Pages: [1]   Go Down

Author Topic: Barbieri Spectro LFP problem  (Read 4366 times)

Stefan Fiedler

  • Newbie
  • *
  • Offline Offline
  • Posts: 12
Barbieri Spectro LFP problem
« on: February 14, 2011, 06:12:25 am »

We find that CMYK profiles for matte inkjet fine art papers (e.g. Hahnemuehle Photo Rag) measured with our new Barbieri Spectro LFP v3 result in less shadow information (smaller shadow gamut) - something like L=22 and shadow posterization effects - compared to profiles measured with the old XRite DTP20 Elite or the standard XRite EyeOne Pro using the exact same profiling software solutions (Monaco Profiler, Ergosoft ColorGPS, basICColor print3) and identical settings which deliver L=15.

With glossy photographic inkjet papers the L shadow-values are almost identical with all three spectrophotometers but we still see posterization effects in critical shadow areas (e.g. dark skintone shadows) with CMYK profiles created with the Barbieri Spectro LFP compared to other spectrophotometers - but again with identical software and settings.

Any ideas why?

Regards,
Stefan
Logged

stretchdcanvas

  • Guest
Re: Barbieri Spectro LFP problem
« Reply #1 on: February 14, 2011, 09:58:20 am »

My first thought regarding the spectro is "don't kill the messenger".

Take a look at how you are building the profile.

Also look at ink reductions for each channel. 

What is your K limit?

Just curious, are you setting your total ink limit in your profile or your RIP?

 
Logged

gromit

  • Full Member
  • ***
  • Offline Offline
  • Posts: 133
Re: Barbieri Spectro LFP problem
« Reply #2 on: February 14, 2011, 05:40:34 pm »

I've been toying with the idea of getting one of these. Since you're getting good/comparable readings on glossy, but not on matte this leads me to think that maybe you're not using the polarization filters or using them correctly.

I don't have firsthand experience with this device.
Logged

Stefan Fiedler

  • Newbie
  • *
  • Offline Offline
  • Posts: 12
Re: Barbieri Spectro LFP problem
« Reply #3 on: February 15, 2011, 03:51:37 am »

The only parameter we have changed for our comparisons is the new spectrophotometer - the Barbieri Spectro LFP v3. We have not changed anything else (profiling software and settings, RIP, printer, inks, paper etc.). We are and always have been using the Ergosoft PosterPrint RIPs Epson 11880 HTM driver (halftone module) because of ist superior dithering etc. Therefore we are setting the ink limits in the profile. No, we are not using a polarization filter. We have always been converting all RGB images to printer CMYK in Photoshop with the relative rendering intent and black point compensation.

Regards,
Stefan
Logged

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Barbieri Spectro LFP problem
« Reply #4 on: February 15, 2011, 10:32:35 am »

First thing I would try is direct comparison of measurements. If there is a discrepancy between your old and new devices in spectral mode it would be useful to see it.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Barbieri Spectro LFP problem
« Reply #5 on: February 15, 2011, 11:57:41 am »

First thing I would try is direct comparison of measurements. If there is a discrepancy between your old and new devices in spectral mode it would be useful to see it.

Except its quite likely the spectral data will be different. Not all the manufactures handle this data identically, hence the reason behind XRGA. Which instrument/measurement is “right”?
http://www.xrite.com/company_press_room.aspx?News=641
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Barbieri Spectro LFP problem
« Reply #6 on: February 15, 2011, 12:02:38 pm »

Except its quite likely the spectral data will be different. Not all the manufactures handle this data identically, hence the reason behind XRGA. Which instrument/measurement is “right”?
http://www.xrite.com/company_press_room.aspx?News=641

Direct comparison does not mean digit by digit. And I did not say that spectral data should be compared.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Barbieri Spectro LFP problem
« Reply #7 on: February 15, 2011, 12:07:12 pm »

Direct comparison does not mean digit by digit. And I did not say that spectral data should be compared.

What measured data would you compare and how would differences in how the instruments operate tell us which is incorrect or correct?
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

gromit

  • Full Member
  • ***
  • Offline Offline
  • Posts: 133
Re: Barbieri Spectro LFP problem
« Reply #8 on: February 15, 2011, 04:04:40 pm »

No, we are not using a polarization filter.

As I see it, the ability of using a polarizer to reduce scatter in textured substrates (such as rag paper) is the main attraction of this spectro.
Logged

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Barbieri Spectro LFP problem
« Reply #9 on: February 16, 2011, 05:24:17 am »

What measured data would you compare and how would differences in how the instruments operate tell us which is incorrect or correct?

I never thought discussing ABC :o

For this particular problem the plot of the Y (or L, but from external conversion) curve over the neutral and close to neutral patches (target generated from a known suitable profile) is the most important indicator. The readings converted from spectrum to Y can differ a little because of device agreement issues but the resolution of the levels should be there and noise levels should be acceptable.

The problem starts not with "incorrect" or "correct", but with the resolution and noise. The device that does not resolve well is either at fault or misused.

While comparing the performance of two devices reliable data can be obtained when the same target is read on both of them. I do not see such measurements posted here so not much apart from the above can be suggested.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Barbieri Spectro LFP problem
« Reply #10 on: February 16, 2011, 09:51:37 am »

For this particular problem the plot of the Y (or L, but from external conversion) curve over the neutral and close to neutral patches (target generated from a known suitable profile) is the most important indicator.

Neutral from a profile built using an instrument you are presumably trying to gauge? Or another instrument you presumably gauged the same way? And its neutral because?

Still sounds like measuring and analyzing with a rubber ruler. But if you are sure its useful....
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".
Pages: [1]   Go Up