Discussion:
Question about M1 measurements
Graeme Gill
2014-08-26 06:19:49 UTC
Permalink
I'm wondering, what is your opinion of the M1 measurement standard. Does it magically
solve all OBA problems, as Idealliance seems to indicate, or is this just another "FWA myth"?
Hi,
nothing can magically solve all OBA problems, short of using paper
that is free of FWA/OBE.

The bottom line is that paper that uses OBE/FWA has an appearance that
varies with the UV content of the illuminant it is viewed under. In color managing
such media, it can therefore help if the the nature of the illuminant is known,
and the color measurement system is able to anticipate the appearance of
what is being measured under the actual viewing illuminant.

As I understand M0/M1/M2, it is an attempt to address the fact that
up till now, the impact of the UV content of the illuminant on
FWA/OBE has largely been ignored and uncontrolled in the graphic arts context.
Both "D50" viewing booths and "D50" instrument measurements
haven't actually taken any care of whether the "D50" matches
the D50 UV content. Introducing the M0/M1/M2 nomenclature
and standard at least makes it clear what's going on in the UV region
when dealing with graphic arts D50 measurements.

As to it's actual practical impact, I suspect that in the short
term, confusion and lack of understanding will make things
a bit worse.

In the long term the drawback of using something like M1 conditions
for everything, is that it is more sensitive to various inaccuracies
and variations than (say) using FWA/OBE free paper, or using M0 or M2 measurements.
I would expect that differences in implementation of the simulated M1 measurements
and approximated M1 illuminants together with any lack of uniformity in FWA/OBE
content of the paper would lead to slightly wider margins of error when comparing
measurements from different sources, and appearance in different viewing booths.

Another source of confusion is that M1 is not automatically the
right choice for every application. M1 is suitable for viewing
in M1 viewing booths, or in direct sunlight. If the final output is
to be viewed under a lower UV illuminant such as normal fluorescent tubes,
incandescent light, prints under glass, or glass filtered sunlight, then
it may not be as appropriate as M0 or M2 conditions. These lower UV
conditions are quite common, because it's generally not good
for humans to be exposed to D50 levels of UV all the time.

Of course, if the actual viewing situation is fixed and measurable,
then using something like ArgyllCMS's illumread in combination
with FWA compensation creates a specific correction, rather than
being stuck with discrete choices such as M0/M1/M2.

Graeme Gill.
Roger Breton
2014-08-26 12:18:35 UTC
Permalink
Graeme,
Of course, if the actual viewing situation is fixed and measurable, then using something like ArgyllCMS's illumread in combination with FWA
compensation creates a specific correction, rather than being stuck with discrete choices such as M0/M1/M2.
I have not looked at your code but do you use a "fixed" correction or a "dynamic" correction in argyll FWA compensation?

Best / Roger
Graeme Gill
2014-08-26 13:07:57 UTC
Permalink
Post by Roger Breton
I have not looked at your code but do you use a "fixed" correction or a "dynamic"
correction in argyll FWA compensation?
Sorry Roger, it's hard to answer such a question. Every reading is different, so by
definition nothing is fixed.

If you're imagining something like a static shift in b*, then no - far from it.
It's a spectral correction model of the way the light interacts with the media,
including the FWA response.

Graeme Gill.
Roger Breton
2014-08-26 13:15:54 UTC
Permalink
So corrections are "illuminant-dependent" then?

/ Roger

-----Original Message-----
From: argyllcms-bounce-***@public.gmane.org [mailto:argyllcms-bounce-***@public.gmane.org] On Behalf Of Graeme Gill
Sent: 26 août 2014 09:08
To: argyllcms-***@public.gmane.org
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
I have not looked at your code but do you use a "fixed" correction or a "dynamic"
correction in argyll FWA compensation?
Sorry Roger, it's hard to answer such a question. Every reading is different, so by definition nothing is fixed.

If you're imagining something like a static shift in b*, then no - far from it.
It's a spectral correction model of the way the light interacts with the media, including the FWA response.

Graeme Gill.
Graeme Gill
2014-08-26 13:22:21 UTC
Permalink
Post by Roger Breton
So corrections are "illuminant-dependent" then?
Dependent on the target illuminant, yes of course - the whole point
is to correct the readings for the target illuminant.

Graeme Gill.
Roger Breton
2014-08-26 17:44:55 UTC
Permalink
Isn't the "target" illuminant always D50 in an ICC profile?

Roger

-----Original Message-----
From: argyllcms-bounce-***@public.gmane.org [mailto:argyllcms-bounce-***@public.gmane.org] On Behalf Of Graeme Gill
Sent: 26 août 2014 09:22
To: argyllcms-***@public.gmane.org
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
So corrections are "illuminant-dependent" then?
Dependent on the target illuminant, yes of course - the whole point is to correct the readings for the target illuminant.

Graeme Gill.
Graeme Gill
2014-08-26 23:43:40 UTC
Permalink
Post by Roger Breton
Isn't the "target" illuminant always D50 in an ICC profile?
For interoperable ICC profiles, yes. But note that traditionally
the "D50" that most instruments produce is M0 based, not M1,
so mixing M0 and M1 based D50 PCS profiles will have problems.

For the business of getting better visual matches in the face
of FWA/OBE, then it's desirable to create non-standard
ICC profiles for the actual viewing illuminant, rather than D50.

Graeme Gill.
Roger Breton
2014-08-27 00:19:46 UTC
Permalink
I love these discussions, Graeme und Gerard :-)

/ Roger

-----Original Message-----
From: argyllcms-bounce-***@public.gmane.org [mailto:argyllcms-bounce-***@public.gmane.org] On Behalf Of Graeme Gill
Sent: 26 août 2014 19:44
To: argyllcms-***@public.gmane.org
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
Isn't the "target" illuminant always D50 in an ICC profile?
For interoperable ICC profiles, yes. But note that traditionally the "D50" that most instruments produce is M0 based, not M1, so mixing M0 and M1 based D50 PCS profiles will have problems.

For the business of getting better visual matches in the face of FWA/OBE, then it's desirable to create non-standard ICC profiles for the actual viewing illuminant, rather than D50.

Graeme Gill.
Giacomo Catenazzi
2014-08-27 08:40:23 UTC
Permalink
Note: color adaption will do the rest for magazines (also thanks the many
white pages).
Post by Roger Breton
I love these discussions, Graeme und Gerard :-)
/ Roger
-----Original Message-----
Sent: 26 août 2014 19:44
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
Isn't the "target" illuminant always D50 in an ICC profile?
For interoperable ICC profiles, yes. But note that traditionally the "D50"
that most instruments produce is M0 based, not M1, so mixing M0 and M1
based D50 PCS profiles will have problems.
For the business of getting better visual matches in the face of FWA/OBE,
then it's desirable to create non-standard ICC profiles for the actual
viewing illuminant, rather than D50.
Graeme Gill.
Gerhard Fuernkranz
2014-08-26 22:55:59 UTC
Permalink
Post by Roger Breton
Isn't the "target" illuminant always D50 in an ICC profile?
The virtual illuminant of the PCS is always D50 by definition, but the PCS isn't the color space of your print, but a mathematically defined virtual one.

The desired viewing light source for your print can be whatever you want (incandescent lamp, fluorescent lamp, etc.), and the point is to build a profile for the print under the actual viewing light source. If you want to view your print under D50 light, then you are of course free to choose D50 as viewing light source.

Print standards are a different story. They define standardized conditions, whose assumptions may or may not match reality. I guess there are more people around who read a magazine under incandescent or fluorescent light, than people reading their magazine in a D50 viewing booth ;-) But admittedly it would not be practical either to sell different variants of the same magazine, each one intended for a different viewing light source. So one needs a compromise, even if it is not perfect, and the standardized compromise is just D50. If you are working in the print business, then you may be forced to comply with standards, regardless whether they are optimal in a particular scenario, or not. Your personal, individual solution is not bound to any standards, though, and you can go the pure scientific way.

Btw: The FWA compensation becomes necessary, if the spectrometer's light source does not match the desired viewing light source for which the profile is built. If the spectrometer uses an incandescent lamp for acquiring the readings, and if the desired viewing light source for the print were the same kind of incandescent light as well, then the FWA compensation would not make a difference.

Regards,
Gerhard
Post by Roger Breton
Roger
-----Original Message-----
Sent: 26 août 2014 09:22
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
So corrections are "illuminant-dependent" then?
Dependent on the target illuminant, yes of course - the whole point is to correct the readings for the target illuminant.
Graeme Gill.
Claas Bickeböller
2014-08-29 06:34:06 UTC
Permalink
Dear all,

as already stated the "corrections" (corrected is the wrong UV excitation of the instrument's light source compared to the desired reference illuminant) have to be illuminant dependent.

I tried to outline it here:
http://www.konicaminolta.eu/en/measuring-instruments/learning-centre/colour-measurement/colour/to-see-what-you-dont-see.html

In addition I wrote an article of possible ways to achieve M1 where I also clearly stated that M1 is perfect for viewing conditions that closely match D50.
For environments that have a UV content differing from D50 (like typical older "D50"-tubes or warehouse lighting), for a perfect correlation between measurements and visual appraisal, M1 is not the best choice.

http://www.konicaminolta.eu/en/measuring-instruments/learning-centre/colour-measurement/colour/iso13655-demystified.html

M0 is also not a good choice except the UV excitation of your instrument matches your viewing condition. But you can never know that unless you measured the light source of your particular instrument. Please note that M0 does _not_ mean tungsten. An instrument with a tungsten lamp as light source is M0, but M0 covers all illuminations incl. M2 and M1 as the UV content is not defined and the demand for matching a CCT of 2856K is only a should.
For tungsten based M0 instruments also keep in mind that the UV content that is emitted from a tungsten lamp changes over time. So for all "FWA-compensations" that assume a certain UV content of a generalized instrument's light source it is not unlikely that they do not work correctly with your particular instrument.
Also all methods that try to derive the FWA content from a single UV excitation will fail for papers that
a) use a lot of shading agents in combination with FWAs
b) do not show the typical strong "FWA-peak" as their "reflective color" (so without FWA emission -> UV-Cut) is yellowish
c) maybe there are other circumstances where they fail

All these are the reasons why we (in our VFS) detect the amount of FWAs using two different UV-excitations and are also able to make use of what we call a "User-Illuminant".

M2 is a good choice if you are in a UV-free environment.

I hope that helps to clarify the "myth".

Best regards

Claas Bickeböller
Product Manager Graphic Arts
Konica Minolta Sensing
Post by Roger Breton
So corrections are "illuminant-dependent" then?
/ Roger
-----Original Message-----
Sent: 26 août 2014 09:08
Subject: [argyllcms] Re: Fwd: Question about M1 measurements
Post by Roger Breton
I have not looked at your code but do you use a "fixed" correction or a "dynamic"
correction in argyll FWA compensation?
Sorry Roger, it's hard to answer such a question. Every reading is different, so by definition nothing is fixed.
If you're imagining something like a static shift in b*, then no - far from it.
It's a spectral correction model of the way the light interacts with the media, including the FWA response.
Graeme Gill.
Loading...