ArgyllCMS profiles: Black Point Compensation gives composite black when checked

[Ink stained Fingers]

The B/W plot of Gamutvision in rel. col. mode should show if the colorprof -nI option has an impact on the blacklevel value although it wouldn't clear any conflict with Lightroom - I don't use it either. So just go ahead and post a profile with this option and we can check if we see any impact

 

[Graeme Gill]

The B/W plot of Gamutvision in rel. col. mode should show if the colorprof -nI option has an impact on the blacklevel value although it wouldn't clear any conflict with Lightroom - I don't use it either. So just go ahead and post a profile with this option and we can check if we see any impact

I don't think Gamutvision will show anything useful, since it has previously showed that existing ArgyllCMS profiles have a perfectly good perceptual intent black level. The problem reported seems to be mainly in Lightroom. So best if someone who has this problem re-create a profile with ArgyllCMS V3.2.0 and report back.

 

[pharmacist]

Can someone have a look with these two profiles: old version without -nI option, and new version with -nI

 

[Ink stained Fingers]

I did a check of the profiles , I'm wondering about the result that both profiles are completely identical, they don't differ by a line width of the plots in the diagram.

I'm displaying here the media black points of both profiles - they are identical to the last decimal digit of the posted values - there are no reading differences at all from 2 different patch sheets which is quite unusual - your spectro is most likely not that accurate.

Did you do separate scans for the -nl option or just 2 calculation runs with -nl on or off with the otherwise same data ?

 

[Graeme Gill]

I did a check of the profiles , I'm wondering about the result that both profiles are completely identical, they don't differ by a line width of the plots in the diagram.

Note that any difference in the profiles using -nI will show up only in the A2B0 and A2B2 tables. The theory is that Lightroom is using round trips through the A2B0 and B2A0 tables in its BPC heuristic.

 

[Ink stained Fingers]

Here is a display of the B/W response curve of Gamutvision, only the black point at L*=0 is of interest if there are any shifts of the output value as reported earlier, this is an overlay of 2 Gamutvision screen shots - with BPC of and on and rel col rendering, the diagrams look exactly the same for the -old profile version , there is not a bit difference in the diagrams, both curves start at the same black point

I think the plots look very much as you would expect them to perform for the BPC off-on options. That's about all of the info I can retreive from these profiles , I cannot see the problem with the black point shift as the title of this thread implies. I don't know if there are other program settings under which this effect becomes apparent specifically with Lightroom.

 

[pharmacist]

I did just calculate twice with the same scanning results on without the -nI option (old) and the new including the -nI option.

 

[Ink stained Fingers]

I think you should go back to the situation as discussed in posting # 20 that the blacklevel raises for the ArgyllCMS profile when BPC gets switched on - the starting point of the discussion and the title of the thread. I remember from a previous discussion about 1 - 2 years ago that the effect was more prominent with a profile for a paper - like a matte paper with a media black point around 20.