Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: BradSmith on May 24, 2016, 06:45:40 pm
-
I'm ordering a custom profile from a service. They've sent me the 2 patch files as tif's. Below is the first of the patches - "Patches pg1 File". I'm on a Mac, so I printed the file using the Adobe Color Print Utility. It printed very dark - see the second file "Print from Ad.Col.Print.Util". So I printed it again through Adobe Color Print Utility on a very different paper, and it printed similarly dark. Then as a validity check, I imported the tif file into Lightroom, with no adjustments, and printed it on a glossy paper with appropriate profile and the print is a very close match to the screen, as I'd expect. I'm an experienced printer, using a calibrated NEC @ 100 cd/m2 with an Epson 3800. In ACPU, I used the correct media type for each of the papers, with color management off in the Epson driver.
Should the prints vary so widely from the appearance of the tif file? I don't want to send these in and find out something is very wrong on my end.
thanks
Brad
-
It's not necessarily wrong that there be a difference of appearance between the print of the target and its appearance on your monitor. What I find troubling is the uneveness of the luminance - the left side being considerably darker than the rest of it. You do not say what your settings were in ACPU and what papers you used. It may help verify whether there are problems to know this
-
There's really nothing to gain visually by viewing those prints. DO triple check all the printer settings, that inks are firing cleanly etc.
-
Mark and Andrew,
Thanks for the quick responses. The paper I want to order the profile for is Epson Legacy Platine. The other paper I printed a target for was Eps. Prem Glossy PP. Here are the print settings for each. (I used media type Luster for the print on Platine because that is what Epson says to use...I think)
Brad
-
Those settings look correct Brad, so if there is a problem, that's not the source of it. Quite possibly, those targets reflect the native behaviour of your 3800, which of course the profile you order should help correct for. I'm still wondering about the uneveness of the luminance between the left side and the remainder that you show. Is this really in the target print, or is it a result of how you reproduced the target for posting? Presumably you did a nozzle check before printing. You may wish to reprint the targets using the same settings on the same two papers to see whether you still get the same kind of results. Machines are machines and as good as they are, sometimes "s..t" happens and we may never know why. If they still come out dark, I think it sensible to send them to your profiling contractor and tell him/her what your observations are.
-
Should the prints vary so widely from the appearance of the tif file?
In the case of Epson printers, yes. I've measured thousands of pages of profiling targets and unprofiled prints from Epson are universally darker than then they appear on screen.
If you think about it, why would you build a printer profile if the screen and unprofiled print matched ?
My concern is whether the profiling business actually has a licence to use i1Profiler for commercial remote profiling as, without separate agreement from X-Rite, it's against the EULA of the software. Unlikely I'd suggest, as they haven't been smart enough to use an optimum patch sample size.
-
If you think about it, why would you build a printer profile if the screen and unprofiled print matched ?
My concern is whether the profiling business actually has a licence to use i1Profiler for commercial remote profiling as, without separate agreement from X-Rite, it's against the EULA of the software. Unlikely I'd suggest, as they haven't been smart enough to use an optimum patch sample size.
Thanks for your feedback based on you having read many profile targets. I was not expecting the output to match the screen, but I guess I was expecting more random variance in color tone/saturation and not such an overly dark output. Regarding "optimum patch sample size", they sent two files to print with a total of approximately 1000 patches. What would you (or the profiling hardware manufacturer) consider a more optimum number of patches to be?
thanks
Brad
-
My concern is whether the profiling business actually has a licence to use i1Profiler for commercial remote profiling as, without separate agreement from X-Rite, it's against the EULA of the software. Unlikely I'd suggest, as they haven't been smart enough to use an optimum patch sample size.
Why are you concerned about such a legal matter when the OP raised a technical question?
-
In the case of Epson printers, yes. I've measured thousands of pages of profiling targets and unprofiled prints from Epson are universally darker than then they appear on screen.
Considering the data is printed without color management, that's not surprising.
-
Considering the data is printed without color management, that's not surprising.
Maybe, but the behaviour of Canon and HP printers is far less radical and distinctive.
-
Why are you concerned about such a legal matter when the OP raised a technical question?
I answered the technical aspect first, but many photographers are rightly highly protective of IPR and should respect the IPR of other creators of protected work.
-
I answered the technical aspect first, but many photographers are rightly highly protective of IPR and should respect the IPR of other creators of protected work.
Firstly, you don't know what kind of license the profiling entity has - so why speculate; I expect a firm like X-Rite can look after its own commercial interests - that is probably what they do best; and secondly, getting back to the more substantive technical side, I am interested to hear your view on what "an optimum number" of patches is, because I don't think this matter is all that unambiguous.
-
Firstly, you don't know what kind of license the profiling entity has - so why speculate;
Having inquired about getting an extension to my i1P licence without any success, I'm a little sceptical that this organisation would supply an unbranded chart if they had a full licence agreement for it.
Maybe I'm wrong and in other territories X-Rite are more cooperative.
I am interested to hear your view on what "an optimum number" of patches is, because I don't think this matter is all that unambiguous.
This has been discussed on here at length before in a few different threads since i1P was first released.
Back read the threads and you'll find references to how there are certain 'sweet spots' in the number of patches that can give better results.
-
Having inquired about getting an extension to my i1P licence without any success, I'm a little sceptical that this organisation would supply an unbranded chart if they had a full licence agreement for it.
Maybe I'm wrong and in other territories X-Rite are more cooperative.
Depending on if you want to go with the letter of the law or not, it's kind of moot. It's super easy for someone who knows what they are doing to remove all the tags that identity the profile was built by X-rite. X-rite keeps (or did keep) changing the EULA. For profile services, one can obtain permission from X-rite to do so. What they are really trying to protect themselves from and should is a big company, like a paper company, supplying profiles built with their products; such companies are supposed to pay a hefty licensing fee for this and should. For someone building a custom profile for one user, I think you're fine. They apparently even allow people who rent their equipment to let the renters of the hardware build their own paper profiles.
-
Depending on if you want to go with the letter of the law or not, it's kind of moot.
Not really.
X-Rite's previous licences for Profile Maker Pro specifically allowed the licensee to offer a general remote profiling service. When i1Profiler was released the EULA changed and no longer allowed it. Remote profiling is only allowable under very specific conditions.
Again this was discussed here in some detail when i1P was first released.
It's super easy for someone who knows what they are doing to remove all the tags that identity the profile was built by X-rite.
Which constitute reverse engineering/modifying the product and is against the EULA too.
What they are really trying to protect themselves from and should is a big company, like a paper company, supplying profiles built with their products;
Probably true, but it also seems to attempt to protect specialist consultants who work making profiles on-site too.
-
Not really.
When i1Profiler was released the EULA changed and no longer allowed it.
And that was again changed due to the uproar. That was then, this is now.
Which constitute reverse engineering/modifying the product and is against the EULA too.
Yes it does depending on your expertise as a lawyer. That part of the EULA can't be enforced if X-rite hasn't a clue what built the profile. I'm not advocating this, I'm pointing out it's not enforceable.
This is like the current silly debate about bathrooms in this country. The 'right' tells us to protect privacy, boys born boys should go to one bathroom while girls born girls go to the other. But to enforce this, everyone would have to provide a birth certificates then pull down their pants, showing their genitals to the bathroom police. So much for privacy. IOW, this is silly can can't be enforced.
Probably true, but it also seems to attempt to protect specialist consultants who work making profiles on-site too.
NOT if said specialist’s contact X-rite for permission which after the flair up over the original EULA, X-rite said they would provide. Again, that was then, this is now.
-
Notification of Revised End User License Agreement (EULA)
for X-Rite i1 Solutions
May 23, 2011
Dear Consultant,
We would like to take this opportunity to thank you for your continued support of X-Rite color management solutions. We certainly value the services and education you provide to the imaging community regarding color management workflows. We also recognize that X-Rite's i1 solutions and legacy profiling applications have become part of your go-to tool set to meet your customers' needs. To that end, we are excited about our next generation i1 Professional Color Management Solutions which includes the new i1Profiler software. We certainly hope that you will embrace and adapt our new i1 solutions to deliver unrivaled color quality to your customers.
As you know, the industry has changed quite a bit since our legacy profiling solutions first came to market - even since our last updates. Some of these changes include a broader distribution of generic profiles, as well as broad based online profiling service models (that do not include significant billable services involving personalized customer interaction). As part of the i1 Pro solutions, X-Rite felt it was an appropriate time to update our EULA (end user license agreement) to better reflect current practices of profile distribution business models.
X-Rite did release a new EULA on April 6, 2011 containing some language that several of you voiced concerns over and have since provided us with valuable input. Based on these interactions we are pleased to provide you with a revised EULA, which is retroactively effective as of April 6, 2011.
It is not our intention to question or monitor your business practice. However, there are two types of profile distribution practices that will require a separate distribution agreement with X-Rite. The first point has been in place for many years, while the second point is a newaddition. Taken from the EULA they are as follows:(i) in conjunction with the sale or promotion of an input, display or output device and/or ink or paper if such sale or promotion extends beyond a single customer-specific application,
(ii) in conjunction with profiling services that are offered electronically or online, including through electronic media, email or other network-based communication channels, without significant billable services that involve personalized customer interaction. We recognize that each business is unique and we are happy to grant individual agreements that can be adapted to your business models and requirements. If you think you may need a special agreement, or have any specific questions about this, please visit our website and complete the inquiry form. We will have an X-Rite sales or marketing representative contact you. Visit xrite.com/specialEULA (http://click.email.xrite.com/?ju=fe2c17747166017f771079&ls=fdf01777726403797c1d7175&m=ff041070756304&l=ff01177075660d&s=fdf515717d6d04787111717d&jb=ffcf14&t=).
For your reference, a copy of the entire EULA can be downloaded by clicking here (http://click.email.xrite.com/?ju=fe3417747166017f771170&ls=fdf01777726403797c1d7175&m=ff041070756304&l=ff01177075660d&s=fdf515717d6d04787111717d&jb=ffcf14&t=). You can also find it online under the Support section of related products.
Thank you for your continued support of X-Rite solutions. We look forward to continuing a mutually beneficial relationship with you.
With kind regards
The i1 Solutions Team
X-Rite, Incorporated
-
Not withstanding your quotes above;
The current EULA that users must agree to on installation http://www.xrite.com/i1profiler-i1publish/support/kb5482 doesn't allow you to offer a remote profiling service without an additional licence.
Maybe that easy or cheap to get in the USA, but in the UK they didn't want to discuss it when I asked.
-
Maybe that easy or cheap to get in the USA, but in the UK they didn't want to discuss it when I asked.
Did try through http://www.xrite.com/custom_page.aspx?PageID=244? (http://www.xrite.com/custom_page.aspx?PageID=244?)
-
I'm a little sceptical that this organisation would supply an unbranded chart if they had a full licence agreement for it.
Take a look at the images of the charts in my original post. Lower right hand corner identified it as an XRite product. I'd call that "branded"
-
Take a look at the images of the charts in my original post. Lower right hand corner identified it as an XRite product. I'd call that "branded"
Yes so would I, and the whole discussion about licensing is completely irrelevant to the OP's concern. Let's take this back to the technical issue raised. It is correct that there is no reason why the colour perception of a printer profiling target on a monitor needs to be close to the colour perception of the print of the target. Let us always bear in mind that the target, if printed properly, reflects the native behaviour of the printer, which the resulting profile will describe numerically and help in the colour management process to reproduce the image file numbers correctly. So in principle there is no reason to worry about the fact that it looks dark. But I do see some cause for concern about the unevenness of the luminance between the left side and the rest. Is this because of how the target print was reproduced into a JPEG for posting here, or is that what it really looks like? This particular matter IS unusual and unexpected.
-
Yes so would I, and the whole discussion about licensing is completely irrelevant to the OP's concern. Let's take this back to the technical issue raised. It is correct that there is no reason why the colour perception of a printer profiling target on a monitor needs to be close to the colour perception of the print of the target.
That IS correct. Now both pages should appear similar in density.
-
Did try through http://www.xrite.com/custom_page.aspx?PageID=244? (http://www.xrite.com/custom_page.aspx?PageID=244?)
No, that wasn't available when I tried and I never received that communication from X-Rite about the option either.
-
Take a look at the images of the charts in my original post. Lower right hand corner identified it as an XRite product. I'd call that "branded"
The point is whose brand ? You weren't sent those charts from X-Rite were you ? All the charts built and supplied by my business are branded with colourprofiles.com
-
But I do see some cause for concern about the unevenness of the luminance between the left side and the rest.
I don't see that. The black reference lines surrounding the chart show no variations along their length.
No obvious cause for concern there.
-
I don't see that. The black reference lines surrounding the chart show no variations along their length.
No obvious cause for concern there.
Well I do, and while it may not be so visible from the black reference lines, it is pretty obvious to me looking at the content of the patch set. But maybe we have different ways of seeing. Anyhow, not something you or I can take any further. Best for the OP to do as I suggested earlier up, which is to send his targets with some commentary to the profiling service, let them advise whether they can work with these, if so get the results and see how well the new profile works.
-
Well I do, and while it may not be so visible from the black reference lines, it is pretty obvious to me looking at the content of the patch set. But maybe we have different ways of seeing. Anyhow, not something you or I can take any further. Best for the OP to do as I suggested earlier up, which is to send his targets with some commentary to the profiling service, let them advise whether they can work with these, if so get the results and see how well the new profile works.
I agree. Once the OP has the profile it can be tested.
-
I will only weigh in regarding the OP's question. I've done a lot of profiling over the years, both for myself (Epson 3880) and others who have had a variety of printers. I've never seen a profile patch set as dark as that shown by the OP. I use a patch set generated by ArgyllCMS and it gives very nice profiles as judged by both the gamut volume and visual inspection using standard reference prints. Quite frankly most of the patch sets I've seen more resemble the PS printed image. I've printed out patch sets with both the ACPU and Photoshop CS4 (the last version that allowed one to print unmanaged patch sets). I'm on Windows so I cannot make any comment about MacOS issues. Something just doesn't look right to me.
EDIT Added: I would note that of course one would expect to see some deviation from what appears on a calibrated monitor. It has been my experience that the deviation at least for my 3880 is no where near that the OP sees.
-
Alan,
Thanks for your input on my original question based on your experience. We'll see what happens. I've printed both pages of patches with ACPU (1000 patches) and have sent them off to have the profile made. I'll update when I have the profile.
Brad
-
Profile Update:
I received the custom profile and it is fine. Problem solved. As several people noted here, the dark patch prints were, apparently, as they should have been for my specific printer, able to produce a good profile.
Thank you all for weighing in with your experiences.
Brad
-
Brad,
I can offer you to create a second profile, just to verify of for other papers. But you have to send the printed charts to Germany as I'm a german resident.
you don't have to pay for this offer except for target shipping
Wolfgang
-
Wolfgang,
I've now printed several different images with the custom profile I purchased and it is excellent. So I don't think I'll need a second, "confirming" profile.
None the less, thank you very much for your very generous offer. Offers like yours are a fantastic example of why Lula is such a wonderful place.
Brad Smith