iSis UV Cut and No UV Cut Choices

[Bill W]

I use an Xrite 530 at press to qualify color (G7 process) and an iSis to measure press sheets. I have noticed for magenta that the UV Cut measurements from the iSis match better to my 530 than the no UV cut measurements. The no Cut adds about 2 to my b* (-5.0 for 530 vs -7.0 for iSis). For CYK, the no Cut is closer to the 530.

Does anyone see any problems with me combining both UV cut and no UV Cut for my final measurements? Number wise it seems this would be a benefit.

Thanks.

 

[meddington]

Do you have any profiling issues that you believe are caused by the difference in your instruments? My personal opinion is to keep separate the UV excluded and UV inlcuded data, using one or the other, but not mixing the two. And also keep separate process control instruments and profiling instruments. You also have to keep in mind that the ISis has a different illuminant than the 530 as well (LED vs incandescent). Even if there are differences between the two device's, this doesn't necessarily mean that there will be a problem with their functions...the ISis could generate completely accurate proofs based off its measurements that deviate from your handheld and vice versa, as long as both are consistent. In a perfect world, they would match, but...
I just wouldn't feel comfortable knowing part of my data excluded UV, while part did not.

 

[Bloodsaler]

The UV cut option will cut all the light come from 400nm below,and in other words,this will help making the result more accuracy

 

[meddington]

The UV cut option will cut all the light come from 400nm below,and in other words,this will help making the result more accuracy

Hi Blood,

Not sure I agree with you there. Whether or not UV exclusion will make the results more accurate depends on other factors, namely the OBA content of the press paper, the OBA content of the proof paper, the UV content of the illuminant viewing conditions and liekwise illuminant of the spectrophotometer. UV cut/ exclusion only addresses the excitation of the OBA by the spectro illuminant, without taking into account the light source's UV content. This "correction" could be hit or miss. IMO, there are more intelligent ways of correcting for OBA issues out there, but all are pretty specific to the viewing conditions and overall a visual analysis is required. However, with better definitions in the standards ISO 3664 (viewing conditions) and ISO 13655 (measurement conditions), the issue of OBA content is better addressed ...maybe not perfect, but better.

Moreover, if the goal is to match industry published data sets (Fogra39, TR003/5/6, etc), these were obtained from devices that did not exclude UV, and to blindly omit the effects of UV when targeting these data sets could be hit or miss as well . I use the term UV exclusion rather than "cut" because the Isis doesn't actually use a UV cut filter, but rather it has an LED illuminant that emits no UV, so there's nothing to cut.

Bill's original question addressed combining UV included and UV excluded data within the same data set. Perhaps this could be successful, but I'm a bit leery if its done for the sole purpose of instrument agreement. Frankendata could be harmless, or even beneficial, but could also lead to torch-bearing villagers.

 

[Bill W]

Greetings Michael,

My concern may be unjustified, but here it is. My magenta on press when measured with the 530 has a b* of (-4.5) - within the tolerances of +/- 2 or 12467-2. When the same patch is measure with my iSis the b* is now( -6) - not within the standards. When looking at the number for b* in my profile it will also look like I am out of spec. Am I too concerned about this - remember I am a prepress guy learning that press tolerances can be "somewhat" broad.

-Bill-

 

[gordo]

My concern may be unjustified, but here it is. My magenta on press when measured with the 530 has a b* of (-4.5) - within the tolerances of +/- 2 or 12467-2. When the same patch is measure with my iSis the b* is now( -6) - not within the standards. When looking at the number for b* in my profile it will also look like I am out of spec.

You have to be very careful when using different instruments and expecting them to agree. According to research into Spectrophotometer Device Agreement by Greg Radencic, PIA/GATF you could see a difference up to a about a Delta E of 7 between two different instruments measuring the same sample. Even two instruments that are the same model may not deliver the same values.

Here are a few issues that vex the process: Quality In Print: Top reasons why color instruments don't agree

This is also why XRite is introducing their XRGA standard.

best, gordon p

 

[meddington]

My magenta on press when measured with the 530 has a b* of (-4.5) - within the tolerances of +/- 2 or 12467-2. When the same patch is measure with my iSis the b* is now( -6) - not within the standards. When looking at the number for b* in my profile it will also look like I am out of spec.

Don't get me wrong, measurement agreement issues are a pretty big point of contention with me, so whether you should be concerned depends on a few things.

1. Is this data being used in an profiling process that is targeting data that was not measured with your Isis (i.e. you are matching Fogra39 or TR006 for example)? This could impart a visual deviation where it might not be warranted as the profile works to make things match on based on the Isis measurements. This is tricky to deal with and my first option would be to complete the profiling procedure, then based on visual analysis (comparing results with a trusted hard/softproof), tweak the profile. I hate tweaking profiles, but sometimes it is indeed necessary. Attempting to manipulate the data prior to profile generation can be done, but you don't always know where to stop and changing data patch by patch can introduce artifacts.

2. Are you sharing the profile or data with others who would judge all or part of it to another data set? This might give the other party cause for concern. If the target measures ok on your 530, you could share the hard copy with those who might be concerned for their own measurement. You could manipulate the data, but I would then be very wary of ever using it with any process connected with the original instrument.

You might consider having the same chart measured off your Isis and another device, compare the two in MeasureTool and look for trends (i.e. all pure magenta is shifted from one data set to another). This can give you an idea of how profound the issue is throughout the profile. Might also consider a target like the Lab-Ref or COLORef to see how your instrument compares to a defacto standard instrument on reference material...not sure how well ISis is supported with these yet.

 

[Bill W]

My "problem" is that the eye1 I use on my iO table returns numbers very close to my 530. Unfortunately, the iSis is "travelable", while my iO is not. Because we have 6 plants that I need to travel to, being able to travel with the iSis is saving us a good amount of money.

I will have to do more measurements on both the iO and the iSis and compare as you suggested.

 

[gordo]

Might also consider a target like the Lab-Ref or COLORef to see how your instrument compares to a defacto standard instrument on reference material...not sure how well ISis is supported with these yet.

Just a sidebar note: According to the Spectrophotometer Device Agreement study by Greg Radencic, PIA/GATF that I referenced earlier, instrument agreement was better on known reference materials (like the TRef) but not consistent when measuring common materials used in industry. The TRef readings were not a predictor of instrument agreement in production use.

best, gordon p

 

[meddington]

Just a sidebar note: According to the Spectrophotometer Device Agreement study by Greg Radencic, PIA/GATF that I referenced earlier, instrument agreement was better on known reference materials (like the TRef) but not consistent when measuring common materials used in industry. The TRef readings were not a predictor of instrument agreement in production use.

I agree with that. The LabRef and TRef are actually on real paper substrates with a laminate for protection. Between 2008 and 2009 they changed from a gloss to a matte laminate and my devices no longer "agreed" as well as they used to. After receiving a new gloss laminate card, "agreement" was again achieved. Still it was interesting to see how different devices fared on different material. I think larger differences would result from papers with increasing amounts of OBAs.

 

[gordo]

I think larger differences would result from papers with increasing amounts of OBAs.

The level of disagreement also varies as a result of paper finish (gloss or matte), varnish, coating, laminates, substrates.

We're measuring with rubber rulers

best, gordon p

 

[meddington]

Hey Bill,

One thing you can try...download the latest version of Colorport from Xrite.

X-Rite: Get exactly the color you need, every time, anywhere in the world.

This allows saving data to new XRGA standard. Maybe this will yeild values that more closely correlate with your handheld spectro. Try saving out the data as XRGA and again as Former Xrite or Gretag and see what kind of differences you get...then post back as we're all curious.

 

[Bill W]

Colorport - another challenge. I posted awhile ago that Colorport and my iSis did not play well together, as Colorport would not see my iSis, while MeasureTool would.

After the Holidays, if I can get Colorport and iSis to work together, I will measure and post results.

 

[meddington]

I posted awhile ago that Colorport and my iSis did not play well together, as Colorport would not see my iSis, while MeasureTool would.

Note that there is a new release of Colorport (2.0.1).
Might trying trashing v1.5.4, install 2.0.1, and cross your fingers.