Paper White M1 and M0 Differences

[Vernon Roberts]

We currently don't have an M1 spectrophotometer. I'm looking into changing our proofing paper for one with a white point closer to GRACoL 2013. I have received a sample roll from a vendor that is supposed to be the white point for 2013. When I measure its white point in M0, I get -3 in b*. I'm assuming that if the same paper was measured with M1, it would show up as b*-4 , picking up more information from the UV reflectance.

Does anyone have experience with the subtleties between a measurement in M0 mode and the same measurement in M1 mode?

I was told once that an M1 measurement would probably be -1 bluer than M0. I wanted to see if someone else can confirm this.

 

[mihas]

I was told once that an M1 measurement would probably be -1 bluer than M0. I wanted to see if someone else can confirm this.

I confirm.
Profiles Fogra51 beta were made with M0 and M1 filter. The difference in b* is equal to 2 on paper. I have also a i1 Pro2 spectrophotometer measurements of the same sample of different filters. The higher CIE whiteness (fluorescence, delta D65 Brightness UV/UVex) - the greater the difference between the filters.
There are at least two colorimetric methods to conversion from M1 to M0, M0 to M1, M0 to M2 etc.

 

[gordo]

We currently don't have an M1 spectrophotometer. I'm looking into changing our proofing paper for one with a white point closer to GRACoL 2013. I have received a sample roll from a vendor that is supposed to be the white point for 2013. When I measure its white point in M0, I get -3 in b*. I'm assuming that if the same paper was measured with M1, it would show up as b*-4 , picking up more information from the UV reflectance.

Does anyone have experience with the subtleties between a measurement in M0 mode and the same measurement in M1 mode?

I was told once that an M1 measurement would probably be -1 bluer than M0. I wanted to see if someone else can confirm this.

AFAIK there is no direct or specific relationship between M0 and M1 as far as b* values are concerned. That is because the UV content for the M0 instrument illuminant is unknown/not specified. So you can't say that if you get -3b* with an M0 instrument that you should expect to see a -4b* with an M1 instrument. The M1 instrument will show a higher value - but that value is not predictable.
I believe the current recommended practice is still to use GRACoL 2006 with an M0 instrument.

 

[benstarr]

Gordo is correct, there is no direct relationship between M0 and M1. It's quite possible that your -3 measured in M0 would be a -4 in M1 but it may not be. If you don't have an M1 measuring device stick with a M0 dataset or upgrade your measuring device if you want to work with M1 data.

 

[Vernon Roberts]

The house stock's b* is hitting -5 in M0. I'm trying to find an appropriate proofing paper. I'm not just sticking with GRACoL 2006, I'm working towards 2013, that's why I was just asking what people's experience was with M0 to M1 measurement and what they've seen. I'm sure we'll pick up the new iSis soon.

Also, I have sat in on a webinar with Don Hutcheson and have heard him directly answer someone question on whether they could use M0 with GRACoL 2013 data set. His response was that they recommended an M1 device, but it would be fine to use the dataset with an M0 device if that's all they had. I have also read in the documents provided with the profiles from IDE Alliance that if you are using an M0 device to stay with 2006.

So I've seen them contradict themselves. I just know my paper is blue and I need a blue proof. lol

 

[gordo]

The house stock's b* is hitting -5 in M0. I'm trying to find an appropriate proofing paper. I'm not just sticking with GRACoL 2006, I'm working towards 2013, that's why I was just asking what people's experience was with M0 to M1 measurement and what they've seen. I'm sure we'll pick up the new iSis soon.

Also, I have sat in on a webinar with Don Hutcheson and have heard him directly answer someone question on whether they could use M0 with GRACoL 2013 data set. His response was that they recommended an M1 device, but it would be fine to use the dataset with an M0 device if that's all they had. I have also read in the documents provided with the profiles from IDE Alliance that if you are using an M0 device to stay with 2006.

So I've seen them contradict themselves. I just know my paper is blue and I need a blue proof. lol

I'm not sure that I would go as far they're contradicting themselves - I think they're just saying that if you need to drive a screw in and all you have is a hammer then use the hammer.

What would have been helpful is to clearly explain what the issues are with using an M0 instrument with data/profiles derived from M1 readings and what, if anything, can be done to mitigate the issues. It would also have been helpful if the instrument manufacturers would have provided an upgrade path to convert an M0 to an M1 reader. "Let's see, a $250 upgrade vs a $6000 sale. Hmmm, which should we provide? Decisions, decisions." LOL

There is another significant difference between an M0 instrument and an M1 besides the UV content and that is the color temperature of the illuminant. M0 instruments have an illuminant with a color temperature of 2856 K, ± 100 K (Incandescent/Tungsten/Warm White (discontinuous spectral distribution) Fluorescent) while M1 instruments are required to match CIE Illuminant D50 the same light you might be using in your viewing booth.

I think that the best you can do is to select a proofing paper with an OBA content as close as possible to your production paper. Use a black light to help with this evaluation since a b* reading can be misleading.

Fun, fun, fun.

 

[mihas]

there is no direct or specific relationship between M0 and M1
The M1 instrument will show a higher value - but that value is not predictable.

Predict the difference between M0 and M1 allows a ratio CIE Whiteness to CIE Lightness.
Calculation of CIE Whiteness just enough based on the difference betwen UV-included and UV-excluded. Manufacturers of paper indicate CIE Whiteness.
Also ISO 11475 allows you to calculate fluorescence component other than CIE Whiteness. Ratio fluorescence component to CIE Lightness allows to link directly M0 to M1.

 

[benstarr]

What proofing system and paper are you working with? It's possible that the vendor has a 2013 dataset created in M1 that you can use without having to upgrade instruments.

From what I've see it can make a big difference if you profile with M0 vs M1. If you want a blue proof try to match to GRACoL 2013 in M0 you'll get a very blue proof, it may work for you but in my experience you won't be happy it will be too blue and not really match your press. However if your paper is measuring -5 in M0 it's likely somewhere between -7 and -9 in M1 so GRACoL 2013 may not even be blue enough for you.

Here's an example of the difference in M0 and M1 how it can't be predicted.

Proofing Paper (with OBA)
M0 b* -3.8
M1 b* -5.3
∆b* 1.5

Press Stocks
M0 b* -2.3
M1 b* -4.5
∆b* 2.2

M0 b* -4.5
M1 b* -7.5
∆b* 2.7

The proofing paper's b* in M0 is in between the two press stocks so you may expect the ∆b* to fall in between the stocks but no it's lower than either one.

 

[mihas]

how it can't be predicted.

∆b* really can not tell the difference. Papers cited in the examples contain diverse fluorescent components. ∆b* ignores that.

 

[Vernon Roberts]

Predict the difference between M0 and M1 allows a ratio CIE Whiteness to CIE Lightness.
Calculation of CIE Whiteness just enough based on the difference betwen UV-included and UV-excluded. Manufacturers of paper indicate CIE Whiteness.
Also ISO 11475 allows you to calculate fluorescence component other than CIE Whiteness. Ratio fluorescence component to CIE Lightness allows to link directly M0 to M1.

This is kinda confusing, but looks interesting. I will look into ISO 11475. I don't know how to look into ISOs. You have to purchase them right?

​What proofing system and paper are you working with? It's possible that the vendor has a 2013 dataset created in M1 that you can use without having to upgrade instruments.

From what I've see it can make a big difference if you profile with M0 vs M1. If you want a blue proof try to match to GRACoL 2013 in M0 you'll get a very blue proof, it may work for you but in my experience you won't be happy it will be too blue and not really match your press. However if your paper is measuring -5 in M0 it's likely somewhere between -7 and -9 in M1 so GRACoL 2013 may not even be blue enough for you.

Here's an example of the difference in M0 and M1 how it can't be predicted.

Proofing Paper (with OBA)
M0 b* -3.8
M1 b* -5.3
∆b* 1.5

Press Stocks
M0 b* -2.3
M1 b* -4.5
∆b* 2.2

M0 b* -4.5
M1 b* -7.5
∆b* 2.7

The proofing paper's b* in M0 is in between the two press stocks so you may expect the ∆b* to fall in between the stocks but no it's lower than either one.

This is what I was interested in seeing. I'm on Heidelberg Prinect System with Epson 9900, using Epson branded proofing paper semimatte.

What would have been helpful is to clearly explain what the issues are with using an M0 instrument with data/profiles derived from M1 readings and what, if anything, can be done to mitigate the issues. It would also have been helpful if the instrument manufacturers would have provided an upgrade path to convert an M0 to an M1 reader. "Let's see, a $250 upgrade vs a $6000 sale. Hmmm, which should we provide? Decisions, decisions." LOL

Epson does have an M1 upgrade for its Spectroproofer on the 9900, instead of buying a whole new unit. I had to wait until Prinect 2016 for M1 RIP support.

 

[arossetti]

it may work for you but in my experience you won't be happy it will be too blue and not really match your press. However if your paper is measuring -5 in M0 it's likely somewhere between -7 and -9 in M1 so GRACoL 2013 may not even be blue enough for you.

How well does the substrate relativity calculator work for something like this?

 

[Vernon Roberts]

How well does the substrate relativity calculator work for something like this?

I used it for our target process standard, so I get a good score on the color report for paper white, it's built into Color Toolbox. But taking the resulting dataset and proofing to it absolute, the proof looked too blue like benstarr said. The Spectroproofer verified it was correct, but visually, it didn't look so. I feel like I don't know what I'm doing sometimes, and perfect color management is just a fancy.

 

[Morgan Manning]

Vernon, Have you considered talking to Jim Ross on the support desk? Can't think of a better guy to talk to about this within the Heidelberg ecosystem.

 

[benstarr]

How well does the substrate relativity calculator work for something like this?

I'n my experience I haven't seen great results (spreadsheet from IDEAliance, I know there are others out there) with the substrate calculator going to much bluer stocks, like what's being discussed here.

 

[benstarr]

Mihas, thanks for the information.

I'll revise my statement to say that M1 Lab values cannot be predicted from M0 Lab values, which is the most easily accessible information for most people. Would that be correct?

 

[mihas]

This is kinda confusing, but looks interesting. I will look into ISO 11475. I don't know how to look into ISOs. You have to purchase them right?

Yes. Or you can find the quote, as an example lesson here: http://rudtp.pp.ru/pdf/Brightness_Wh...resc_comp.xlsx

I disappoint spectrophotometers sellers, but M1 to Mo and M0 to M1 conversion of colorimetric Characterization Data (reference or sample) is possible.
Here are some ways to conversion between the spectral filters:

1) Argyll CMS (spec2cie modul). For example - M0 to M1 conversion bat-file:

::start
Set argpath=%~dp0
Start /b /wait %argpath%bin\txt2ti3.exe %1 "%~dpn1"
start /b /wait %argpath%bin\spec2cie.exe -v -i D50 -f M1 -o 1931_2 "%~dpn1".ti3 "%~dpn1"_M1.ti3
start /b /wait %argpath%bin\spec2cie.exe -v -i D50 -o 1931_2 "%~dpn1"_M1.ti3 "%~dpn1"_M1_CIE.ti3
Del "%~dpn1".ti3
Del "%~dpn1"_M1.ti3
::End
pause

M1 to M0:

::start
Set argpath=%~dp0
Start /b /wait %argpath%bin\txt2ti3.exe %1 "%~dpn1"
start /b /wait %argpath%bin\spec2cie.exe -v -I D50 -i D50 -f M0 -o 1931_2 "%~dpn1".ti3 "%~dpn1"_M0.ti3
start /b /wait %argpath%bin\spec2cie.exe -v -i D50 -o 1931_2 "%~dpn1"_M0.ti3 "%~dpn1"_M0_CIE.ti3
Del "%~dpn1".ti3
Del "%~dpn1"_M0.ti3
::End
pause

2) A more accurate tool - http://rudtp.pp.ru/spectralcalc_en.php
Here works my heuristic algorithm. It is taken into account fluorescence component as far as possible with the single filter.

3) In addition to the spectral conversion there is a McDowell 2005 colorimetric method. He was published with formulas entirely in ISO 12647-2:2013 as a Tristimulus Correction Method. It allows you to change the white point and followed her all the color coordinates on this substrate. McDowell 2005 method is implemented in Heidelberg Color Tool and in Spectral Calculator http://rudtp.pp.ru/spectralcalc_en.php

 

[Vernon Roberts]

Vernon, Have you considered talking to Jim Ross on the support desk? Can't think of a better guy to talk to about this within the Heidelberg ecosystem.

I have talked his ear off, lol. He is the one that suggested that I might see -1 from M0 to M1. The Colortool Box + Jim Ross has helped me make big gains in color management. The software does all the work.

 

[mihas]

Mihas, thanks for the information.
I'll revise my statement to say that M1 Lab values cannot be predicted from M0 Lab values, which is the most easily accessible information for most people. Would that be correct?

I think yes. It is better than use appropriate datasets: instrument M0 and dataset M1 or vice versa.

 

[mihas]

Before I forget. Paper White M1 and M0 Difference = 0 if fluorescence component of paper = 0.

 

[gordo]

Some words from idealliance on the topic:

QUOTE:

The main difference between the GRACoL 2006 and the new GRACoL 2013 dataset is the white point. The target white point for GRACoL 2006 (95 L*, 0 a*, -2 b*) was based on the legacy ISO 12647-2 standard, which in turn was based on typical commercial printing stocks available in 1994. The GRACoL 2013 white point (95, 1, -4) is slightly bluer, in line with today’s typical commercial stocks.
Because the two data sets are nearly identical, most print buyers and designers should see little visible difference. In fact the differences are smaller than the typical variation between two good offset press runs.
The really good news is that legacy image files and proofs produced using GRACoL 2006 should not need to be adjusted or converted for printing or proofing to GRACoL 2013 (and vice-versa), except in rare situations.
Calibration and verification of GRACoL 2013 requires use of M1 capable instruments. This means that use of GRACoL 2013 in production cannot be done unless you have M1 capable instruments.
IDEAlliance recommends the following transition:
• If you only have M0 measurement devices keep using GRACoL 2006.
• If you have M1 instruments you may choose to move to GRACoL 2013. In the GRACoL 2013 workflow M1 instruments are to be used for calibration and verification. (M0 instruments can still be used for process control.)

END QUOTE