Getting yellow cast on prints

[fellmonster]

Hi Pauly,

we have a very similar setup (Arizona, Barbieri – lot's of Backlit). We have seen yellow casts most of the time then OBAs (optical brigtheners) cross paths with measuring standards M0/M1/M2. Recently we had to do real "shoot out" to figure out what was measuring best and – more important – what was most appealing. For sure there are standards like Gracol or Fogra which MUST be profiled and measures with a specific setting most POS/LFP applications will require a "looks great" approach.

What I would suggest: make three profiles on the same media one with M0/M1/M2 each (as the Spectro LFP and the SpectroPad can only do M0/M2 (plus M3 - but i doubt this will help) – I would suggest to use the X-Rite's i1Pro2) and compare results using absolute colorimetric rendering.

We were very surprised to see a STRONG yellow cast with M1, a slight (but preferred/wanted) cast with M1 and no cast with M0. As always: you mileage may vary and we definitely had quite a bit of optical brigtheners in our media.

One more thing: you wrote BLUEback – long time ago we were fighting with strange casts on out profiles/prints only do discover that by whatever reason we forgot to measure in reflective. We used transmissive mode and caught the cast from the blue back of the media...

Cheers, Lorenz

Postscriptum: sometime strange things happen . As part of a students examination we have discussed and tested the influence of measuring modes last week. Using some really cheap offset paper on a digital press (xerox versant 80 with an efi controller) we have done M0/M1/M2 with an i1Pro2 while all other parameters stay the same. Guess the two pictures speak for themself...

 

[pauly92]

Hi Lorenz,

Interesting results. I'l have to give that a go.

measuring transmissive with blueback will give horrid results, not just a cast! But i haven't made that mistake yet.

I can only do M0 and M1. i dont have a uv cut filter for any of my devices.

The blueback is a really white media, has no yellow cast like my outdoor paper i've used in the past.



I've just been in and out the office - meetings ect haven't had the chance to do any profiling - Hopefully some time next week i'll have time.

 

[fellmonster]

Hi Pauly, I really doubt that (any) Blueback packer is REALLY white. It is white – sort of . We have used different brands and different types in the past – most of them we having LOTS of OBA (optical brighteners) – actually they were more BLUE than white. At least they were BLUE to our measuring devices. Would really give it a try with different measuring modes. PLUS, takin the tremendous amount of cash you spent on hardware into account: I would suggest to get a hand on a current i1Pro2 which will enable you to test M0/M1/M2 really good (and quick). Best, Lorenz

 

[AngusPady]

Anytime I see a Yellow cast it is usually because the device is seeing blue. i.e. Big optical brighteners / Fluorescence. Is the B value of the "problem media" in the -5 or higher/lower... Usually with media's with a -4 to a -8 I get a strong yellow cast when using the M0 filer. Try M1 to see if that helps.

 

[pauly92]

Sorry for being a little absent. Huge workload + holidays..

I'll be doing some measuring this week, well tomorrow.

The LFP only uses M0. The Spectropad uses M0 & M1
I use the LFP in most cases, easier & most accurate.

I'll do some tests with M0 and M1. Also on different medias as i have Aluminium Composite sheets from a previous job sitting here. Which is a satin finish.

I'll also be doing this with the suggestions of mike.
The ink limits will be pulled to max chroma, not the automated settings onyx likes to do.

 

[pauly92]

So i've used a few different types of medias just to see where it's all at.

I've only been using the LFP,
I've created all new ink restrictions, following this method. https://www.onyxgfx.com/wp-content/uploads/2015/09/Advanced-Ink-Restrictions.pdf
They all end up getting restricted roughly in the same levels (70-80%) All end up quite linear in the linearisation/calibration

I've used the IT8 7.4 patch from Barbieri (CMYK_17) http://www.profile-xpert.com/targets/index.htm

After reading it though gateway, it exports it as a .mxf which includes the reference and spectral data.

I'll drag that file to i1 Profiler which it recognises the patches ect once dragged in. (i1 Profiler also has it's own IT8 7.4 & ECI2002 patches which i have yet to use)

Settings i use my most preferred black ink settings. I keep everything else as 0

I'll make the profile...
I'll do a soft proof in photoshop. Near spot on with the dark reds, browns & Blues.

Go to print, I'm back with the same thing, a slight yellow cast. Getting similar results across everything.
I tried it on some PVC sheets (L96.7 a0.7 b-1.8) & Aluminium which is a satin finish (L90.1 a0.9 b-1.8)

Everything is very similar across all medias. Same print, comes out identical on both PVC and Aluminium. (small variances due to the nature of the material) but same yellow cast/warm print.

I have some white paper with some OBA (L93.8 a1.7 b-6.6) But yet to try it.

Note: The cast is though out the print, including solids.
Also targets are printed with "all icc off" and i do check them in the colour checker section with the tool that there's no profile on them.

 

[pauly92]

Another update.

I've been using the IT8 7.4 charts from the i1 Profiler.
I turned off the largest dots on yellow and also all inks.
I've tried different medias.
And still getting consistent results.

Soft proof is real good.
When i print it, Same deal. cast is there.

I even changed the embedded ICC on the image to my created one. loaded it into onyx. put on the profile i was using. Turned all ICC off.
Printed the same as usual.

I'm pretty stumped at the moment. If i pull back any more ink, I loose a lot of gamut and causes my prints to be quite bad.
I built a profile with the onyx engine, With the biggest dots off, it appears to have came out better slightly. But still struggles to get the punch in the reds. and the browns have to much yellow in it.

 

[pauly92]

A few tweaks here and there and everything is solved. A bit in the ink restrictions and some in the ICC profile itself.

Thanks to all of those for suggestions.