Conservation Display Ratings

[MHMG]

Hi Everyone,

I mentioned recently in a previous thread that I would let LL forum readers of "Printers, Papers, and Inks" know when I was ready to post a new method for rating the lightfastness of digital print media. I have just finished posting the results and a new article about "Conservation Display Ratings" on my website. You can find a news post about it on the news page at:

http://www.aardenburg-imaging.com

Conservation Display ratings are now beginning to appear in the Light fade test results list which can be reached by clicking on the first link on this page:

http://www.aardenburg-imaging.com/acceleratedagingtests.html

This brings you to the light fade test results list which shows what products I currently have in light fade testing and the current status of those tests.  If the products have reached sufficient exposure to calculate a Conservation Display rating, then the results are also located in the list and provide a nice "executive summary" of fading performance without having to download the full report (which you can also get by clicking on a chosen report in the column marked "test report".  

Some of the test results and new Conservation Display ratings are free to the public (click on the blue colored links).  Others are for members that are helping to produce the test samples and supplement the costs of the tests through subscription fees, all of which I mention only to make sure that folks realize there is indeed some free stuff there!

The AaI&A conservation ratings differ mainly from the consumer-oriented "display life" ratings because they track the early stages of deterioration when prints still remain in overall excellent condition rather than the point at which easily noticeable and often objectionable fading has occurred. That said, a much more thorough explanation of the Conservation Display ratings can be found in the article entitled "An Overview of the AaI&A Conservation Display Ratings" which can be found on the documents page of my website:

http://www.aardenburg-imaging.com/documents.html


As noted in my news article, this research has been a long time in development. Collectively, the I* metric and the Conservation Display rating concept are my "treatise tacked to the cathedral door" regarding the lightfastness of modern digital print media. Whether the industry will embrace the concepts remains to be seem, but I think that LL readers represent my key constituency.

Cheers,

Mark

[bill t.]

Just one quick comment, it might be a good idea to change the "worst 10" terminology to something else.  While initially perusing your article I stumbled across the long table of tests, and was shocked to see that several of my Epson 7800 combinations were among "the worst 10!"

Of course I simply hadn't absorbed the whole article or the meaning of the table.  But a photographic art buyer doing research might make the same misinterpretation upon trivial examination, we are after all used to seeing similar consumer oriented tables displaying ranking information.  "Worst" is a common ranking term.

Other than that it looks like you have a great system there!

[MHMG]

Just one quick comment, it might be a good idea to change the "worst 10" terminology to something else.  While initially perusing your article I stumbled across the long table of tests, and was shocked to see that several of my Epson 7800 combinations were among "the worst 10!"

Other than that it looks like you have a great system there!

I do understand.   The purpose of the rating justification column is to let people know at a glance which of criteria caused the lower and upper limit of the rating. Was it loss of tonal accuracy, loss of color accuracy, etc.   I had tinkered  with wording like "80%I* color for worst 10%" etc., and ended up dropping the second percent sign, then realized later it would cause some to think it was worst 10 samples when it is really worst 10%. So,yes, I'm still tinkering with this wording to try to make it clearer, and will definitely be coming up with a better phrasing.  But the science is correct, and that has been my main concern.

The whole concept of an exposure range is no doubt going to cause many people to take issue with the conservation display ratings.  Yet if I had "dumbed it down" further and just specified an average overall number, products that are really very different in fading performance could end up with similar ratings (as they are now in current industry tests). We need the range in order to inform how each system responds to image-specific content.


Best regards,

Mark

[Kirk Gittings]

I'm surprised this is not generating more discussion. As I understand the numbers they are significantly lower than Wilhelms estimates. Is this primarily do to a tighter definition of what constitutes significant fading or color shift?

[MHMG]

I'm surprised this is not generating more discussion. As I understand the numbers they are significantly lower than Wilhelms estimates. Is this primarily do to a tighter definition of what constitutes significant fading or color shift?

Yes, the Conservation Display ratings are based on tighter criteria for "acceptable" fade. The design goal was to track small but measurable changes that still enable the viewer to see prints in very good to excellent condition rather than in a more noticeably faded state that may or may not be objectionable depending on your personal standards and the intended application.

My test conditions may also be a little tougher than WIR test conditions due to use of a more blue-rich  full spectrum light source, but other inter-lab differences make it hard to know for sure just which lab uses the "harshest" environment.  In the absence of an ISO standard on light fade testing, the key here for any independent lab is to standardize on a test environment that does a decent job simulating real world conditions.  .

A third factor to consider is the non linearity of fading between different systems. When fading criteria are set loose, they allow many different pathways of change before reaching the test endpoint, and this can cause misranking of the products. Some systems will drop quickly to nearly the failure point but then settle down to hang in just above the limit for a longer period of time which leads to a better score than is actually deserved. In reality, the viewer will see the changes earlier in the "life cycle" of the print with a system that does this compared to one that does the opposite (i.e., maintains initial color well early on, but then "caves in" later). Yet both systems might end up with the same final rating.  The tighter conservation display criteria avoid the problem of system non linearities affecting rank results.  In other words, mathematicians will confirm that all fading curves look linear if you evaluate them in small enough discrete increments.


best regards,

Mark

[Kirk Gittings]

Thanks, I will be following your work with great interest. Competition in this field will be a great thing.