500K.

Everything looks gray and dim! So are these luminance levels really
the recommended values for a printer-workflow environment?

This probably accounts for why *some* people think that my
web-gallery photos are too dark.


Here is an example which looks dark to me:
http://www.metalvortex.com/myphotos/antarctica/old_buildings.htm

Here is the histogram for it:
http://www.theknack.net/img_histogram.png

Not only is the distribution of pixels skewed left (i.e. dark), but the
upper 15% (or so) of the color range is almost entirely unused.

Personally I would always use the histogram and not rely on calibration
and subjective adjustments (which are subject to ambient lighting on the
monitor, etc). I have a bunch of old scans which are too bright because
I relied on a faulty monitor.

[A complimentary Cc of this posting was sent to
Kulvinder Singh Matharu
], who wrote in article :
OK. Spyder says 90 cd/m2.

At ISO200 in P-mode and spot-metering (Canon D40 with 28-135mm
f3.5-5.6 IS USM):

ISO200
f5.0
1/50 second

Thanks. But this puzzles me; at ISO100, this would be f/5 and 1/25
sec. Mine is f/8 with f/80, which gives about 740 cd/m2...

Strange,
Ilya

In article , timeOday-
says...
-Kulvinder Singh Matharu wrote:
- I've just been calibrating my new printer but during the process I
- found recommendations that monitor luminance level should be between
- 90 cd/m2 and 100 cd/m2.
-
- I checked my (calibrated) monitor and it was something like 190
- cd/m2. So I've now adjusted my monitor to 100 cd/m2, gamma 2.2 and
- colour temp of 6500K.
-
- Everything looks gray and dim! So are these luminance levels really
- the recommended values for a printer-workflow environment?
-
- This probably accounts for why *some* people think that my
- web-gallery photos are too dark.
-
-
-Here is an example which looks dark to me:
-http://www.metalvortex.com/myphotos/antarctica/old_buildings.htm

a comparison...
http://vircen.com/rpd/index.cgi?mode...album=adjusted

gamma, lightness, contrast adjusted.

-Here is the histogram for it:
-http://www.theknack.net/img_histogram.png
-
-Not only is the distribution of pixels skewed left (i.e. dark), but the
-upper 15% (or so) of the color range is almost entirely unused.
-
-Personally I would always use the histogram and not rely on calibration
-and subjective adjustments (which are subject to ambient lighting on the
-monitor, etc). I have a bunch of old scans which are too bright because
-I relied on a faulty monitor.
-

--
Bob Crownfield

crownfield wrote:
In article , timeOday-
says...
-Kulvinder Singh Matharu wrote:
- I've just been calibrating my new printer but during the process I
- found recommendations that monitor luminance level should be between
- 90 cd/m2 and 100 cd/m2.
-
- I checked my (calibrated) monitor and it was something like 190
- cd/m2. So I've now adjusted my monitor to 100 cd/m2, gamma 2.2 and
- colour temp of 6500K.
-
- Everything looks gray and dim! So are these luminance levels really
- the recommended values for a printer-workflow environment?
-
- This probably accounts for why *some* people think that my
- web-gallery photos are too dark.
-
-
-Here is an example which looks dark to me:
-http://www.metalvortex.com/myphotos/antarctica/old_buildings.htm

a comparison...
http://vircen.com/rpd/index.cgi?mode...album=adjusted

gamma, lightness, contrast adjusted.

The snow loses all detail in this latter version, so perhaps you went a
bit overboard. Also, there is nice detail available in the lattice brick
in the top middle of the image.

Then again, any version might be "right" depending.....

--
John McWilliams

In article ,
says...
-crownfield wrote:
- In article , timeOday-
- says...
- -Kulvinder Singh Matharu wrote:
- - I've just been calibrating my new printer but during the process I
- - found recommendations that monitor luminance level should be between
- - 90 cd/m2 and 100 cd/m2.
- -
- - I checked my (calibrated) monitor and it was something like 190
- - cd/m2. So I've now adjusted my monitor to 100 cd/m2, gamma 2.2 and
- - colour temp of 6500K.
- -
- - Everything looks gray and dim! So are these luminance levels really
- - the recommended values for a printer-workflow environment?
- -
- - This probably accounts for why *some* people think that my
- - web-gallery photos are too dark.
- -
- -
- -Here is an example which looks dark to me:
- -http://www.metalvortex.com/myphotos/antarctica/old_buildings.htm
-
- a comparison...
- http://vircen.com/rpd/index.cgi?mode...album=adjusted
-
- gamma, lightness, contrast adjusted.
-
-The snow loses all detail in this latter version, so perhaps you went a
-bit overboard. Also, there is nice detail available in the lattice brick
-in the top middle of the image.

on my monitor, it does not. variations are visible.
the brightest areas are over, but only the very brightest.

beng a little more careful...
http://vircen.com/rpd/index.cgi?mode...album=adjusted
-
-Then again, any version might be "right" depending.....
-
-

--
Bob Crownfield