Welcome to OnyxTalk - Global Support Community, please Login or Create an account to get full access to the forums.
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

ICC - improving colour.

Offline Mikulas Posted 11-30-2015 - 05:40 PM
Post: #26
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

OK, last one for today... (it is 2:30am here)
Just to clear it up, my gamut size is good or bad?
Your gamut size is ... BIG! Hard to say whether bad or good, but on the good side. But the shape of the gamut is problematic. You have too much volume in dark-blues. It´s maybe fine for RGB to CMYK, but not so good for CMYK to CMYK...

Your primaries are hell too dark ! Eek!Confused
Your 82% cyan is similar to 100% ISO Coated v2 cyan.
Your 84% magenta is similar to 100% ISO Coated v2 magenta.
What is see from Ink Restriction table - you have made absolutelly no restriction ??? How is that? Something went wrong ? Check it !!!!!!

You have to go thru the whole process from beginning again - proper ink restrictions:
- eyes are lying, you primaries have a different hue!
- check the numbers - densities are good base (search internet)
- or match Lab value of primaries (read ISO Lab values from Photoshop )

Just FYI - ISO offset colors were optimized to give the best average stable smooth results.
Modern HQ colors often give bigger gamut but produce lot of problems. They usually have very nonlinear primaries (linearized L but bowed hue)

And last - if Barbieri is giving you correct numbers (this is vital)... one day you will have perfect color profiles. Cool
(This post was last modified: 11-30-2015 05:56 PM by Mikulas.)
Back to top Find Quote
Offline Pauly Posted 11-30-2015 - 06:27 PM
Post: #27
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

Hmmm I see that onyx didn't restrict the ink! this is no good.
(the one i manually did isn't the one you're reading, what you've got is what ever onyx outputs)

I'll do another one (again) and get the ink restricted correctly and see what happens. I'll send the data again when i've done it.

Thanks again!!!

My device is new and was checked before leaving and darn expensive compared to the xrite stuff!!
Back to top Find Quote
Offline walter_cfl Posted 11-30-2015 - 06:30 PM
Post: #28
Member
55 Posts
Member

Onyx Version:
X11, X12
Printer(s):
Latex, UV, LED-UV

Yo Pauly,

I missed it, sleeping while you guys partying.. LOL

He's right, seem your primaries a bit too aggressive, well, just use the "HOOK" method to determine your max primaries, easy.

I think you need to verify the status of "Ink Restriction + Linearization + Ink Limit" before proceed to icc profile genaration, do a test print with above 3 parameters only, your test images should have nice smooth tone and detail in the shadow, if you need to cut max density, you could do it on the linearization curves instead of ink restriction.

The "Kick Back" is true.


If Fogra, SWOP, G7 or whatever standard is less of a concern in printing glass, icc is only good for analyzing, not good for printing.

you back to sleep, I'm back to work. Cheers!
Back to top Find Quote
Offline Pauly Posted 11-30-2015 - 07:04 PM
Post: #29
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

I'm still at work, only 2pm!

All the images i use (that require more accurate colour) have imbedded profiles which is photoshop default.
The main problem is my dark reds are brown and reds are orange.
It's not a huge deal when the image has a full range of colours and specific colours aren't as noticeable. But the images i'm printing are one colour, all red,, basically a light painting with reds whites and blacks.
I have a custom profile where after applying the icc, i changed the ink restrictions to add more ink to get the red we want.

I will do another one and see where i get.
Back to top Find Quote
Offline Pauly Posted 11-30-2015 - 09:17 PM
Post: #30
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

New restrictions. used the same readings at first. pulled ink back where it needed to be. (iso 12647 Coated v2)
you can see results in calibration.
Did calibration, used Nfactor - 4
did ICC
Same results.

Note: All readings are done in Barbieri GateWay (externally) if i read them though onyx, it opens a limited version of gateway which is full of bugs

https://www.dropbox.com/s/c1v5jhfp0za77p8/new.rar?dl=0
(This post was last modified: 11-30-2015 09:20 PM by Pauly.)
Back to top Find Quote
Offline walter_cfl Posted 12-01-2015 - 12:26 AM
Post: #31
Member
55 Posts
Member

Onyx Version:
X11, X12
Printer(s):
Latex, UV, LED-UV

created a dummy profile with your data

31 patches.. haha.. no harm but not necessary.. use 20 save you some material.

Try the "Saturation Boost" in the CM preset, this will give you the red you've been looking for, but this preset will ignore any embedded profile.

the gateway version within onyx are older version gateway with bugs in the script, get over it.. LOL

holala, almost time to home
Back to top Find Quote
Offline Pauly Posted 12-01-2015 - 02:31 AM
Post: #32
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

I've used the saturation boost. prints orange. not deep enough reds. (I'll bring my DSLR to work and see if i can take a photo of the difference)

Did you use my icc .spec file? or did you just import my ICC.

I use my device wirelessly, and the embedded software doesn't give me a reference file on the device so i can see if i've missed a patch.
Back to top Find Quote
Offline walter_cfl Posted 12-01-2015 - 06:35 AM
Post: #33
Member
55 Posts
Member

Onyx Version:
X11, X12
Printer(s):
Latex, UV, LED-UV

Spec file used.

I meant the ink restrictions and calibration patches.

After that run a simulation of onyx test file in job editor and yes, I can see now where the orange you talking about, in mid tone where the equal m and y value present, it suppose look like medium red but now it contain very heavy yellow.

Just curious, if profiling it the same way on paper, will it still give you orange?

Nvm, I guess it still print orange. Usually if a red looks orange is because of viewing light in onyx icc parameter set to D65 instead of D50.

Ask you something, how is the environment where you do your chart scanning? Near window? Just curious if... day light affect your reading since scanning on glass, you notice the lamp from your barbieri is yellowish, if letting day light interfering through the glass edge, this will confuse onyx thinking it saw bluish patches, hence forcing more yellow ink, this is unlikely, but just a thought.
(This post was last modified: 12-01-2015 07:31 AM by walter_cfl.)
Back to top Find Quote
Offline Pauly Posted 12-01-2015 - 01:44 PM
Post: #34
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

Light is set to D50. on both my spectro and onyx.

On glass i have the lights off, but paper i dont. But these tests are all on paper. makes it easier. this is how i basically know it's onyx, or something ive missed ect than my media i'm using.

Here's a photo. As you can see, the bottom one is what we want. the top 3 is what onyx likes to put out.
(i made a custom profile to achieve the bottom, but only for red. by adjusting ink restrictions after the icc has been done)
Top may be chroma boost 0.50. the rest chromaboost is off
[Image: IMG_5357.jpg]
(This post was last modified: 12-01-2015 01:45 PM by Pauly.)
Back to top Find Quote
Offline walter_cfl Posted 12-01-2015 - 09:48 PM
Post: #35
Member
55 Posts
Member

Onyx Version:
X11, X12
Printer(s):
Latex, UV, LED-UV

the way i look at your ink restriction.

I noticed there are 233 submarine passed by,

everybody keeping what they know and don't know.

lame.
(This post was last modified: 12-01-2015 09:59 PM by walter_cfl.)


Attached File(s) Thumbnail(s)
           
Back to top Find Quote
Offline Pauly Posted 12-02-2015 - 01:42 PM
Post: #36
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

That's where roughly they are after i restricted them.

I got in contact with canon, They are requesting my profile to investigate.
Back to top Find Quote
Offline Mikulas Posted 12-02-2015 - 05:56 PM
Post: #37
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

Did you use my icc .spec file? or did you just import my ICC.

I´ve found no ICC in you last Dropbox link...
Back to top Find Quote
Offline Pauly Posted 12-02-2015 - 06:28 PM
Post: #38
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

https://www.dropbox.com/s/yc4qrg2qdz33cg...t.Icc?dl=0

Here it is.

Do you guys get any issues like this though onyx? or only me?
Back to top Find Quote
Offline Mikulas Posted 12-03-2015 - 07:44 AM
Post: #39
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

Good question, Pauly...
World of Color management is full of "issues". Wink Without them this forum (and many others... and companies too) would not exit. And we would not be able to write a single line without having experience with such issues...LOL
Onyx is OK. It is the special media you use. With usual paper or film things are 100x more straightforward.
Back to top Find Quote
Offline Pauly Posted 12-03-2015 - 04:55 PM
Post: #40
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

Good point! I looked at some of the onyx youtube videos and they seem to get amazing results.. dammed if i know how.

My paper and Glass look the same. i know i'm in the right track with the glass. all the densities are right on glass, it's only the reds now. My clients are happy with my red solution which is good for me.

All i have to do is wait for an answer from Canon and ODGS support
Back to top Find Quote
Offline virtu Posted 12-04-2015 - 12:38 AM
Post: #41
Frequent Flyer
127 Posts
Frequent Flyer

Onyx Version:
10.2.5
Printer(s):
Spuhl virtu RS-35, Canon ipf9100, HP5000, RS-640

http://grab.by/MuFw
Back to top Find Quote
Offline Mikulas Posted 12-04-2015 - 04:51 PM
Post: #42
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

Pauly, give me some days please. I will catch and report some of the biggest problems. But I need time.
I would be very surprised if you get some miracle answer from that support. Wink The reason is - as I told you already 5 times - your ICC profile is mapping that red area in a very strange way. Instead of leaving the colors where they are, profile is mapping these colors to inside... That´s strange because your gamut is bigger than most CMYK standard profiles =no moving necessary.
Now ask me why. Yell




virtu, do you know where is the left photo coming from ?

Search the internet - "average face study" - it´s really interesting !
Most beautiful are average faces....
Back to top Find Quote
Offline Pauly Posted 12-04-2015 - 11:47 PM
Post: #43
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

Mikulas, When ever you can Smiley
I do understand what you're saying now.

The guy i've been dealing with, he's a certified onyx tech, and up there in Canon & ODGS. You could be possibly right, but we'll see. His finding it strange also.
Back to top Find Quote
Offline Pauly Posted 12-06-2015 - 03:20 PM
Post: #44
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

I was reading up on one of the OCDG user guides.

For most medias, (use generic profile yada yada yada)
Ok do a test print using generic.
looked like junk
Recalibrate it (now i can as before i couldn't with out a specro)
re print.
Not bad, much better than the original. reds are closer that's for sure.

Looked at the calibration table, noticed a different Nfactor for all colours. Interesting.

So my thoughts now are.
Should i copy the nfactor and see where i get with a new profile. - But after ICC no matter what i do, print is always the same.
Recalibrate printer with those Nfactor settings. But then the ICC would invalid as all the ink calibration settings are different.
Back to top Find Quote
Offline Mikulas Posted 12-06-2015 - 06:33 PM
Post: #45
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

Using generic Onyx profiles ...yes! This is also what I wrote you some pages back... Grin Grin
Most probably you will never reach the quality of pro-level profiles. So it´s a good idea to use them...

If using generic Onyx profiles, you should ONLY duplicate them, and do re-calibration! This puts all 1-95% values back in place (compensating for media nature). Of course with assumption that your media must be friends with the original Ink Restrictions. If not, you will get ugly artifacts. But then necessary pulling back restrictions is moving the 100% density = changing the gamut boundary = requires re-profiling...

N-factor... ignore this mysterious Factor simply. Smiley It is only there to set the shape of tone value curves after calibration (former Target Dot Gain) . N=2 was found as a good average. Changing it will do a small job (and one must exactly know what it does). It is good for "super-precise" profiles.

1/ What generic profiles have you found for glass ?
2/ Look at the graph below. It´s your measurement in Lab. What you see are the M=0-100% with Y=48,64,82,100% combinations. It is far far from being nice... If would really think of averaging at least 3 measurements...
(This post was last modified: 12-06-2015 06:44 PM by Mikulas.)


Attached File(s) Thumbnail(s)
   
Back to top Find Quote
Offline Pauly Posted 12-06-2015 - 07:24 PM
Post: #46
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

I haven't tried it on glass just as yet. I'm not sure how i'm going to get it to work on glass as i paint the white on, i dont use the white ink. So reflections may affect it. we'll see.
The profile was built using onyx. Because it shows me the spec readings if i ask for them. i exported them also. I can also see what settings they used in the builder. nothing different that what i do. So this also confuses me a little.


I get how Nfactor works. But yes not really necessary.

I'm assuming the chart should be more linear?


I'm going to download all the different generic profiles that i can make use of, and re calibrate them. and see where i get
Back to top Find Quote
Offline Pauly Posted 12-06-2015 - 11:54 PM
Post: #47
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

So i downloaded a few Generic profiles. not many to choose from, paper, film, ect. They are all made in onyx. I know this as when i open them up to have a look, all the readings are there, and everything is adjustable. I can even see the adjustments for the ICC profile.
Most settings are similar to mine.

One thing i noticed is that the ink restrictions are mostly to the max. Then the calibration is used to pull back the inks. makes sense to me.

ink limits are either on 400 or 350.

I opened the ICC readings and plotted the chart. print screened it then opened my chart. They both had a very similar shape. with similar ranges ect.

So i decided to toy around a little. I got one of my profiles i made for paper, and used their paper ICC readings and built (with the same setting they used which is the same as mine) a new profile. test print. exactly the same as the previous profile i had built.

Right now the generic profile is fine to use. (Thanks Mikulas Razz ) But to get it to work on glass, i need it on density mode. I cannot use the generic density mode because it lays down to much ink. I havent had a change to see what i can do with it just yet though.

I'll see what i come up with tomorrow. I am getting closer! I know i'm doing something wrong somewhere. Not sure where though!!

I might even get a generic profile and build the icc with my own readings and see how that goes.
Back to top Find Quote
Offline Mikulas Posted 12-07-2015 - 06:31 AM
Post: #48
Banned
55 Posts
Banned

Onyx Version:
10.2
Printer(s):
L26500, W8400+Onyx / Mimaki JV4, JV5 +TexPrint

Another info I have for you - concerning GCR settings. I had a look at your 3-channel combinations.
a/ CM+Y, CY+M and MY+C and compared them to
b/ CM+K, CY+K and MY+K

Normally chromatic darkening (a/) should be a bit less saturated than darkening with K (b/).
This is normal behavior because the last color channel is not only darkening, but also desaturating - pulling color back to gray axis quicker than K.

In your case I can confirm what the manufacturer says about the inks - chroma curves are really overlapping up to approx. 50% of the tone value. I.e. you get the same chroma with 0/100/100/50 as with 50/100/100/0. Ditto with all other channel combinations.

I.e.- setting the GCR starting point for K somewhere around 50% is fine for your inks+media !

-C-O-N-F-I-R-M-E-D-
Grin Grin Grin

One thing i noticed is that the ink restrictions are mostly to the max. Then the calibration is used to pull back the inks.
Ink limits are either on 400 or 350.


A nice small trick from Onyx.
Normally restrictions help to maximize the useful ink coverage per channel. But if you are following some density standards, then you can simply set the 100% CMYK points in the linearisation graph manually to where you want them... and the linearisation curve will restrict the max. ink...
Most modern inks+media can handle more ink density than any standard, so yes it´s possible in this way too...
But not sure whether it´s also suitable for glass... Nuetral


And yet another one for today:
Ink restrictions in your last posted profile wasn't done properly. You were hunting density instead of chromacity. Have a look at magenta chromacity - from 85% up is not evolving anymore. Chromacity gain is marginal compared to ink increase.
0/70/0/0 ..... 62,09
0/75/0/0 ..... 70,29
0/80/0/0 ..... 70,12
0/82/0/0 ..... 69,41
0/85/0/0 ..... 73,28
0/90/0/0 ..... 73,37 !
0/95/0/0 ..... 75,05 !
0/100/0/0 ..... 76,09 !
0/100/0/0 ..... 75,96 !

Have a look at cyan chromacity - fine up to 90%. Then jumping in a strange way with no significant chroma gain (some ink values have 2 patch measurements):

85/0/0/0 ..... 55,38
90/0/0/0 ..... 61,53
95/0/0/0 ..... 63,57 !
100/0/0/0 ..... 61,93 !
100/0/0/0 ..... 64,03 !

Have a look at yellow chromacity - there is one erratic reading at 85%. Otherwise yellow chromacity gain is fine up-to 100% Wink

82/0/0/0 ..... 83,03
85/0/0/0 ..... 82,89 !
90/0/0/0 ..... 89,96
95/0/0/0 ..... 92,37
100/0/0/0 ..... 98,59
100/0/0/0 ..... 98,25

And lastly, have a look at black lightness (L-coordinate in Lab).
There no reasonable density increase after 90%, and values are erratic like hell:
0/0/0/75 ..... 24,8
0/0/0/75 ..... 22,8
0/0/0/85 ..... 23,2
0/0/0/90 ..... 13,3
0/0/0/95 ..... 13,1 !
0/0/0/100 ..... 11,0 !
0/0/0/100 ..... 10,3 !

Have a nice day!
(This post was last modified: 12-07-2015 12:53 PM by Mikulas.)
Back to top Find Quote
Offline Pauly Posted 12-07-2015 - 02:23 PM
Post: #49
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

Thanks for those efforts!!

For the promoter and white we use for glass, I can use ink calibration on it to restrict. Because the paint acts different with the ink it helps.

I have a few things i'm going to try today and i'll report back.
Back to top Find Quote
Offline Pauly Posted 12-07-2015 - 09:00 PM
Post: #50
Frequent Flyer
109 Posts
Frequent Flyer

Onyx Version:
12.0
Printer(s):
Oce Arizona 480XT

I did some toying around today.
new media profile.
i made notes of the generic profile ans saved the spectral data.

Printed and read the ink restrictions. adjusted them to the same as the generic one. (curves are quite close also)
copied and made multiples with the same restrictions
did a calibration then did the ICC with same settings (i have made a setting in the selection box)

Made another one, skipped calibration did the icc.
copied that one, then did a calibration on it,

Printed all 1 profiles next to eachother with my test file and all the same. same as what i usually get. also printed the generic to compare and it's different.

made a new media. used the same restrictions. calibrated it. then imported the icc.
i copied that one but instead of importing the icc, i imported the .spec file.
printed both.
the one with the imported icc looks like the rest. the one with the imported .spec file looks slightly better than what i had, but not as good the generic.
Back to top Find Quote


Possibly Related Threads...
Thread: Author Replies: Views: Last Post
  Quick sets colour management byebyegt1 1 2,205 09-06-2013 10:26 AM
Last Post: Correct Color
  XL1500 4 Colour - Tarp Wenzil Chettiar 1 1,493 02-22-2012 08:33 AM
Last Post: Correct Color
  Creating a spot colour for Yellow fluorescent Dye-Sub Ink Scody 1 2,616 09-20-2011 08:08 AM
Last Post: Correct Color
  True Colour Management vs "Standards" Fingers 21 14,028 05-10-2011 03:46 AM
Last Post: MarcoRoos
  Colour Management - valued differently worldwide? Fingers 7 4,096 07-30-2010 06:47 AM
Last Post: Correct Color
  HP 5500 Wider Colour Gamut ttlcntrlprint 3 3,501 05-29-2010 03:44 AM
Last Post: Correct Color
  Spot Colour oss_flatbed 5 3,323 09-02-2008 02:44 PM
Last Post: Douglas
  Input Colour Overview Hotspur 1 2,298 05-25-2008 10:53 AM
Last Post: Scott


User(s) browsing this thread
1 Guest(s)
myRepono - Website & mySQL Database Backup Service
Community software by MyBB