[HN Gopher] Lightfastness Testing of Colored Pencils
___________________________________________________________________
Lightfastness Testing of Colored Pencils
Author : picture
Score : 193 points
Date : 2025-07-04 23:15 UTC (3 days ago)
(HTM) web link (sarahrenaeclark.com)
(TXT) w3m dump (sarahrenaeclark.com)
| hoppyhoppy2 wrote:
| https://web.archive.org/web/https://sarahrenaeclark.com/ligh...
| gennarro wrote:
| Very spammy images. Is this a legit resource? The test is
| interesting though 6 hrs/day in sunlight is very extreme for
| artwork.
|
| Also I'm wondering is a fixer would help or hurt the testing.
| This is common with some art, like pastels.
| hoistbypetard wrote:
| What did you find spammy about the images? The ads for the
| artist's coloring books and calendars and such?
| ethan_smith wrote:
| Accelerated testing (6hrs/day) is standard practice in
| materials science - it compresses years of normal exposure into
| months while maintaining relative degradation patterns.
| Fixatives might alter results by adding UV inhibitors, but most
| artists want to know worst-case baseline performance.
| hinterlands wrote:
| Although someone will challenge me on that, I'm 100% sure that
| large chunks of the text are AI-generated. That said, the
| website itself has been around at least since 2017 (the text
| just wasn't as verbose - e.g., https://sarahrenaeclark.com/diy-
| gift-bag/).
|
| So, I suspect it's legit. It's a case of an author leaning on a
| crutch for writing, but we're here to judge the results, not
| the phasing.
| humblebeekeeper wrote:
| Why do you think it's AI? It reads to me like someone who has
| a special interest and a data driven mindset.
|
| I've seen plenty of people "rate every X" in youtube videos
| or blogs before, this one is just more data oriented than
| most.
| hinterlands wrote:
| First, it just reads that way. It's the default style if
| you ask ChatGPT to write a couple of paragraphs that
| explain why lightfastness is important.
|
| Second, while I know there are reasons to be skeptical
| about AI text checkers, the author's earlier (less verbose)
| style doesn't get flagged at all, while the style in more
| recent articles gets classified as heavily AI-assisted.
| humblebeekeeper wrote:
| > First, it just reads that way. It's the default style
| if you ask ChatGPT to write a couple of paragraphs that
| explain why lightfastness is important.
|
| It doesn't read that way to me, and I've read lots of
| ChatGPT text. We've come to opposite conclusions, I'm
| curious what qualities you are identifying/keying off of?
| capnrefsmmat wrote:
| In our studies of ChatGPT's grammatical style
| (https://arxiv.org/abs/2410.16107), it really loves past
| and present participial phrases (2-5x more usage than
| humans). I didn't see any here in a glance through the
| lightfastness section, though I didn't try running the
| whole article through spaCy to check. In any case it
| doesn't trip my mental ChatGPT detector either; it reads
| more like classic SEO writing you'd see all over blogs in
| the 20-teens.
|
| edit: yeah, ran it through our style feature tagger and
| nothing jumps out. Low rate of nominalizations (ChatGPT
| loves those), only a few present participles, "that" as
| subject at a usual rate, usual number of adverbs, etc.
| (See table 3 of the paper.) No contractions, which is
| unusual for normal human writing but common when assuming
| a more formal tone. I think the author has just affected
| a particular style, perhaps deliberately.
| wzdd wrote:
| Tangent, but I'm curious about how your style feature
| tagger got "no contractions" when the article is full of
| them. Just in the first couple of paras we have it's,
| that's, I've, I'd...
| capnrefsmmat wrote:
| Probably because the article uses the Unicode right
| single quotation mark instead of apostrophes, due to some
| automated smart-quote machinery. I'll have to adjust the
| tagger to handle those.
| ezconnect wrote:
| She didn't even promoted a single pencil brand on the
| conclusion. She just shows the data and let the viewers decide.
| humblebeekeeper wrote:
| There's:
|
| 1 header image
|
| 1 image showing in process
|
| 1 image explaining lightfastness
|
| 3 images explaining the importance of lightfastness
|
| 1 image explaining the measurement process
|
| 1 image linking to another article diving much deeper into the
| methodology
|
| 1 image linking to another article on a different color pencil
| concern (layering)
|
| 1 image representing each brand-line's lightfastness
|
| Every single one of those images seemed relevant to the concept
| presented and clarified something that would have been
| difficult to articulate succinctly in writing. For example, the
| "how was this measured" is a lot easier to understand once
| you've seen the grid of squares before and after than it would
| be to try and articulate the fading of colors in small squares
| in text.
|
| There's LOTS of individual images on specific brands, but given
| their wild degree of variance, I think it's really useful to
| perceptually see what's going on with each one.
|
| I'm curious, where do you feel the images were "spammy"? It's a
| conclusion I _heartily_ disagree with, but would love to
| understand.
| dcrazy wrote:
| I think gennarro is reacting to the very SEO-friendly
| organization of the article. Every content farm produces
| articles with this kind of flow, often with Wikipedia-style
| tables of contents at the beginning. But they do it because
| it's very similar to the structure an actually informative
| article would take! So we can't tell for sure whether the
| author adopted an SEO-friendly structure for her informative
| and original content, or if her content just happens to be a
| good model of the style that content farms have chosen to
| imitate.
| hoistbypetard wrote:
| I really enjoyed reading this even though I have no direct
| interest in the lightfastness of the pigments in colored pencils.
|
| It was just fun to see what someone who is deeply invested
| thought important to test, explain and research about something
| I'd have previously called a matter of aesthetic preference (as
| opposed to a thing you can benchmark).
| sunrunner wrote:
| > test, explain and research about something
|
| There's more high quality engineering discipline in this 'non-
| engineering' article than in seemingly a lot of self-professed
| software engineering today ;)
| ChrisMarshallNY wrote:
| Heh. As a [former][0] artist (and musician), and an engineer,
| I can confirm that the Venn diagrams overlap, quite a bit.
|
| Pigment color is a real heavy-duty field. There's a guy named
| Michael Wilcox[1] that is famous for his work on pigment
| color.
|
| [0] https://news.ycombinator.com/item?id=40917886
|
| [1] https://michaelwilcoxschoolofcolour.com/about-michael-
| wilcox... (Has an annoying popup on every page).
| sunrunner wrote:
| Thank you for these links!
|
| > Pigment color is a real heavy-duty field
|
| I'm not surprised. As $DAYJOB involves a reasonable amount
| of requirements for colour accurate previewing in a print
| context I still feel like I'm never as sure about all steps
| of the colour pipeline as I should be, and this adds a
| whole new area to know I'll never feel like I know enough
| about.
| kedihacker wrote:
| Michael Wilcox's site has virus on every click or link it
| redirect to a casino or some other site at the end.
| ChrisMarshallNY wrote:
| Sadly, I'm not surprised (looks like the site was written
| twenty years ago). It must be one of those "roulette"
| ones, because I have not encountered it (or Safari kills
| it), but it does have that stupid region popup.
|
| Here's a video interview with him:
| https://www.youtube.com/watch?v=nd3hCBPqzYU
| thread_id wrote:
| This is a really nice work product. I totally appreciate the
| value of this.
| sllabres wrote:
| What a effort. I did something similar for some pens and
| different inkjet colors a long time ago, but not nearly as broad
| or as methodical. The inkjet inks (especially red) were already
| blown after a short time (>4 months). But black still holds up 20
| years later till now, only a little bit faded, so one can see the
| tracks of each row.
|
| Foils (laminate or adhesive foils) or protective spray (UV) did
| not change the result at all. But one film tore and gave the
| whole thing an interesting, crackle-like appearance. However, the
| colors all faded in the same way, whatever protective used
| compared to direct exposure to sunlight.
| andrewla wrote:
| I've noticed this as well -- at one point I had switched from
| Canon inkjet refills to a generic third party refills, and
| years later the photos from the knockoff have faded to an
| astonishing degree while the Canon ink prints look bright and
| vibrant.
|
| Going through old photo albums that my parents have show a lot
| of fading as well, even though the pictures themselves were
| kept in photo albums in the dark for many years. We have
| negatives for some of those photos, which when scanned are
| bright and vibrant, but the prints vary significantly.
| dylan604 wrote:
| uh-oh, did you just anecdotally confirm that buying first
| party inks is the better decision? careful, you might just
| confirm to the makers that their DRM policies are legit.
| PaulHoule wrote:
| Third party ink is crap. On various forums you will read
| stories about people who attempt borderless printing and
| have terrible inksplosions. Quite often they are using
| third party inks and get no sympathy from people who are
| serious about inkjet printing.
|
| Inkjet manufacturers put a lot of effort into ink
| formulations and often these are better in terms of VoC and
| other parameters. Sometimes you find certain first party
| inks are not at all lightfast (like the ink for the Epson
| ET-3750 which I found fades badly in less than six months)
| but there is very little independent testing of ink
| performance. If there was there might be a market for ink
| that performs _better_ than first party ink.
| bayindirh wrote:
| If I remember correctly, HP produces pigment based color
| inks for their more serious inkjet lines, and these
| systems generally use 4 or more cartridges.
|
| However, even though I balk at my 4515 for having dye
| based color inks, their fade resistance is astonishing.
|
| HP always uses pigment based black inks on every inkjet,
| so text is always at archival quality on good paper.
| PaulHoule wrote:
| I still like the tank so I upgraded to an ET-8550 which
| uses mainly dye-based inks except for a pigment black.
| Photos and art reproductions I print from that still look
| OK 2.5 years later of being on display, even in a south-
| facing room in my house which seems to be a good
| "accelerated aging" environment. Epson makes an Pro ET
| line which uses pigment-based ink, but when I tried to
| get one during the pandemic I couldn't find one so I got
| the ET-8550.
| grues-dinner wrote:
| It's a bit of a pity because the printer makers have been
| trying so many scams that their actual valid selling
| points are also assumed to be a bamboozlement.
|
| But also, it's a fairly niche requirement to have a non-
| photo printed document need survive sunlight for years.
| I'd guess 95% plus of pages that get printed on a normal
| office printer have minimal need for colour precision,
| use crappy paper and go in the bin within the year or end
| up in an archive box, likely never to be seen by human
| eyes again.
| bayindirh wrote:
| I always say the same thing. HP and Xerox run their (ink
| and toner) labs _for a reason_ , and their ink might be
| expensive but it's high quality and dependable.
|
| I seldom print photos with my entry level HP inkjet, using
| HP inks. Even though my printer uses dye based color
| cartridges, none of the photos I print have faded even
| though they are framed and displayed 7/24/365. Ink is not
| colored water.
|
| If you print/read/shred in two days, 3rd party ink and
| lower quality paper is OK, but if you want to be able to
| read or look at that thing after 10+ years, you need better
| quality paper and inks.
|
| I use fountain pens a lot, and difference between ink
| quality becomes much more evident there. There are writing
| inks, there are archival inks (which are _not_ Indian
| inks), and they behave totally different.
| codazoda wrote:
| Crazy amount of effort here. Awesome.
|
| I did this for a single color from a single printer--the black
| toner from my Brother laser printer. I left it in my West
| facing office window for about 18 months. On the BACK SIDE I
| labeled it with pen. The pen on the back faded to almost
| nothing but the toner did not fade at all.
|
| I did not do monthly scans, that would have been a better
| "experiment", but I was satisfied that a B&W laser print would
| last a very long time.
|
| Maybe I should lightfast test my Brother Laser and my HP Inkjet
| (with Black Pigment based ink).
|
| I thought that pigment based inks would be both waterproof and
| lightfast. Since I started to airbrush watercolor over my HP
| prints I am now very aware that these pigment based inks are
| not waterproof, even after long drying times.
| wrp wrote:
| I did this for colored mechanical pencil leads. Short answer, use
| the Staedtler Mars Micro Color if you need lightfastness.
| dylan604 wrote:
| It's interesting how pink tends to be the worst of the colors
| according to the charts. I wonder what it really looks like on
| paper. Does it disappear completely so it is very hard to see, or
| is it just invisible in the data as it rounds to 0 but leaves
| something visible on the paper? If you did an image in pink
| duotone with the worst offenders, would you have a blank sheet of
| paper after 26 weeks? Or does it look like something done in
| white duotone?
| ChrisMarshallNY wrote:
| Red, in general, sucks.
|
| Look at all the red-white-and-blue bumper stickers. They are
| usually white-and-blue.
|
| Around here, we have school buses with a sticker on the back,
| announcing that they don't turn right on RED (with "RED" being
| in heavy letters, and colored red).
|
| They frequently say that they don't turn right on.
| PaulHoule wrote:
| Red ink absorbs high energy photons so yeah it tends to get
| trashed. Mineral reds are good (the red on a red metal barn)
| and some chemical reds but not others. I remember on Jan 1,
| 2002 walking around Binghamton where there were many white-
| grey-and-blue flag lawn signs all over town that had been
| hurriedly installed the prior September and going in to watch
| a parade on TV where an announcer said "You know what my
| favorite bumper sticker is? These colors don't run"
|
| The flag I got from the US Army in 1993 when my father passed
| away is still in great shape while many of them don't look so
| good.
| Cthulhu_ wrote:
| You can (could?) see it in a lot of red cars too, they fade
| to a colour that I can only describe as "faded red car". They
| either use a different, more stable paint nowadays or avoid
| red paint in general.
| humblebeekeeper wrote:
| As someone who is heavily tattooed, I'd LOVE to see this analysis
| for tattoo inks.
|
| Fun fact: UV light makes tattoo particles smaller, which makes
| them easier for your lymphatic system to carry them to your lymph
| nodes. The particles are easy to transport into the lymph nodes,
| but difficult for your body to remove from your lymph nodes,
| meaning that for heavily tattooed people like myself, surgeries
| can be a potentially very colorful endeavor! (Or, if you have
| primarily black tattoos, it can be a spooky endeavor, I suppose.)
| smokel wrote:
| I did similar testing with ball-point pens. Eventually I ran into
| the ISO 12757-2 standard on archival ink, thinking that it would
| be a great idea to use such pens for my intricate drawings.
|
| Turns out the drawings, some of which I actually sold, faded into
| oblivion within about a year. After slightly more careful reading
| of the actual standard, I learned that the drawings were supposed
| to be _archived_ , i.e. kept in a box or a drawer, and not to be
| framed for full-time viewing pleasure.
|
| The typical blue ink in the famous BIC ball-point pens (i.e. non-
| ISO 12757) turns black after some time of sunlight exposure,
| which seems fine.
| LeratoAustini wrote:
| Thanks for mentioning this. I'm just getting into ballpoints in
| my art (literally just posted my latest piece online this
| morning, using Schneider Slider pens which use ink conforming
| to ISO 12757-2) and have diligently sourced 'archival' inks
| without looking deeply enough into what it means (a lot of
| online discussions imply that it's the same as lightfastness).
|
| About BIC, I thought I'd seens some tests online showing that
| over time the ink yellowed and eventually disappeared, so I've
| been avoiding them.
|
| Clearly I need to look more deeply into it, especially for work
| I might sell.
| smokel wrote:
| With ball-point ink, I think (but haven't checked recently)
| that there is unfortunately _no_ safe option.
|
| Some contemporary artists sell digital prints, but if you are
| selling the real deal, at least ensure that it does not sit
| in full sunlight all day, because it will fade quickly.
|
| In most museums original prints and drawings are often
| exhibited for short periods in darkened rooms. I think this
| takes away from the pleasure, and I prefer to see my art as
| perishables, which one can enjoy for a few years. An
| additional problem is that paper is terribly fragile stuff.
| Framing it (acid free paper) properly (with acid free tape)
| helps, but can be very expensive and will affect the
| aesthetics of course.
|
| Some collectors of printmaking art store everything in a
| drawer and take it out once a year to enjoy it with a good
| glass of wine.
| interloxia wrote:
| I know very little about ink, but I'm a little surprised,
| especially for something marked as archival.
|
| My only point of comparison is an etching which is a few
| hundred years old. The ink has no obvious signs of fading. It
| had some light restoration about forty years ago but we usually
| have it on display, at least in my life time.
|
| When I've seen other prints of the same etching in galleries,
| they are usually in the same or slightly worse condition
| despite, probably, being kept much more carefully.
| chris_st wrote:
| There's a book where the author tested watercolors which my
| watercolor teacher said was _extremely_ controversial when it
| came out, but she was able to replicate some of his results, and
| so changed which colors she bought.
|
| I tried the colors she recommended, and got good results. I also
| tested various black pens, and found (25 years ago!) that Micron
| pens were colorfast. Some black (gel, IIRC) pens faded to a nice
| sepia.
| chris_st wrote:
| And OH YEAH THE BOOK: Michael Wilcox's "The Wilcox Guide to the
| Best Watercolor Paints".
| __mharrison__ wrote:
| Sad. I was raised that Prismacolors were the end all be all of
| colored pencils.
| bayindirh wrote:
| In short, Caran d'Ache has a name and that price tag _for a
| reason_.
| CGMthrowaway wrote:
| It's my understanding that natural pigments are in general less
| colorfast than synthetic ones. If one prefers the color quality
| of natural pigments, would choosing the "worst" pencils here be
| indicative of that?
| mattgrice wrote:
| Pinks and oranges are notoriously not lightfast. Unless you use
| cadmium, which is kind of not something you want to be handing
| out to people who might be licking their pencils. Ever notice why
| yellow cars always end up looking pale lemon-yellow? Lightfast
| non-toxic cheap red pigments aren't that easy.
|
| Many rose/fuschia colors are not lightfast. Pigments that are not
| permanent are called 'fugitive' in the arts. Rose madder and
| alizarin crimson are not lightfast. So are a lot of others:
| cochineal, geranium lake. That's why you see 'green' cherubs when
| you visit the Getty museum also why Van Gogh's paintings are not
| the color they were when he painted them. Most reds for art are
| now quinacridone or cadmium.
|
| One of Windsor and Newton's (oldschool paint supply manufacturer,
| its fun to browse history on their website) most expensive paints
| is 'rose madder geniune.' They claim it is permanent but I don't
| know how they did it.
|
| Note: x 'lake' means x dye turned into a pigment.
___________________________________________________________________
(page generated 2025-07-08 23:01 UTC)