Datacolor User to User Support Group.

So we've all calibrated our screens and our printers.  There seems to
be plenty of software out there to calibrate our scanners too.  There
does not however seem to be much out there for calibrating our cameras.

This is tricky though.  The image captured is somewhat subjective. 
Normaly I shoot with a colorchecker chart just as a reference.  I did
however run across the ACR Calibrator script which is supposed to help
calibrate Adobe Camera Raw when importing images.  After some
experimentation it has turned out that under controlled lighting
conditions even the lens used has a subtle effect on the colors
captured by my camera.  Unfortunately after measuring my chart with
PFP I noticed that the colors published don't (cough) exactly match
... which isnt unusual with large batch production methods.

How careful are the rest of you with color when capturing images? 
Would it be accurate to say that, like printing, having the colors
captured being pleasing is not nearly as important as being deadly
accurate?

This is something that can be a blessing if used with a good deal of
thought given it. One benefit of camera calibration is that it will
also calibrate the meter used to determine exposure. I always use a
handheld incident meter for exposure determination and afterward use a
spot meter to see where the values in the scene fall and make needed
adjustments for the shadows. I have found the incident favors the
highlights and I don't have many blown out values there. My procedure
for calibrating my camera goes like this.

1: I place my color checker in direct sunlght evenly lit.
2: I take a meter reading and make my first exposure 2 full stops
below what the camera's ISO rating is. My Nikon D70 touts an ISO of
200, so my initial exposure is at iso 50. I then make exposures at 1/3
stop less exposure; corresponding to iso 50-64-80-100-125-160-200.
3. Opening each exposure one at a time in Camera Raw and putting all
settings on ZERO except the color temp, which I leave at the 5500
temp; I open the image in Photoshop. I use the color eyedropper and
read the values on the bottom squares to see if the fourth one from
the left is reading 127,127,127 which indicates middle gray (Zone
V-five for Zone System users). The exposure that gives me that reading
on that square is the film speed I dial into my handheld incident
meter and also my spot meter so both meters will result with identical
exposures for evaluation of the scene.
4. I follow the instructions with ACR calibrator (either the one from
Thomas Fors or Rags Gardner) and run the script on the marked section
of the color checker image to completion. The resulting numbers I copy
down on paper and on the next image I open and make them the default
RAW camera settings for all following uses.
5. I also have my monitor calibrated and use Colorvision's PrintFix
PRO for creating printer profiles for my ink/paper combinations.

For Black and white prints I feel it is absolutely necessary to have a
good color image to start with because that way the tonal ranges will
be more accurate in starting out.

This is just my personal feelings in this matter. Your reasoning and
uses will undoubtedly be different, but that is what makes photography
so personal. We are all precious and unique.

Your friend in Photography,

Johnny Eades




--- In colorvision_group@yahoogroups.com, "Tom" <ttrostel@...> wrote:

>
> So we've all calibrated our screens and our printers.  There seems to
> be plenty of software out there to calibrate our scanners too.  There
> does not however seem to be much out there for calibrating our cameras.
> 
> This is tricky though.  The image captured is somewhat subjective. 
> Normaly I shoot with a colorchecker chart just as a reference.  I did
> however run across the ACR Calibrator script which is supposed to help
> calibrate Adobe Camera Raw when importing images.  After some
> experimentation it has turned out that under controlled lighting
> conditions even the lens used has a subtle effect on the colors
> captured by my camera.  Unfortunately after measuring my chart with
> PFP I noticed that the colors published don't (cough) exactly match
> ... which isnt unusual with large batch production methods.
> 
> How careful are the rest of you with color when capturing images? 
> Would it be accurate to say that, like printing, having the colors
> captured being pleasing is not nearly as important as being deadly
> accurate?
>

Hi

In general I don't calibrate my 1Ds. The exceptions are where I'm doing a whole lot
of product shots under constant conditions -and- the client has the savvy to
appreciate the difference. Note that second point which means not too often ;-)

I modified the ACR script to use measured values from my colorchecker SG card.
I do use the Eye One Match software sometimes when shooting in odd lighting
- there are examples of both approaches in the article at:
http://www.northlight-images.co.uk/article_pages/eye_one_photo_SG.html

bye for now

Keith Cooper
Keith's Photodiary in Colorado
http://www.northlight-images.co.uk/article_pages/landscape_photo_travels.html

--- In colorvision_group@yahoogroups.com, "Keith Cooper" <yg_1@...> wrote:
>
> Hi
> 
> In general I don't calibrate my 1Ds. The exceptions are where I'm
doing a whole lot
> of product shots under constant conditions -and- the client has the
savvy to
> appreciate the difference. Note that second point which means not
too often ;-)

I suppose that was the original question.  In general we don't
remember colors exactly as we see them.  Many of us remember scenes as
being more colorful than they really are for one.  Technicaly for
middle of the day shots my family remembers the scene with a warmer
color temperature and a slightly different tint than it actually was
... white balance nonwithstanding.

Other than tint and white balance differences which can be measured
even from lens to lens is there a compelling value to having extremely
accurate color on the input end?  One could almost argue, as in the
case of images taken in the harsh blue light of noon day sun,
shouldn't match reality at all.  People don't buy images which are
mirrors of reality.

Is there then a compelling reason to use strict color calibration on
the input end other than fixing extreme lighting conditions?

One of the points I was trying to make might have been overlooked
because this is primarily a Color newsgroup. I referred to the need
for reliable color rendition so that in working up an image for black
and white that the tones of the color would be more accurately related
to the corresponding tones of grayscale tones. There is a very good
website www.zuberphotographics.com that speaks in detail of this.

Your friend in Photography,

Johnny Eades



--- In colorvision_group@yahoogroups.com, CDTobie@... wrote:
>
> 
> In a message dated 5/11/06 9:03:28 AM, ttrostel@... writes:
> 
> 
> > Is there then a compelling reason to use strict color calibration on
> > the input end other than fixing extreme lighting conditions?
> > 
> 
> For certain forms of technical documentation (museums documenting their 
> artwork, for example) the most literal and exact match is wanted.
More often its a 
> matter of trade colors for a company, where "fixing" them after the
fact is 
> more practical, and generally would still be required even if a
profile was 
> used. For fine art photography, its artistic intent, not camera
output, that 
> matters, and correcting visually on screen with a calibrated monitor
is the usual 
> choice, though getting the best built in profiles, and the mosts
convenient 

> tools, in your RAW convertor make this process as painless as possible.
> 
> C. David Tobie
> Product Technology Manager
> ColorVision Business Unit
> Datacolor Inc.
> CDTobie@...
> www.colorvision.com
>

On 5/11/06 6:29 PM, "Johnny Eades" <jeades1@...> wrote:

> One of the points I was trying to make might have been overlooked
> because this is primarily a Color newsgroup. I referred to the need
> for reliable color rendition so that in working up an image for black
> and white that the tones of the color would be more accurately related
> to the corresponding tones of grayscale tones. There is a very good
> website www.zuberphotographics.com that speaks in detail of this.

Johnny,

I just uploaded to the File section a PDF titled:

Photoshop_Channel_Mixer_Settings_B_W.pdf 

The PDF has Cannel Mixer values that replicate many popular B&W film responses.

Best,

John V.

--- In colorvision_group@yahoogroups.com, "John Vitollo" <jvlist@...> 
> Johnny,
> 
> I just uploaded to the File section a PDF titled:
> 
> Photoshop_Channel_Mixer_Settings_B_W.pdf 
> 
> The PDF has Cannel Mixer values that replicate many popular B&W film
responses.
> 
> Best,
> 
> John V.
>


Heh ... I still have a soft spot for Tri-X Pan ... 100 ASA though the
faster speed grain was sort of interesting.

I'm not sure what those exact numbers mean, because they will depend
on what light the film has been shot in - and of course they would
also be changed by the use of filters. However, there is an
interesting feature: for Ilford and Agfa films, the B number is lower
than the G number. For Kodak films it is the other way round. 

Why is this? Maybe it is because the mixer settings have been
estimated from the published spectral response curves. Agfa and Ilford
published wedge spectrograms made in tungsten light (2850 K for
Ilford), at one density only. Kodak publish equal energy response
curves at two densities. This makes Kodak films look as if the blue
response is higher than the green response, and vice-versa for Ilford
and Agfa. It's just the way the spectral sensitivity is shown. You
need to compare wedge spectrograms or equal energy curves. You can't
compare one with the other - that is misleading, and whoever wrote
those channel mixer numbers appears to have been misled. Not that it
matters one bit for imitating B&W film digitally, but it does matter
if you wish to understand the differences between real B&W films.

Best,
Helen

 

--- In colorvision_group@yahoogroups.com, "John Vitollo" <jvlist@...>
wrote:

> Johnny,
> 
> I just uploaded to the File section a PDF titled:
> 
> Photoshop_Channel_Mixer_Settings_B_W.pdf 
> 
> The PDF has Cannel Mixer values that replicate many popular B&W film
responses.

> 
> Best,
> 
> John V.
>

>> I just uploaded to the File section a PDF titled:
>> 
>> Photoshop_Channel_Mixer_Settings_B_W.pdf 
>> 
>> The PDF has Cannel Mixer values that replicate many popular B&W film
>> responses.

On 5/12/06 8:47 AM, "Helen Bach" <helenbach@...> wrote:

> I'm not sure what those exact numbers mean, because they will depend
> on what light the film has been shot in - ..... whoever wrote
> those channel mixer numbers appears to have been misled. Not that it
> matters one bit for imitating B&W film digitally, but it does matter
> if you wish to understand the differences between real B&W films.
> Helen


Good points on what light the film has been shot.

The numbers were taken from a photo.net post a while ago...I just packaged it in a nice 
PDF. Just how accurate are the numbers/films and who worked on them is anyone's guess. 

But at the very least the PDF shows Photoshop users is there is more than just converting 
RGB to Grayscale.