1492: "Dress color"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: 1492: "Dress color"

Postby Diadem » Fri Feb 27, 2015 2:19 pm UTC

Qaanol wrote:The picture making the viral rounds clearly and unambiguously has colors that are, approximately, periwinkle and bronze. If it were under blue light, that would mean the original would be white and gold.

The dress in question doesn’t come in white and gold though. It does come in blue and black, however, which is exactly what you’d expect if the picture were under yellow light. And as it happens, you can look at the background of the viral picture to see that indeed the entire scene is washed with yellow.

I have seen lots and lots of people linking to this 'original dress'. It utterly confuses me. How is that the same dress? Both the shoulders and the lace pattern are completely different.

Apart from that: The colors in the photo are clearly gold and white (but a very blueish white). Just blow the picture up with an image editor and you can clearly see that. Of course that hardly tells us anything about the original color, because lighting has so much influence.
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

User avatar
Coyoty
Posts: 195
Joined: Wed Jun 06, 2012 5:56 pm UTC

Re: 1492: "Dress color"

Postby Coyoty » Fri Feb 27, 2015 2:25 pm UTC

Does this explain how Ceres looks like it has two bright spots on it, but may instead have two spots that are less bright than the rest of the planet, or are contrasting colors?

Oh crap, I just opened the door to the planet/asteroid/dwarf/Kuiper/moon debate...

User avatar
orthogon
Posts: 3006
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1492: "Dress color"

Postby orthogon » Fri Feb 27, 2015 2:28 pm UTC

OK, we just had a session over lunch, and it's stranger than I thought. Some people see black and others see gold when looking at the same image on the same screen at the same time. A subset of people who see gold see the other colour as white, but the vast majority see it as a pale blue. And unlike an optical illusion, none of us was able to "make it go back and forward" between the two colourings. It just is the colours it is, and incredulity accompanies the suggestion that it could be otherwise.
xtifr wrote:... and orthogon merely sounds undecided.

MOH
Posts: 3
Joined: Fri Feb 27, 2015 2:25 pm UTC

Re: 1492: "Dress color"

Postby MOH » Fri Feb 27, 2015 2:38 pm UTC

There's been about 15 of us in work arguing back and forth over this all morning. I've spent far more time than I ever intended googling pictures of dresses on amazon.

Leaving aside the odd olive/white, and lilc/sand suggestions, the majority have been in the white/gold camp, whereas a few of us see it as black/blue.
For me, that may come partly down to being first asked "what colour is the dress", rather than what colour it looks in the picture, so my brain automatically corrected for the lighting and I've been biased towards black/blue since.

While trying to convince one of the other camp by talking about the lightling, I was briefly able to see it as gold, but now bizarrely even though I know the picture actually contains gold I can't "see" anything but black. I can't see how it's white at all.

Back to the comic: I never would have realised both sides are the same colour, if someone hadn't pointed it out.

I'm expecting to wake up tomorrow to a Community-style civil war based on rival perceptions of the dress colours



Side note: Given the comic number, shouldn't it have been titled something like "1492 - Con dress of paradox"?
Last edited by MOH on Fri Feb 27, 2015 4:11 pm UTC, edited 1 time in total.

User avatar
mathmannix
Posts: 1415
Joined: Fri Jul 06, 2012 2:12 pm UTC
Location: Washington, DC

Re: 1492: "Dress color"

Postby mathmannix » Fri Feb 27, 2015 2:41 pm UTC

Here is the original image of the dress.
Spoiler:
Image

It looks blue and brown to me. While in my mind, this could mean that it is white and brown (or possibly gold), but the shadowing makes it look blue and brown, I cannot see it being black. It is either light blue and brown, or white and brown.
I hear velociraptor tastes like chicken.

User avatar
Adam H
Posts: 1267
Joined: Thu Jun 16, 2011 6:36 pm UTC

Re: 1492: "Dress color"

Postby Adam H » Fri Feb 27, 2015 2:43 pm UTC

orthogon wrote:OK, we just had a session over lunch, and it's stranger than I thought. Some people see black and others see gold when looking at the same image on the same screen at the same time. A subset of people who see gold see the other colour as white, but the vast majority see it as a pale blue. And unlike an optical illusion, none of us was able to "make it go back and forward" between the two colourings. It just is the colours it is, and incredulity accompanies the suggestion that it could be otherwise.
What I don't understand is how people can ONLY see it as black in the picture. It's one thing to be like "well yeah i can understand why you think it's gold but that's because of the lighting", but to refuse to accept that the photo shows gold (or at least a goldish color) is crazy. When you isolate the gold/black it still looks much more gold than black. Now if it was between gold and grey, then you'd have a legitimate argument, because the main argument against black is that it's such a light color.

Or are the gold-seers arguing that the dress cannot possibly be black because of the photo? That's just as crazy. Can't everyone just agree that the dress is actually blue and black (as confirmed by the photographer) even though the photo shows pale blue and muted bronze?

The trick to it being black is that it's very shiny, so it reflects the yellow light.
-Adam

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 2:50 pm UTC

mathmannix wrote:I cannot see it being black. It is either light blue and brown, or white and brown.

Try to start looking at it from the bottom-left corner of the picture.

speising
Posts: 2289
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: 1492: "Dress color"

Postby speising » Fri Feb 27, 2015 2:51 pm UTC

the blue isn't even pale. it's a well saturated violet. to WB this to white is absurd.
in any case, it's a lot darker than the backgroud of this website here.

User avatar
jc
Posts: 349
Joined: Fri May 04, 2007 5:48 pm UTC
Location: Waltham, Massachusetts, USA, Earth, Solar System, Milky Way Galaxy
Contact:

Re: 1492: "Dress color"

Postby jc » Fri Feb 27, 2015 2:57 pm UTC

Cauchy wrote:
slinches wrote:The proper plural form [is] math.

Whoa there. "Math" isn't a plural noun. You don't have one mathematic, two mathematics. "Mathematics" is a mass noun, and so is "math". "Maths" is some weird contrivance that shouldn't exist.


Funny how often this sort of peevery shows up. This one is really just one of the many examples of the spelling differences between the UK and the US. The "maths" abbreviation goes way back in the UK, but it's rare in the US.

Lately I've been doing some transcribing of old dance music books from the 18th and 19th centuries, and you see a lot of this due to the space limitations in the usually small "pocket-size" booklets. It's common to see longer words shortened to a few initial letters plus one or sometimes two final letters. The best-known modern example of this is "Mr" (with or without a final dot) for "Mister". These books use terms like "gentn" for "gentlem[ae]n", typically with the final 'n' as a superscript, often but not always with a dot below the 'n'. Because superscripts are so hard to type on computers, people often transcribe this as "gent.n", as a hint that the dot may be present, and it's understood that the final 'n' was printed above the dot. The most common examples of this are "first", "second" and "third", abbreviated to "1.st", "2.d" and "3.d", again with the letters elevated and the dot either below the 's' or 'd', or missing entirely. Other common examples are "mid.e" and "all.de", shortened forms of "middle" and "allemande". Shortening "mathematics" to "maths" is an example of this UK tradition, though you hardly ever see the 's' as a superscript nowadays due to the limits imposed by computers.

Such abbreviations mostly died out in the US around 150 to 200 years ago, but they're still common on the other side of the Pond.

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 2:59 pm UTC

speising wrote:the blue isn't even pale. it's a well saturated violet. to WB this to white is absurd.
in any case, it's a lot darker than the backgroud of this website here.

According to my Photoshop color picker, its saturation (in the original picture) is about 30%, sometimes (someplaces?) less.

User avatar
jovialbard
Posts: 613
Joined: Thu Apr 18, 2013 11:08 pm UTC

Re: 1492: "Dress color"

Postby jovialbard » Fri Feb 27, 2015 3:02 pm UTC

1) The comic has it backward. The dress isn't against a darker background making it appear lighter, it's in a lighter background.

2) There is no conceivable way that dress is white. There is literally white in the picture an inch away from the dress. Look at the cow pattern to the left, now back at the dress, now at the cow pattern, now at the dress. Are those the same color?

3) For that matter, look at the black on the cow pattern where it touches the black lace on the dress. Those are very close to the same color, though one is more textured. If the dress were gold and receiving the same light exposure as the cow pattern then it would very clearly be gold against the black of the cow pattern.

4) Let's look at where the "gold" is coming from, most notably the top of the dress. Do you see the dress curving out away from the camera or in toward the camera? Perhaps this is where the illusion lies? I see the dress curving out away from the camera, which tells my brain that obviously as it is curving back it is catching a light from above that doesn't fall on the rest of the dress as directly. The very bright, yellowish indoor lighting of the changing room or wherever we are is illuminating the black fabric in contrast to the less direct ambient light, making it look bronze. If you see it as curving in toward the camera then I suppose it would appear that the closer portion of the dress is catching the light from the flash and revealing it's true color while the rest of the dress appears black/dark because it is not illuminated as well. If that is the case then it's simply an old woman/young woman type spacial illusion that is also messing with our perception of color.

5) Someone on the internet is wrong!
I wrote a book exploring the nature of identity and loss of self! And another book about babies!
Also, vlog and short-story blog

User avatar
Keyman
Posts: 311
Joined: Thu Jun 19, 2014 1:56 pm UTC

Re: 1492: "Dress color"

Postby Keyman » Fri Feb 27, 2015 3:08 pm UTC

jc wrote: The best-known modern example of this is "Mr" (with or without a final dot) for "Mister".

WITHOUT a dot???? :wink:

But seriously... I think the point of the cartoon is not what the 'real' color of the dress, nor why some see it differently. It happens. People are weird. It's that the original frickin' picture made the frickin' national news as the frickin' lead story this morning. What the hell are people thinking...when there are llamas on the loose!!!!
Nothing could be more ill-judged than that intolerant spirit which has, at all times, characterized political parties. - A. Hamilton

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 3:13 pm UTC

jovialbard wrote:For that matter, look at the black on the cow pattern where it touches

On the image, it's not black, but brown. With RGB values around (100, 80, 60), sometimes even lighter.
Last edited by Kit. on Fri Feb 27, 2015 3:17 pm UTC, edited 1 time in total.

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1492: "Dress color"

Postby Whizbang » Fri Feb 27, 2015 3:17 pm UTC

I...

I just don't see white and gold, no matter how long I look at it. How is this even a debate?

User avatar
orthogon
Posts: 3006
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1492: "Dress color"

Postby orthogon » Fri Feb 27, 2015 3:17 pm UTC

Kit. wrote:
jovialbard wrote:For that matter, look at the black on the cow pattern where it touches

On the image, it's not black, but brown. With RGB values around (100, 80. 60), sometimes even lighter.

Dammit, it's turned charcoal grey now.
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
jovialbard
Posts: 613
Joined: Thu Apr 18, 2013 11:08 pm UTC

Re: 1492: "Dress color"

Postby jovialbard » Fri Feb 27, 2015 3:20 pm UTC

Kit. wrote:
jovialbard wrote:For that matter, look at the black on the cow pattern where it touches

On the image, it's not black, but brown. With RGB values around (100, 80. 60), sometimes even lighter.


True, which is a sort of very dark grayish brown. Though to my mind that's what you'd expect from a washed out black. If it were actually brown or bronze or gold and under a bright light like that, then you would expect it to be much more vibrant. There's something about seeing the blue as white that makes the washed out black look even more brown/gold...

eta: charcoal is a good description of that color orthogon
most "black" clothing isn't really truly black anyway, it's some kind of very dark brownish grey.
Last edited by jovialbard on Fri Feb 27, 2015 3:34 pm UTC, edited 1 time in total.
I wrote a book exploring the nature of identity and loss of self! And another book about babies!
Also, vlog and short-story blog

AbnerRavenwood
Posts: 2
Joined: Mon Oct 13, 2014 10:26 pm UTC

Re: 1492: "Dress color"

Postby AbnerRavenwood » Fri Feb 27, 2015 3:33 pm UTC

There are four lights.

HoopleDoople
Posts: 4
Joined: Mon Sep 23, 2013 6:35 am UTC

Re: 1492: "Dress color"

Postby HoopleDoople » Fri Feb 27, 2015 3:33 pm UTC

When I saw the original pictures I had no prompts and assumed the colors were blue and brown. The darker part I can see how the lighting makes the color ambiguous between black, brown, and gold (particularly at the top part of the dress) - had I been given the color pair prompts I would have likely identified it as black. No matter how hard I try, however, I cannot see the blue as anything but different shades of blue. This is true for both the original picture and the comic. I find this disappointing as my favorite optical illusions are the ones that you can make go back and forth between different ways of viewing them.

Those who saw the dress as white and gold initially, after learning the dress was actually blue and black are you:

1) Still only able to see the dress as only white and gold even though you know it is incorrect?
2) Now only able to see the dress as blue and black?
3) Able to go back and forth between the two color pairs?

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 3:45 pm UTC

jovialbard wrote:If it were actually brown or bronze or gold and under a bright light like that,

How about this?

Spoiler:
81A.jpg
That's the same picture, but with Kodak 81A filter applied (a filter that can be used to remove blue cast from sky-lit photos). Does it still look like the object in a bright light?

richP
Posts: 190
Joined: Wed Aug 17, 2011 3:28 pm UTC

Re: 1492: "Dress color"

Postby richP » Fri Feb 27, 2015 3:46 pm UTC

Wow, this is one of those "only on xkcd" threads. Start with a comic that's not funny or sarcastic, but rather a "hey that's neat" type. Step into the forums expecting a real sleepy discussion and find that we have:
+ Arguments about the original picture from who knows where.
+ Technical banter about white balance and other photography issues.
+ Paging all grammar Nazis.

And here we are, two pages in after discussing what shade of black a dress is, and nobody's managed to even make a weak 50 Shades of Gray reference? Ok, I'll try. "The Red Room of pain is actually an oversaturated light puce. Discuss...."

User avatar
Keyman
Posts: 311
Joined: Thu Jun 19, 2014 1:56 pm UTC

Re: 1492: "Dress color"

Postby Keyman » Fri Feb 27, 2015 3:48 pm UTC

HoopleDoople wrote:When I saw the original pictures I had no prompts and assumed the colors were blue and brown. The darker part I can see how the lighting makes the color ambiguous between black, brown, and gold (particularly at the top part of the dress) - had I been given the color pair prompts I would have likely identified it as black. No matter how hard I try, however, I cannot see the blue as anything but different shades of blue. This is true for both the original picture and the comic. I find this disappointing as my favorite optical illusions are the ones that you can make go back and forth between different ways of viewing them.

Those who saw the dress as white and gold initially, after learning the dress was actually blue and black are you:

1) Still only able to see the dress as only white and gold even though you know it is incorrect?
2) Now only able to see the dress as blue and black?
3) Able to go back and forth between the two color pairs?

I've tried the "Magic Eye" trick. I've tried the "look at the lower left corner". I've tried backing away from my monitor. It is white and gold, and I therefore I take egregious exception to "even though you know it is incorrect". I do not "KNOW" that. Because is isn't.
Nothing could be more ill-judged than that intolerant spirit which has, at all times, characterized political parties. - A. Hamilton

User avatar
jovialbard
Posts: 613
Joined: Thu Apr 18, 2013 11:08 pm UTC

Re: 1492: "Dress color"

Postby jovialbard » Fri Feb 27, 2015 3:51 pm UTC

Kit. wrote:How about this?
...
Does it still look like the object in a bright light?

Yeah, it does, but also the filter made the "black" fabric even blacker, it's now more like (63,54,49) toward the bottom away from the direct bright light exposure, both darker and less brown than the original picture was.

eta: Keyman, did you try looking at the cow pattern to the lower left? Is the white on that cow pattern the same as the dress?

eta2: I realize it may not really be a cow pattern, but when I see black and white splotches I think cow, sorry fashionistas.
I wrote a book exploring the nature of identity and loss of self! And another book about babies!
Also, vlog and short-story blog

FancyHat
Posts: 341
Joined: Thu Oct 25, 2012 7:28 pm UTC

Re: 1492: "Dress color"

Postby FancyHat » Fri Feb 27, 2015 4:03 pm UTC

CharonPDX wrote:The illusion is that some people see it as "a blue and black dress in bright light, washing it out," and others see it as a "white and gold dress in a shadow, dimming it" The fact that the background is severely overexposed allows BOTH interpretations - the dress could be the one thing in the picture that is in shadow. As it is the only "dark" thing in the image, it is easy to jump to that conclusion.

I was suspecting much the same thing. And it's a testable hypothesis - science can be done!

As a first, rough draft for how it could be done:-

Find enough people who haven't been exposed to that dress, the photos, etc, and show them the photo and ask them two questions:-

  • Is the dress blue and black, or is it white, possibly with a blue tint, and gold?
  • Is the dress in shadow, or brightly lit?
Vary the order of the questions and the possible answers, and see if there's a correlation between answers that supports the shadow/brightly-lit perception hypothesis.

This is an opportunity to inform and educate the general public about science!

Edited to add: Also, the lighter material: is it slightly shiny, or is it matt?

I'm a white, possibly with a blue tint, and gold, perhaps dark gold, person, and the impression I get is that the dress is in shadow, in relatively poor light compared to the brightly-lit, over-exposed background. And I perceive the light parts of the dress to be slightly shiny, and not really matt.

Edited to also add: Also, the gold (black) bits look slightly shiny to me, too.
Last edited by FancyHat on Fri Feb 27, 2015 4:13 pm UTC, edited 2 times in total.
I am male, I am 'him'.

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 4:06 pm UTC

jovialbard wrote:Yeah, it does,

For me, it looks like an image with brightly-lit, overexposed background and underexposed foreground.

SuperCow
Posts: 13
Joined: Fri Jul 13, 2007 3:53 pm UTC
Location: Indiana

Re: 1492: "Dress color"

Postby SuperCow » Fri Feb 27, 2015 4:10 pm UTC

This is possibly the most ephemeral subject Randall has ever made a comic about.

Most xkcd strips have enough longevity that their relevance extends beyond the original social and cultural environment the strip was posted in.

When people look back at xkcd 1492 a year from now, will anyone know or care about the topic?

User avatar
orthogon
Posts: 3006
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1492: "Dress color"

Postby orthogon » Fri Feb 27, 2015 4:14 pm UTC

FancyHat wrote:Find enough people who haven't been exposed to that dress, the photos, etc,

There's your biggest problem, right there.

SuperCow wrote:This is possibly the most ephemeral subject Randall has ever made a comic about.

Most xkcd strips have enough longevity that their relevance extends beyond the original social and cultural environment the strip was posted in.

When people look back at xkcd 1492 a year from now, will anyone know or care about the topic?


Yeah, they'll remember the day our understanding of light, perception and psychology was turned on its head. Randall will be seen as a founder of the new era of physics. This very thread will be cited in academic papers for centuries to come. Yes, my friends, we are making history here.

That, or it's just a ridiculously slow news day. Isis, Schmisis.
xtifr wrote:... and orthogon merely sounds undecided.

SuperCow
Posts: 13
Joined: Fri Jul 13, 2007 3:53 pm UTC
Location: Indiana

Re: 1492: "Dress color"

Postby SuperCow » Fri Feb 27, 2015 4:16 pm UTC

Also, this strip makes me curious about how Randall works. Clearly he couldn't have created it prior to the day before it was posted, as the subject matter in question didn't exist yet.

Does he create all of his strips the day before their posting? Or, does he have a backlog of strips created ahead of time, and only interrupts the sequence with timely strips when some cultural event occurs that he deems strip-worthy?

User avatar
jovialbard
Posts: 613
Joined: Thu Apr 18, 2013 11:08 pm UTC

Re: 1492: "Dress color"

Postby jovialbard » Fri Feb 27, 2015 4:17 pm UTC

Kit. wrote:
jovialbard wrote:Yeah, it does,

For me, it looks like an image with brightly-lit, overexposed background and underexposed foreground.


Then how do you perceive the relative brightness at the center of the top of the dress? Where is that brightness coming from and why is it there if not from a light shining directly down on the dress rather brightly?

SuperCow wrote:When people look back at xkcd 1492 a year from now, will anyone know or care about the topic?


I think it will make me laugh. Only one ha though, and then I'll move on.
I wrote a book exploring the nature of identity and loss of self! And another book about babies!
Also, vlog and short-story blog

chris857
Posts: 100
Joined: Thu Dec 02, 2010 5:04 pm UTC

Re: 1492: "Dress color"

Postby chris857 » Fri Feb 27, 2015 4:21 pm UTC

The dress looks very blue + black/dusky grey/sort-of dusky gold to me. But the blue cannot be white to my eye.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1492: "Dress color"

Postby Klear » Fri Feb 27, 2015 4:29 pm UTC

speising wrote:i just took this badly overexposed photo of my desk phone. which colour is it?
Spoiler:
Image


I see a window, a lamp, and part of a gramophone record showing a rocker in sunglasses.

edit to fix nesting, apparently. subsilver ftw
-Angua

User avatar
Ambignostic
Posts: 10
Joined: Fri Feb 18, 2011 3:01 pm UTC

Re: 1492: "Dress color"

Postby Ambignostic » Fri Feb 27, 2015 4:30 pm UTC

So are you telling me that Superman's hair really is black?
SwanFace2_custom.jpg
One for the blue hair crowd
SwanFace2_custom.jpg (33.28 KiB) Viewed 4035 times

User avatar
Exodies
Posts: 919
Joined: Wed Apr 18, 2012 12:36 pm UTC
Location: Blackwaterside

Re: 1492: "Dress color"

Postby Exodies » Fri Feb 27, 2015 4:31 pm UTC

There is a painted wall in London (maybe Marylebone High Street) which from any angle looks purple. However when you photograph it close up looking straight on the picture comes out sliver/grey. I've experienced this with two different cameras. This example is with an iPhone. No post processing has been done. Any ideas?

Image

Image

More on-topic: does her bum look both bigger and smaller in that dress?
Ware2 guv? The Book of Dave

DanMcQueen
Posts: 9
Joined: Tue Aug 31, 2010 12:22 pm UTC

Re: 1492: "Dress color"

Postby DanMcQueen » Fri Feb 27, 2015 4:35 pm UTC

Screw the dress... Randall, what about the llamas?

User avatar
Quizatzhaderac
Posts: 1605
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: 1492: "Dress color"

Postby Quizatzhaderac » Fri Feb 27, 2015 4:35 pm UTC

HoopleDoople wrote:Those who saw the dress as white and gold initially, after learning the dress was actually blue and black are you:

1) Still only able to see the dress as only white and gold even though you know it is incorrect?
2) Now only able to see the dress as blue and black?
3) Able to go back and forth between the two color pairs?
In the brightened photo, I see pure white and glossy gold. In the photo in question and the left of the comic I see white with a blue tint and mate gold. In the comic on the right and the darkened photo I see very dark gold and blue.

I am not able to make anything go back and forth. The only things that changed are seeing off-white/black where I saw simple white/black before and the the background in the original and brightened photos looks "wrong".
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Keyman
Posts: 311
Joined: Thu Jun 19, 2014 1:56 pm UTC

Re: 1492: "Dress color"

Postby Keyman » Fri Feb 27, 2015 4:46 pm UTC

jovialbard wrote:eta: Keyman, did you try looking at the cow pattern to the lower left? Is the white on that cow pattern the same as the dress?

I did. And I can perceive a difference in shade, not a difference in color. The 'white' in dress may possibly be a very light ice-blue but still, but as my eyes stray from that contrast point, it 'returns' to white. And there is no way the gold is black.

I am totally convinced people see different colors. My two daughters see it differently (while I didn't tell them which I saw, it's still possible they are messin' with their old man :evil:) But a friend sees it blue/black, and she wouldn't screw with me. I think the real interesting part of this is actually a separate question/issue from what color it actually "is", or what it really "should be" if the color balance was "correct". I know for people who look at the exact same image and see two separate and mutually incompatible color schemes.

Now this is purportedly the actual dress....
Image
(coincidentally posted under a comment from "llamaswill")

Does the blue/black faction, looking at the original picture, see this blue/black? Or different shades of blue?
Nothing could be more ill-judged than that intolerant spirit which has, at all times, characterized political parties. - A. Hamilton

Kit.
Posts: 1099
Joined: Thu Jun 16, 2011 5:14 pm UTC

Re: 1492: "Dress color"

Postby Kit. » Fri Feb 27, 2015 4:48 pm UTC

jovialbard wrote:
Kit. wrote:
jovialbard wrote:Yeah, it does,

For me, it looks like an image with brightly-lit, overexposed background and underexposed foreground.

Then how do you perceive the relative brightness at the center of the top of the dress? Where is that brightness coming from and why is it there if not from a light shining directly down on the dress rather brightly?

The brightness is coming from the dress, of course. Is it related to how the dress is illuminated? Yes, probably, but there could be different reasons for that.

The question you might want to ask is: "Do you see the sharp shadows on the dress, the shadows that are not normally associated with blue cast?".
My answer would be: "No, unless I pay too much attention to this dull subject."

bittyx
Posts: 194
Joined: Tue Sep 25, 2007 9:10 pm UTC
Location: Belgrade, Serbia

Re: 1492: "Dress color"

Postby bittyx » Fri Feb 27, 2015 4:56 pm UTC

Exodies wrote:There is a painted wall in London (maybe Marylebone High Street) which from any angle looks purple. However when you photograph it close up looking straight on the picture comes out sliver/grey. I've experienced this with two different cameras. This example is with an iPhone. No post processing has been done. Any ideas?


Assuming the wall *is* purplish or whatever (as in the first picture), the second one might be the cameras' automatic white balance getting weirded out for some reason?

User avatar
Exodies
Posts: 919
Joined: Wed Apr 18, 2012 12:36 pm UTC
Location: Blackwaterside

Re: 1492: "Dress color"

Postby Exodies » Fri Feb 27, 2015 4:59 pm UTC

Thank you. I will return with a manual control camera in two weeks.
Ware2 guv? The Book of Dave

User avatar
jovialbard
Posts: 613
Joined: Thu Apr 18, 2013 11:08 pm UTC

Re: 1492: "Dress color"

Postby jovialbard » Fri Feb 27, 2015 5:06 pm UTC

Keyman wrote:Does the blue/black faction, looking at the original picture, see this blue/black? Or different shades of blue?


Yes, those are the colors I perceived the dress to be the first time I looked at the dress, but washed out by the very bright yellowish light shining down on the dress from above making the colors lighter and shifted toward yellow. And of course lightening black makes it look grey and shifting grey toward yellow makes it look brown, but my mind perceived that lighting effect and interpreted it as those colors.

eta: so the dress looks to be the color light grey to you, which you perceive to be white in a shadow?
I wrote a book exploring the nature of identity and loss of self! And another book about babies!
Also, vlog and short-story blog

speising
Posts: 2289
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: 1492: "Dress color"

Postby speising » Fri Feb 27, 2015 5:12 pm UTC

Keyman wrote:Now this is purportedly the actual dress....
Image
(coincidentally posted under a comment from "llamaswill")

Does the blue/black faction, looking at the original picture, see this blue/black? Or different shades of blue?


different shade of blue, obviously, because whitebalance. a bit like this:
tumblr_inline_nkeezsjAuH1svicb3.jpg
whitebalance changed and overexposed


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: gmalivuk and 151 guests