Pages: [1]   Go Down

Author Topic: Drying Time  (Read 3840 times)

Victor Engel

  • Newbie
  • *
  • Offline Offline
  • Posts: 32
Drying Time
« on: April 19, 2011, 11:00:12 am »

I recently got a Colormunki, which includes a feature in the software to wait for drying time. The drying time in the software, in my opinion, is inadequate for such paper as Colorlife. I can wait a day or two after a print to do the scan. That's not a problem. In fact, the software allows you to state that you already have a printout.

My question is about the second print. It doesn't seem to have a similar flow in the software that would make it convenient to let the second print dry a day or two before scanning it. Or did I just miss that?

On a related question, has anyone done scans to test how much the color changes over time on various papers (I'm specifically interested in Colorlife, but a more generic answer is appropriate)? Is the 10 minutes built into the software sufficient?
Logged

Wil

  • Newbie
  • *
  • Offline Offline
  • Posts: 13
Re: Drying Time
« Reply #1 on: April 19, 2011, 12:09:04 pm »

I have been doing some tests on this exact question.  Till now, all the profiles I have made have been with xRite i1 Pro, Match 3.  My printer is a Canon Pro9000 II.

Because the price of OEM ink is higher than human blood, I obtained some OPC ink to try.  I made profiles for three different papers, CostCo Kirkland Glossy, Red River Arctic Polar Satin, and Red River Polar Matte.  In the past, I have always let the test target dry for a day before making the profile and used this method with this new ink.  After the profile is made, I make a test print from a file that has the full gamut of Adobe RBG.

The test prints from all three papers were a little flat and not what I wanted.  During the next few days, I observed the test prints, and to my surprise, they improved each day for more than a week.  By week's end, they were as good as I have ever printed.  I don't know if this is just a property of this ink, or a general situation.  I have not observed it with OEM ink, but I was not looking for it with that ink.

That made me think that waiting three days to make the profile was the way to go, but for more advice, I called the xRite people who said that the profile should only be made after the target was completely dry -- 20 min, 1-hour, 3-days, two weeks, etc.

With that advice, I did make profiles after three days and was able to observe the initial test prints beside the new ones.  But I did something in addition.  I also made profiles and test prints from these targets after drying for only 20-minutes.  Now I had three sets of test images with the only variable being the drying time to make the profile.

I was shocked to learn that the profiles made after drying 20-minutes ultimately made the best prints.  I have two 12 x 18 prints with these 20-min profiles on my wall above my desk.  They seem to get better as the days go by.  After a week, they are an almost perfect match with my monitor, except of course, for the difference in reflected and transmitted light.

All the profiles were good, and I could only tell a difference when the three prints were viewed side-by-side.  My advice from this experence is to make the profile when you desire, but a little better results will be obtained if the profile is made as soon as the target won't smear when being scanned.

Good Luck!
Wil
Logged

Victor Engel

  • Newbie
  • *
  • Offline Offline
  • Posts: 32
Re: Drying Time
« Reply #2 on: April 19, 2011, 12:29:06 pm »

That's surprising and good to hear. It's definitely the most convenient. Any speculation on why this may be the case? It seems very counterintuitive.
Logged

Victor Engel

  • Newbie
  • *
  • Offline Offline
  • Posts: 32
Re: Drying Time
« Reply #3 on: April 19, 2011, 12:56:45 pm »

Looks like my issue about the second print is mentioned in this detailed review of Colormunki.
http://www.northlight-images.co.uk/reviews/profiling/colormunki_printing.html
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: Drying Time
« Reply #4 on: April 19, 2011, 01:27:56 pm »

There was some discussion here several months ago about the failure of the software to permit longer drying times for the second set of patches.  It was noted you can overcome this by going through the setup and canceling print before the second set prints and then reading the pre-printed 2nd set; so you can do thing with 24 hour intervals if need be.
Logged

Victor Engel

  • Newbie
  • *
  • Offline Offline
  • Posts: 32
Re: Drying Time
« Reply #5 on: April 19, 2011, 01:43:21 pm »

There was some discussion here several months ago about the failure of the software to permit longer drying times for the second set of patches.  It was noted you can overcome this by going through the setup and canceling print before the second set prints and then reading the pre-printed 2nd set; so you can do thing with 24 hour intervals if need be.

I thought about this, but it assumes the software will calculate the same values the second time around. I'm not prepared to make that assumption.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Drying Time
« Reply #6 on: April 19, 2011, 01:56:50 pm »

I thought about this, but it assumes the software will calculate the same values the second time around. I'm not prepared to make that assumption.

Easy to test, even without a Spectrophotometer or a tool like ColorThink.

Run this process twice and save off the two TIFFs (iteration of 2nd target). Open both in Photoshop and use Apply Image>Subtract. The details are here: http://digitaldog.net/files/Apply_Image.pdf
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bill t.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3011
    • http://www.unit16.net
Re: Drying Time
« Reply #7 on: April 19, 2011, 02:19:37 pm »

Run this process twice and save off the two TIFFs (iteration of 2nd target). Open both in Photoshop and use Apply Image>Subtract. The details are here: http://digitaldog.net/files/Apply_Image.pdf

Now that's brilliant!  There's a poetry to solutions of that sort that I find pleasing.

Of course all this begs questions about the consequences of changing ink cartridges, new batches of the same media, humidity, temperature, printer idle time, phase of the moon and on.  The mind boggles.
Logged

Alan Goldhammer

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4344
    • A Goldhammer Photography
Re: Drying Time
« Reply #8 on: April 19, 2011, 02:53:01 pm »

Now that's brilliant!  There's a poetry to solutions of that sort that I find pleasing.

Of course all this begs questions about the consequences of changing ink cartridges, new batches of the same media, humidity, temperature, printer idle time, phase of the moon and on.  The mind boggles.
Maybe we don't need profiles after all.  Just let the printer manage colors as it knows best! ;D
Logged

Wil

  • Newbie
  • *
  • Offline Offline
  • Posts: 13
Re: Drying Time
« Reply #9 on: April 19, 2011, 06:06:57 pm »

That's surprising and good to hear. It's definitely the most convenient. Any speculation on why this may be the case? It seems very counterintuitive.

Back to the drying time (I like to call it cure time) thing and why the 20-min cure time ultimately produced the best profile with this particular ink -- other inks may produce different results.

The results didn't make sense to me either, and my reasoning didn't satisfy me.  In an effort to get a better handle on all of this, I purchased ColorThink 2.3 to look at the profiles.  I am just getting my hands around ColorThink, so there is a lot to learn about it, but from what I know, I can display the gamut of profiles.  For me, that is enough to look at this situation.

1.  Before getting to these printer profiles themselves, I looked at the Dell 2408 monitor profile and the Red River OEM ink profiles for their Arctic Polar Satin and Polar Matte papers for the Canon Pro9000.  I was surprised to learn that all of these profiles were larger than the Adobe RGB color space in some areas and almost as large in all the areas.  I knew all these items were good, but didn't suspect them to be that good.

2.  Next I compared each of the new printer profiles I made with Adobe RGB and the Red River OEM ink profiles.  Then the light started to shine.

The gamut of the profile made from the three day cure was almost identical to the Red River OEM ink profile -- a good sign for this ink.

The gamut of the 20-minute and one day cure profile was not quite as large as the three day cure.

3.  The visual evidence is that after curing for a week, the test print from the 20-minute cure was the best.  Visually, it looks like its gamut is at least as large as OEM ink.

4.  I don't know how to measure the gamut of a test print, but since the print from the 20-min profile looks the best after a week cure, I conclude that the gamut of prints from this ink increases over some time frame.

5. This is the hard part.  If 4 (above) is correct, here is what I think it all means,

5.1  Both the profile target and final prints will cure at the same rate.

5.2  If a profile is made from a target that has cured for a few days, it will try to make a print whose initial state will be determined by the final cure state of the test target.  Then the print will do its own curing, and its final state will be different from the final state of the profile.

5.3  If a profile is made from a test target as soon as possible (it doesn't smear), the print it tries to make will then initially have the gamut of the initial target.  It will then cure in the normal way and produce the expected results of increased gamut.

6.0  While all of this was done with a specific third party ink, I think the results indicate that getting the profile to be in accord with how the print comes off the printer is the best practice.

Good Luck!
Wil
Logged

Victor Engel

  • Newbie
  • *
  • Offline Offline
  • Posts: 32
Re: Drying Time
« Reply #10 on: April 20, 2011, 06:25:36 pm »

I'll have to think about this and post another reply later. I do have a couple of side remarks, though.

How would you compare ColorThink to Gamutvision? I've been thinking of getting Gamutvision.

About increasing gamut over time, that is easily explained (I think) by considering ink spread. Right after a print, the ink will be a sort of pointillism picture. After some cure time, the ink will have time to spread out, essentially filling in some white voids. The white will tend to "wash out" the color with respect to the blended final result. Or looked at from the other direction, with the ink spread, the color should tend to saturate, given enough ink volume, anyway.
Logged

Wil

  • Newbie
  • *
  • Offline Offline
  • Posts: 13
Re: Drying Time
« Reply #11 on: April 21, 2011, 09:35:56 am »

I'll have to think about this and post another reply later. I do have a couple of side remarks, though.

How would you compare ColorThink to Gamutvision? I've been thinking of getting Gamutvision.

I know nothing about Gamutvision.

Good Luck!
Wil
Logged

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Drying Time
« Reply #12 on: May 10, 2011, 03:26:10 am »

has anyone done scans to test how much the color changes over time on various papers (I'm specifically interested in Colorlife, but a more generic answer is appropriate)?
Not having read many credible facts about drying time, I decided to my own tests yesterday.

Methodology;
Print out a chart then measure it as often as seemed reasonable.
I used a i1 Easy RGB 1.2 chart (3 rows of 15 patches) to allow fast measuring times. Measuring with an i1 Spectro, then comparing readings in PMP 5’s measure tool.

I know from previous tests that I get around >0.15 dE average variation between readings of fully dry charts with a maximum dE of >1
So once the difference between reading gets to that level I considered the chart to be stable.

For my test I used an Epson 3800 and Jessops gloss paper (a decent non-OEM cheap gloss paper).
I found that readings became stable after about 25 minutes drying time. Darker colours seeming to take longer to stabilise than lighter colours. After 25 minutes there was a more random distribution of errors amongst colours suggesting that measurement variability was a more likely cause of variation than drying issues.
This fits well with Epson’s own ColorBase utility for calibrating their pro inkjets requiring a minimum 30 minute drying time before measuring it’s charts.

So to answer your question;
Quote
Is the 10 minutes built into the software sufficient?
The answer is no, that’s not really enough time.

I’d assume that different inks, papers and environments will behave differently too, so there’s room for further investigation.

I’m not convinced visually evaluating prints made from profiles made at varying measurement times is necessarily a valid approach, unless you know the veracity of the underlying data. Having now spent quite some time checking the range of variations in spectro readings, I’ve seen occasional variations in measurement that could lead to different profiles being built, possibly due to inaccurate calibration or maybe other changes like temperature.
For really good profiles it would seem prudent to make multiple measurements and check that they don’t vary more than expected.

Finally, Gamutvision is a great tool and has a mass of options and facilities, but like most colour tools you need to understand what it’s telling you to get the best from it.

Paul
Logged
Pages: [1]   Go Up