Wilhelm Research Update
Dennis Yep
think this is
the 4th or 5th time. Does anyone know why?
Some "experts" consider Wilhelm to be a very credible and reliable
source of information, but
now I am not so sure.
LeftCoast C.C.
"Dennis Yep" <dy...@cas.org> wrote in message
news:3AD467BF...@cas.org...
Ted Felix
> Wilhelm Research postponed the "major update" on their web page again. I
> think this is the 4th or 5th time. Does anyone know why?
Sounds like they are just a bit behind schedule. I've done fade
testing. It's a lot of grueling, thankless work:
http://tedfelix.com/FadeTesting/
> Some "experts" consider Wilhelm to be a very credible and reliable
> source of information, but now I am not so sure.
I wouldn't hold the postponement against them. I might hold the
source of their funding against them, however. My understanding is that
it comes from the manufacturers.
--
Ted Felix
http://www.tedfelix.com
Ted Felix
David Chien
1. Too busy. Tests taking longer than expected.
2. Not enough time. On vacation?
3. Waiting until the makers of inkjets say the results are 'good enough' to
be published. After all, with Epson having seen the spectacle surrounding
the orange-fading issue re: their printers ala 870/1270/780/etc., maybe
they're paying Wilhelm a lot right now to wait until they can figure out how
to fudge the number to their desired values. In any case,
http://www.p-o-v-image.com/epson/ to read about this mess. Since they're
not a public company, who knows which companies they'll bend to.
4. Weird results requiring retests.
5. Trying to figure out the best test methods to use for inkjet print
longevity. Nobody's really done much in this area before, so it may need
finessing.
6. They're being paid to 'shut up'.
7. Misc. other problems like server crashes, PC with the tests crashed and
lost data, etc.
BHilton665
>some more ideas..
>3. Waiting until the makers of inkjets say the results are 'good enough' to
>be published. ... maybe they're paying Wilhelm a lot right now to wait until
> they can figure out how to fudge the number to their desired values.
>
>who knows which companies they'll bend to.
>
>6. They're being paid to 'shut up'.
You must be a real asshole to sit around thinking up stuff like this. Hope
Henry Wilhelm sues you for defamation of character (I'll forward your post to
him).
What totally deflates your nonsense is that Wilhelm has always been seen as a
strong willed, independent tester who publishes his results (and his methods)
and isn't afraid to stand up to those who attack him, even when the attacks
come from the most powerful players in the field.
The best evidence of this is his "relationship" with Kodak. When he published
his original book and longevity findings Kodak products didn't fare too well
compared to Cibachromes (now called Ilfochromes). I think Ciba prints were
rated at 29 years or so, the Kodak prints 14-20 years or less, depending on the
paper and process.
Kodak was pissed, challenged the results and commissioned another scientist to
run tests, which showed that prints on Kodak papers lasted 75 years. This is
on the web but if you read it carefully you'll see that the guy assumes much
lower light levels than Wilhelm and if you extrapolate the differences in light
levels you'll see that yes, the prints will last 5 times longer because he
exposed them to 5 times less light. Big surprise! You could say the same
thing about the Ilfochromes and then they'd be rated at almost 150 years.
The advantage of Wilhelm's tests is that all media are tested the same way and
establish which ones will last longer, relatively speaking. You can argue that
he assumes light levels many times higher than a museum would use to show
prints (true), so most of the products tested will actually last longer under
those conditions, but the relative rankings remain useful.
As for the Epson PGPP screw-up, no one saw that coming and his original number
of 9-10 years for PGPP was for prints under glass. It's to his credit that
he's revising his tests to take into account the failures exhibited with the
dye inks on coated papers and try to test for them.
Finally, I just got a blurb announcing a new "Metallic" paper from Kodak from
the printer who does my Light Jet prints. He sent a sample and says it's meant
to look like Ilfochrome classic polyester. Then he says "As far as longevity
is concerned, Kodak says it is 'on par with all other color papers.' A more
definite answer we cannot give as Kodak has not agreed to let Wilhelm test the
paper." Saying longevity is "on par with all other color papers" is meaningless
to me since other color papers have such a wide range of time-before-fade.
If he were getting "paid to 'shut up'" and fudge the numbers (as you allege)
you'd think he'd do it for the biggest photo paper company in the world. He's
a scientist who publishes how he runs the tests so anyone can repeat his work
and it's slanderous of you to say he is crooked.
Bill
Mark Morgan
-That only leads to confusion and mistakes.
I'm glad they are willing to wait until they get it right, rather than to
get it on time.
This is not unusual for in that line of work.
"Dennis Yep" <dy...@cas.org> wrote in message
news:3AD467BF...@cas.org...
Mickey Ward
"Ted Felix" <nos...@no.spam> wrote in message
news:3AD494C2...@no.spam...
> Sounds like they are just a bit behind schedule. I've done fade
> testing. It's a lot of grueling, thankless work:
No offense, but "a bit behind" schedule is the understatement of the year.
These test reports were originally scheduled, if memory serves correctly, to
be posted back in November!! That's 5 months!! Grueling? OK. I'll buy that,
but why post a date, only to be postponed again and again when the date is
finally reached? They are making fools of themselves, not to mention the
damage being made to their credibility.
I could accept a couple of postponements made in order to include recent
hardware arrivals, but this is ridiculous. Stuff comes out every day. That
kind of rationale will prevent the report from ever being posted.
>
> I wouldn't hold the postponement against them. I might hold the
> source of their funding against them, however. My understanding is that
> it comes from the manufacturers.
I think you hit the nail on the head there. I've grown very suspicious of
the reason for these repeated postponements. The only reasonable motive I
can come up with is to assist certain vendors in temporarily saving face,
and keeping sales rolling on a big inventory of problematic hardware, at
least until they've had time to either rectify the problems, or release
improved replacement models. In the case of Epson, neither has occurred with
respect to their 12xx line of photo printers plagued with "fading" issues.
B.Server
>>From: David Chien chi...@uci.edu
>
>>some more ideas..
>>3. Waiting until the makers of inkjets say the results are 'good enough' to
>>be published. ... maybe they're paying Wilhelm a lot right now to wait until
>> they can figure out how to fudge the number to their desired values.
>>
>>who knows which companies they'll bend to.
>>
>>6. They're being paid to 'shut up'.
>
>You must be a real asshole to sit around thinking up stuff like this. Hope
>Henry Wilhelm sues you for defamation of character (I'll forward your post to
>him).
Appropriate; as you seem to have an emotional investment in his work
second only to himself.
Personally, I am surprised that we are waiting tests since I have seen
his name lauded high and low by the digirati to the effect that
Cibachrome is dead, long live the inkjet and the fading problem is
merely an artifact belonging to our sad, misguided film days of yore.
Maybe the reason that he has no published new material is that the
things he is testing are so perfect that they never fade and thus
cannot be measured in a convenient length of time?
RAS
real problem is that his whole basis for estimating print life is in total
disarray.
His test method uses long exposure to light to try to simulate the effect of
years of light exposure. He now admits to at least three problems.
First, as we all know from the Epson orange problem, he did not do testing
for the effects of contaminants in the air.
Second, he did not test for the problems that resin coated black and white
papers experienced from chemical reactions over time under the coating.
Third, he now says there is a reciprocity effect on prints. In other
words, one ten hour exposure is not the equivalent of a one hour exposure on
each of ten days. Anyone who has printed in a darkroom knows that there is
a huge reciprocity effect on exposing paper. Wilhelm now seems to think
that also applies to the processed prints.
I wish he well in sorting this out, and I would love to see others working
in this area. At present there is almost nothing to rely on and the
manufacturers have even less credibility than Wilhelm.
"Mickey Ward" <micke...@nospam.msn.com> wrote in message
news:lL6B6.220$5B2....@eagle.america.net...
Bob Salomon
While Wilhelm does charge app. $2500.00 per ink/paper/printer test for
test results that he allows to be used for publication he also tests on
his own and those results can not be used for publication.
Assuming that he is postponing his results for the reason you stated
below the result would be that the other vendors that pay him for
results would crucify him if he was postponing posting for the benefit
of other manufacturers.
He is a testing lab not a tool for the use of only select customers to
the detriment of others.
Mickey Ward wrote:
>
SNIP
> I think you hit the nail on the head there. I've grown very suspicious of
> the reason for these repeated postponements. The only reasonable motive I
> can come up with is to assist certain vendors in temporarily saving face,
> and keeping sales rolling on a big inventory of problematic hardware, at
> least until they've had time to either rectify the problems, or release
> improved replacement models. In the case of Epson, neither has occurred with
> respect to their 12xx line of photo printers plagued with "fading" issues.
>
> >
> > --
> > Ted Felix
> > http://www.tedfelix.com
--
HP Marketing Corp. 800 735-4373, US distributor for: Ansmann, Braun,
CombiPlan-T, Ergorest, Gepe, Gepe-Pro, Giottos, Heliopan, Kaiser, Kopho,
Linhof, Rimowa, Rodenstock, Sirostar, Tetenal ink Jet Papers and
cleaning cloths, VR Frames, Wista, ZTS
Ted Felix
> Third, he now says there is a reciprocity effect on prints.
Actually he's pointed that out all along. It's nothing new, and it's
why I gave up on my tests. They just aren't even remotely realistic.
David Chien
>
> Actually he's pointed that out all along. It's nothing new, and it's
> why I gave up on my tests. They just aren't even remotely realistic.
Well, any sort of early failure in the inks/papers through accelerated
testing will reveal indications of what components may fail early in real
life. While they may not pinpoint the 'exact' range of years/months these
prints will last, they, along with previous histories of printed materials
longevity, can be used to suggest an expected lifespan of prints.
Naturally, they'll have to be adjusted as time goes on to match real-life
results, and just like how people found out how long it took film to fade
in real-life, maybe we'll have to wait as long to find out just how long
inkjet prints really last.
In any case, premature failure of the inks/papers vs. conventional
printed materials that can easily last 100+ years, will put an upper limit
on the expected lifespan on inkjet printed materials. This alone can
provide valuable insight for those users who expect archival output from
their equipment.
Tests may not be perfect, but they can provide useful information that
will get you 'ballpark' figures to work by.
d =)
David Chien
> you'd think he'd do it for the biggest photo paper company in the world. He's
> a scientist who publishes how he runs the tests so anyone can repeat his work
> and it's slanderous of you to say he is crooked.
Never said he was crooked - only going through the possibilities.
After all, there aren't that many comapnies out there doing the same sort of
longevity tests, and if there are no other tests from other companies to compare
the results to, and if its a private company, the possibility, however slim, of
manipulation of the results could occur -- after all, it's the reason many
scientists have their results peer reviewed before and after publication.
Unfortunately, most of us do not have the equipment or money to run similar tests
to backup his claims, so who really, really (100%) knows for sure? Do you know?
Again, I'm not implying anything with the possibilities, just listing them.
But, after all, many magazines write up 'nice' reviews of many products based on
their sponsors' desires, and anybody in the public relations/marketing departments
of big magazines all realize that's what goes on in real life.
As for the Kodak thing, maybe FujiFilm got to them first? It's a
possibility....
--
But, we all do have to question his test methods and results today --- why was
it he/his company didn't see one sign of the orange-fading problem yet many Epson
870/1270/etc. users including myself easily saw it occur within days/weeks?
Gee, there must be a big ==reason== Epson called me directly to offer a
full-buyback of the printer on this note, as well as many others (see
groups.yahoo.com for the Epson Inkjet Mailing list archives as well as
www.p-o-v-image.com/epson/).
Yes, it maybe that all the prints were under glass, but still it points to
defects in his test methods, methods that perhaps haven't been fully thought
through initially.
And why is it that they've already got 'preliminary' Canon S800 longevity
results posted, but not a single note re: the Epson 870/1270s? How could they
apply their 'new' testing methods to the just released Canon inks & papers and say
'26-28 years...' yet nothing at all about the year old Epson 870/1270?
Hmmm, could they be 'working' with Epson to hide the results until they can
either 1) fix the orange-fading problem, 2) wait for the angry 870/1270 owners
storm to pass 3) change the tests to make their papers appear better?
--
Again, we don't know because it's a private company that reports to no one at
all, a company w/o open review and full disclosure of the exact test methods (we're
talking about every bit of detail on every piece of equipment, setup, etc. used so
anyone can duplicate their tests exactly), and they've not fully explained how such
orange-fading problems got past them. But, if you start going through the
possibilities, some of them could make sense.
Hopefully, none of the possibilities is true, and Wilhelm's latest report will
provide significant, useful insight into the problems we've had and continue to
produce usable, reliable data we can all rely on. However, like anything that
hasn't been duplicated by others & uses methodology that hasn't been fully
revealed, we really won't know for 100% certainty whether the results are valid --
just being a good scientist here with skeptism.
d =)
Ted Felix
> Tests may not be perfect, but they can provide useful information that
> will get you 'ballpark' figures to work by.
Unfortunately, I never bought Wilhelm's book, but I did read it. My
favorite graph was the one that showed how the accelerated testing
indicated one paper would outlast another, but the un-accelerated
testing showed quite conclusively that it was the other way 'round.
Ted Felix
> As for the Kodak thing, maybe FujiFilm got to them first? It's a
> possibility....
No, I think Kodak was the first huge print life catastrophy. So, they
are always the first to be mentioned. Yeah, they probably covered it
up, so I wouldn't give them too much credit. But it is indeed hell to
be on top.
> But, we all do have to question his test methods and results today --- why was
> it he/his company didn't see one sign of the orange-fading problem yet many Epson
> 870/1270/etc. users including myself easily saw it occur within days/weeks?
I have Wilhelm's "Inks and Media for Desktop Inkjet Printers, Years of
Print Display Before Noticeable Fading Occurs" report from 6/14/2000.
It does indeed mention the 870/1270 and it goes into the orange fading
problem in detail. Too bad his website is basically down, or you might
be able to find it in the archives there. But from what I remember, he
doesn't keep a very good archive of his reports online. Let me know if
you want me to email this report to you.
> Yes, it maybe that all the prints were under glass, but still it points to
> defects in his test methods, methods that perhaps haven't been fully thought
> through initially.
Absolutely. And that's probably why he's redoing the whole thing and
it's taking forever. Take it from one who's tried it. Fade testing is
endless and frustrating.
> And why is it that they've already got 'preliminary' Canon S800 longevity
> results posted, but not a single note re: the Epson 870/1270s? How could they
> apply their 'new' testing methods to the just released Canon inks & papers and say
> '26-28 years...' yet nothing at all about the year old Epson 870/1270?
See my comments above. 6/14/2000 he released 870/1270 data.
> Hmmm, could they be 'working' with Epson to hide the results until they can
> either 1) fix the orange-fading problem, 2) wait for the angry 870/1270 owners
> storm to pass 3) change the tests to make their papers appear better?
Wilhelm's tests are usually performed "under glass", so his longevity
figures for the 870/1270 are fairly long. From the 6/14/2000 report:
Epson Stylus Photo 870 and 1270 Printers (Epson Inks)
Epson Matte Paper - Heavyweight 24-26 years
Epson Premium Glossy Photo Paper 9-10 years(2)
Epson Photo Paper 6-7 years
The (2) is a footnote that explains the orange fading problem.
> just being a good scientist here with skeptism.
I'm sure we all expect no less.
Big Bob
make 'suggestive' accusations, then claim 'well, I didn't say that WAS the
case, just a possibility...' This is pure, "conspiracy theory" speculation
on your part, and you make it worse by your after-the-fact ideas on what
they "should have done" as though you would have known any better. Indeed,
if you're so wise and all-knowing, perhaps you could set up a lab and show
them how it's done. Do that, and succeed, and you'll be believable.
Until then I'll file it away with your "600+ ppi-$25000 printer-dual RAID
SCSI drive-digital not as good as a point and shoot" answers that are
totally unrealistic in the real world, but this one is just plain
ridiculous.
Big Bob
"David Chien" <chi...@uci.edu> wrote in message
news:3AD5EE0...@uci.edu...
Mickey Ward
"Bob Salomon" <bobsa...@mindspring.com> wrote in message
news:3AD4C653...@mindspring.com...
> Your idea is seriously flawed!
>
> While Wilhelm does charge app. $2500.00 per ink/paper/printer test for
> test results that he allows to be used for publication he also tests on
> his own and those results can not be used for publication.
>
> Assuming that he is postponing his results for the reason you stated
> below the result would be that the other vendors that pay him for
> results would crucify him if he was postponing posting for the benefit
> of other manufacturers.
>
> He is a testing lab not a tool for the use of only select customers to
> the detriment of others.
>
The fact that he has not yet posted his test results on his own website has
not prevented vendors from documenting them themselves on their own websites
to promote their products. Two examples are Epson and Canon, both of whom
have used his ~25+ year longevity claims for their 12xx and S800 printers,
respectively.
My guess is that there are more ugly facts that he is witholding that would
not serve the involved parties' best interests, regardless of how useful we,
the consumers, might find them towards making our purchasing decisions.
Bob Salomon
It must have something to do with black helicopters.
Or perhaps he simply isn't ready to announce results and is doing
further testing to try to eliminate the fiasco last year with PPG.
--
jbuch
The phenomena of failure of "accelerated testing" is that if stuff
obeyed linear equations and the equations were simple, then acceleration
of variables such as time swapping for intensity or temperature would
work.
When the relatioships are non-linear and otherwise complex, accelerated
aging can't be accomplished by swaps of intensity or temperature for
time. Instead, you end up exploring something that has no reflection of
what will actually happen with natural aging.
It is beginning to look or sound like exactly that is happening with
inkjet color prints.
One hopes that some of the really bright ink chemists are paying
attention, as something revealing has been revealed by the evident
breakdown of accelerated aging as a useful tool.
What is it? Can I use it? What can I do about it?
Jim
-----------------------------------------
Keepsake gift for young girls.
Unique and personal one-of-a-kind.
sartha
> What is Wilhelm?
A man who sold his soul.
Sartha
harrison numbugger
sartha wrote:
>
> > What is Wilhelm?
>
> A man who sold his soul.
>
how much did he get?
> Sartha
______________________________________________________________________
Posted Via Uncensored-News.Com - Still Only $9.95 - http://www.uncensored-news.com
With Seven Servers In California And Texas - The Worlds Uncensored News Source
Pavel Dvorak
No, that was Peter Schlemiel.
Pavel
David Chien
> >case, just a possibility...' This is pure, "conspiracy theory" speculation
> >on your part, and you make it worse by your after-the-fact ideas on what
Why would listing the possibilities = suggestive accusations? A cop can list
the possible reasons he's investigating someone, but it isn't accusing them of
anything.
Not considering all the possibilities means there are gaps in the thinking and
can lead to overlooking the obvious -- just look at what happened to the Mars
probe when it wasn't considered that there might be the use of English vs.
Metric measurement systems in the design of the software - a multi-million
dollar space debris.
Innocent until proven guilty, and no one, not even I, has proven anything yet
re: Wilhelm's delay.
Consipracy theory? Watching way too much X-files.... Listing possibilites
has nothing do to with consipracies - again, like if you've ever worked for a
big company like PC Magazine, you'll know that the coverage will be biased
accordingly towards the biggest ad buyers and sponsers. It's not like we
haven't seen politicians being 'bought' by big interests now, have we? Being
bought up or influenced isn't a conspiracy, just a possibility that may very
well be likely given how prevelent it has affected many markets and people
across a variety of fields.
Besides, why not post why you think Wilhelm's taking so long, eh?
>The phenomena of failure of "accelerated testing" is that if stuff
>obeyed linear equations and the equations were simple, then acceleration
>of variables such as time swapping for intensity or temperature would
>work.
>When the relatioships are non-linear and otherwise complex, accelerated
>aging can't be accomplished by swaps of intensity or temperature for
>time. Instead, you end up exploring something that has no reflection of
>what will actually happen with natural aging.
True. But that's only one part of what users want. They not only want to
know how long a print will last, but also, how stable are the inks and papers
they use. An ink that'll last centuries, but only in very strict conditions, is
of little use when it'll fade/decay outside that stabilized environment.
Mickey Ward
> Everything always seems to be a conspiracy.
>
> It must have something to do with black helicopters.
>
> Or perhaps he simply isn't ready to announce results and is doing
> further testing to try to eliminate the fiasco last year with PPG.
>
Yeah. I guess you're right. I'm sure he's a very nice man who wants nothing
more than to make the world a better place.
Bob Hickey
Bob Hickey
M5 -readout
some sample prints from each maker I might be interested in. I'll stick
these to a wall and in 2015~2020 or so I'll take a look at them and make a
decision.
Not likely 8-) but probably the only way to be really sure about what you
are getting!
Loring Palleske
Ted Felix wrote:
His test methods really only apply to fading due to UV exposure. The orange shift is
caused by exposure to ozone (so epson says). I imagine people who live in less
industrial areas may get more longevity.
sartha
"Dennis Yep" <dy...@cas.org> wrote in message
news:3AD467BF...@cas.org...
> Wilhelm Research postponed the "major update" on their web page again. I
> think this is
> the 4th or 5th time. Does anyone know why?
Clients pay Wilhelm for good news only.
Sartha
Ron Baird
Just a comment on what I have seen.
In my past, I was a traditional kinda guy, and I worked with film and paper.
In my travels around this technology as well as that of Inkjet etc., I have
visited the testing facilities at Kodak as regard paper testing, etc. I can
tell you that they include rooms that are unique in the world in terms of
size and light what kind of light can be exposed in them. There are quite a
few millions of dollars spent to test our products and make sure they stand
and deliver. Of course all of it done under the ISO and more. Of course,
there are a good many people that simply put prints in their office windows
(if they have them) but the bulk of the testing done is through true light
exposure. Testing under glass, in mount, and lots of other ways are noted
as well.
Since I am recently late to the party I am not sure what the issue might be
to prompt the questions, but I can assure you that the testing we do is
quite extensive and thorough.
Results in testing are always based on the kind of testing done and how it
is measured. Unless it is always the same for everyone, ISO standards, it
is not always the same. Just an observation.
OK, guess I better get on with things.
Talk to you all soon.
Ron B
Kodak
"Ted Felix" <nos...@no.spam> wrote in message
news:3AD5FADC...@no.spam...
BHilton665
>In my past ... I worked with film and paper.
>In my travels around this technology as well as that of Inkjet etc., I have
>visited the testing facilities at Kodak as regard paper testing, etc. I can
>tell you that they include rooms that are unique in the world in terms of
>size and light what kind of light can be exposed in them. There are quite a
>few millions of dollars spent to test our products and make sure they stand
>and deliver. Of course all of it done under the ISO and more.
Hi Ron,
Glad to get the Kodak side of this discussion. Could you check around and get
the official party line on why Kodak disagrees with Wilhelm's methods? And why
they won't let him publish results on products like the new "Metallic" paper?
>Since I am recently late to the party I am not sure what the issue might be
>to prompt the questions, but I can assure you that the testing we do is
>quite extensive and thorough.
OK, I think everyone can agree that Kodak does a lot of product testing. But
the question is basically why the Kodak results (and the Ilfochrome results,
and the watercolor painting results etc) are so much more optimistic than
Wilhelm's results.
>Results in testing are always based on the kind of testing done and how it
>is measured. Unless it is always the same for everyone, ISO standards, it
>is not always the same. Just an observation.
I think this was the root of the problem, prior to release of Wilhelm's book in
1994 (?) or so. All the manufacturers tested to various conditions and were
free to pick the tests that made their products look best.
The biggest advantage of Wilhelm's research for the consumer is that he tests
everything the same way, so at the least you know what the relative life-spans
of the prints might be. I think the biggest disadvantage is that practically
everyone agrees his test assumptions for light levels are a quite a bit higher
than what you'll see in a typical home or, for certain, in a museum. But to me
this just means his life-span numbers are probably very conservative. I can
live with that.
Before his book came out Cibachromes were rated by Ciba-Geigy at, if memory
serves me, around 90 years before fading. I think Kodak's Type C and Type R
prints (what most of us use) were rated by Kodak at 75 years or so (may be
wrong, this is from memory). Wilhelm rated them much shorter, 29 years for
Ciba's, 15-20 I think for Type C and R prints from Kodak. Watercolor paintings
were also rated pretty short, to the consternation of many collectors.
Here's an interesting web site on the Kodak page that describes why Kodak
thinks prints are good to 150 years, instead of the 15-20 years Wilhelm allots
them.
http://www.kodak.com/cluster/global/en/consumer/education/imageStability.shtml
The author, Robert McComb, is a well-respected authority with a Ph.D, a
"26-year veteran of the Library of Congress, Preservation Research and Testing
Office". If you read the page carefully you can see the veiled shots at
Wilhelm's methods. But look at what he actually says and compare it to
Wilhelm's methods and it's easy to see why they reach different conclusions ...
he doesn't actually say the prints last 150 years before fading, he says
"prints now commonly last generations before consumers notice a color change.
And they'll last 150-plus years before the change becomes objectionable." But
'generations' could mean as little as 40 years. And 'objectionable' is not
very scientific. Take a spectrometer to 'em and measure when they first begin
to fade. Wilhelm, on the other hand, does rate prints on when the first sign
of any fading appears, presumably a lot tougher test than "when consumers
notice a color change".
Also, look at the assumed viewing conditions ... McComb extrapolates his data
based on a viewing/storage environment of 72 F, 40% humidity, 120 Lux. Wilhelm
is slightly tougher for temperature and humidity (75 F, 60% RH), but
significantly more extreme for light levels, assuming 12 hrs/day at 450 Lux.
Everyone would agree that you'll see much shorter print life at 450 Lux than at
120 Lux.
This is the heart of the issue. If there were a single standard spelling out
exactly how the tests should be done and how to measure the results with
something objective like a spectrometer then consumers could accept the results
from different companies. But when companies can test to various conditions
and apply non-scientific standards for judging when color shift occurs then it
becomes mostly marketing hype.
Bill
Lugh-Clyde
agree with you that the tests are different. Alas, they are probably both
valid - based on their assumptions.
I guess the real question is: can accelerated testing be linearly
extrapolated to be exactly the same as non-accelerated testing. Kodak is
saying no and Wilhelm is saying yes. (I'm very sure that Kodak is doing some
accelerated testing; their current products haven't been out for 90 years.)
The only way to really tell is to run both to exactly the same standards and
compare. The problem with that is it takes 29 or 90 years. Along the way you
find out that some assumptions on both sides are wrong and you start a new
test. Another 90 years later...
Frankly, I'd like to see Kodak's testing and Wilhelm's. What I'm really
looking for is a relative comparison. Then I'll pick current products that
will do good, better, and best.
Clyde
Ted Felix
> I guess the real question is: can accelerated testing be linearly
> extrapolated to be exactly the same as non-accelerated testing. Kodak is
> saying no and Wilhelm is saying yes.
Actually, Wilhelm also says no. I don't have his book handy or I'd
give you the page number. He has a discussion of the "reciprocity
failure" that occurs when doing accelerated testing vs. slower
"real-life" testing. It was this discussion that caused me to stop my
fade testing since I realized the results were pretty worthless.
BHilton665
>That was a very interesting site. You also make some very good points.
Thanks.
>I guess the real question is: can accelerated testing be linearly
>extrapolated to be exactly the same as non-accelerated testing. Kodak is
>saying no and Wilhelm is saying yes.
I think the Kodak sponsored site also says "Yes" to accelerated testing.
Author McComb says "Responsibly gauging image stability at the 120 <Lux> level
requires accelerated tests which exposes prints to a prudent 5400 lux for about
9-12 months." I think the "prudent 5400 lux" is a veiled shot at Wilhelm, who
presumably uses much higher intensity light to shorten the tests to a few
months instead of "9-12 months".
McComb probably has a good point here. The greater the test intensity the more
likely you'll see reciprocity problems, I'd venture. But while 9-12 months may
have been OK for testing processes like Type C or R, the digital world is
moving so fast that Wilhelm would barely be publishing results on a product
before it's outdated, right?
>Frankly, I'd like to see Kodak's testing and Wilhelm's. What I'm really
>looking for is a relative comparison.
Or the Epson (or Ilfochrome or Canon or Somerset or whatever) products tested
to the same standard as Kodak's. That would be great and it's likely the
projected print lives would be a lot higher than Wilhelm's numbers, but right
now only Wilhelm tests various products to the same standards (or if others do
it they aren't publishing like he is).
>The only way to really tell is to run both to exactly the same standards and
>compare.
I agree, but right now Kodak won't let Wilhelm publish results on their
products :)
Bill
Edward Bigelow
first to fade in "normal" use.
Similarly the older "standard" Epson inks which faded less quickly in the
sun, still faded fairly quickly.
So it's not that accelerated testing isn't remotely realistic, it's just
that it has never been an exact process, and there are a lot of variables.
It can definately be a reasonable relative indicator of "fade-ability"
And it's not yet a science, even though Wilhelm sometimes measured years
before fade accurate to one decimal.
Ted Felix <nos...@no.spam> wrote in message
news:3AD574D8...@no.spam...