1814: "Color Pattern"

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

Moderators: Moderators General, Prelates, Magistrates

User avatar
Wee Red Bird
Posts: 187
Joined: Wed Apr 24, 2013 11:50 am UTC
Location: In a tree

Re: 1814: "Color Pattern"

Postby Wee Red Bird » Thu Mar 23, 2017 9:29 am UTC

Znirk wrote:Wikipedia suggests they do, although they don't go as far as going silent on the <é>:

/mwɑːrˈeɪ/; French: [mwaˈʁe]


The audio on the OED site sounds like she is loudly air kissing.

User avatar
orthogon
Posts: 2961
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1814: "Color Pattern"

Postby orthogon » Thu Mar 23, 2017 4:15 pm UTC

Another point in favour of this, compared to the Moray ones, is that describing the Moiré phenomenon in a rhyming couplet comprising two pairs of anapests is a damned sight harder than doing it for an eel. Of course, it would be even better if you could describe the effect as well as the cause, but you'd only have six syllables for each. There's a challenge.
xtifr wrote:... and orthogon merely sounds undecided.

Seamus
Posts: 9
Joined: Mon Sep 13, 2010 11:14 pm UTC

Re: 1814: "Color Pattern"

Postby Seamus » Thu Mar 23, 2017 4:17 pm UTC

When an eel bites your toe
and it just won't let go,
That's a moray!

User avatar
Heimhenge
Posts: 239
Joined: Thu May 01, 2014 11:35 pm UTC

Re: 1814: "Color Pattern"

Postby Heimhenge » Thu Mar 23, 2017 7:41 pm UTC

Keyman wrote:
capngwyn wrote:When you swim in a creek
and an eel bites your cheek
it's a moray...

Here in the upper midwest USA, that word is sometimes pronounced "crick".


Yeah, that's how I read it in my head (crick). In fact, I was expecting something a bit more painful coming at the end of that second line. :)

This comic reminded me of the first time I saw a moire pattern. Didn't know what it was called back then, but we were visiting some relatives who had a double screen door on their entrance. I walked past it and saw some motion in my peripheral vision. Spent some time moving my head around and watching the effect. Curiously, my cousin swore he saw nothing but maybe I wasn't describing what I saw clearly enough. Makes me wonder if the eye-brain system on some people is "immune" to this illusion? Any psychologists on this forum?

User avatar
GlassHouses
Posts: 129
Joined: Thu Nov 24, 2016 12:41 pm UTC

Re: 1814: "Color Pattern"

Postby GlassHouses » Fri Mar 24, 2017 12:48 am UTC

Heimhenge wrote:This comic reminded me of the first time I saw a moire pattern. Didn't know what it was called back then, but we were visiting some relatives who had a double screen door on their entrance. I walked past it and saw some motion in my peripheral vision. Spent some time moving my head around and watching the effect. Curiously, my cousin swore he saw nothing but maybe I wasn't describing what I saw clearly enough. Makes me wonder if the eye-brain system on some people is "immune" to this illusion?

But it's not an illusion! Those patterns are real!
I think your cousin was just blasé about Moiré.

User avatar
Heimhenge
Posts: 239
Joined: Thu May 01, 2014 11:35 pm UTC

Re: 1814: "Color Pattern"

Postby Heimhenge » Fri Mar 24, 2017 5:38 am UTC

What do you mean "real"? Sure, the intersection of grid patterns in overlaid screens is real. But the "wavy band" that seems to move across the screen when your head moves ... that's gotta be happening in the brain. Otherwise why wouldn't my cousin see it?

Mikeski
Posts: 1034
Joined: Sun Jan 13, 2008 7:24 am UTC
Location: Minnesota, USA

Re: 1814: "Color Pattern"

Postby Mikeski » Fri Mar 24, 2017 6:37 am UTC

Heimhenge wrote:What do you mean "real"?

Not an illusion. Something that can be photographed. Something that math works on. Something that's used in science and technology (moire strain gauges, electron microscopy).

Sure, the intersection of grid patterns in overlaid screens is real. But the "wavy band" that seems to move across the screen when your head moves ... that's gotta be happening in the brain.

Nope. You can photograph it. You can make videos of it (Yes, it's an advertisement... still pretty cool).

(And you don't need to use a digital camera, which can produce moire patterns of its own, due to the image sensor's pixel grid lining up with some other grid being photographed.)

Otherwise why wouldn't my cousin see it?

He didn't understand what you were talking about? He was standing at a different distance where the screens lined up differently? He was just messing with you? He was nearsighted and wasn't wearing glasses with a strong enough prescription to make out the pattern?

racemaniac
Posts: 3
Joined: Fri Apr 18, 2014 8:06 am UTC

Re: 1814: "Color Pattern"

Postby racemaniac » Fri Mar 24, 2017 11:53 am UTC

Anybody got any more verses for the it's a moiré?
The comic itself is already good inspiration:
When a rainbow is seen
On a pic of your screen

it's a moiré

reval
Posts: 87
Joined: Fri Sep 23, 2016 2:56 pm UTC

Re: 1814: "Color Pattern"

Postby reval » Fri Mar 24, 2017 3:06 pm UTC

Heimhenge wrote:Otherwise why wouldn't my cousin see it?

A lot of people don't believe what they're seeing. If what they see doesn't match up which what they think they ought to be seeing, the idea can get dismissed at a fairly preliminary level. I believe your cousin was being honest. Seeing something new and unexpected is not easy, and you have to train yourself to observe properly.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3566
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1814: "Color Pattern"

Postby Soupspoon » Fri Mar 24, 2017 3:40 pm UTC

Could just be not focussing on the right thing. If you're directed to look towards a meshed doorway that is (under normal circumstances intended to be visually unimpeding) you might look through it, the doorway itself thus outside your assumed - necessary focal depth.

User avatar
somitomi
Posts: 635
Joined: Fri Nov 06, 2015 11:21 pm UTC
Location: can be found in Hungary
Contact:

Re: 1814: "Color Pattern"

Postby somitomi » Fri Mar 24, 2017 4:45 pm UTC

There's a photo I took
of my computer screeeen,
bands of different colours
in it can be seeeeen.
Two grids misaligned
create a pattern made of liiiight
and the pattern keeps on changing
playing tricks on my eyesight.


Whoops, that's a diifferent song, but it keeps popping into my mind whenever patterns are mentioned.
—◯-◯

User avatar
orthogon
Posts: 2961
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1814: "Color Pattern"

Postby orthogon » Sat Mar 25, 2017 9:23 pm UTC

The "rainbow" thing is interesting. Presumably the camera's raster aligns with pixels of different individual primaries in the monitor as the Moiré pattern runs through. And/or vice versa. I saw some examples on the image search, but mainly alternating between two colours. Is it possible to get patterns that explore the full range of colours?
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3566
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1814: "Color Pattern"

Postby Soupspoon » Sat Mar 25, 2017 10:00 pm UTC

If there are three independent (albeit coperiodic, by various complicated factors) relationships of red-sensor on red-pixel, blue-sensor on blue-pixel and green-sensor on green-pixel and all sensors most tightly record "only FOO-ranged pixel brightness", by design, then you'll have three monochromatic moirés with either slightly or completely (and varyingly so) offset peaks and troughs, in locations where "true" brightnesses are already edgy.

The better the quality of the data (the less optical blurring or pixel->sensor vagueness) for a given resolution, you'll get a resulting RGB=(0%|100%,0%|100%,0%|100%) component producing obvious mini-rainbows even (especially?) from an uncoloured source. (Not entirely sure that linked image is produced the same way, but it does show the kind of "colour from where there was none", that isn't a prismatic effect.)

User avatar
mheney
Posts: 12
Joined: Wed Apr 29, 2009 8:15 pm UTC

Re: 1814: "Color Pattern"

Postby mheney » Sun Mar 26, 2017 3:28 pm UTC

To quote Dr. Zachary Smith - "Oh, the pain, the pain!"

User avatar
peterdavidcarter
Posts: 42
Joined: Wed Dec 14, 2016 6:36 pm UTC

Re: 1814: "Color Pattern"

Postby peterdavidcarter » Sun Mar 26, 2017 3:34 pm UTC

I ran this through a sentiment analysis program written in Python and it returned:

    Rainbow
    Another behind

and

    When the spacing is tight/and the difference is slight

Given this was not actually a sentiment analysis, but rather just a selection of seemingly unrelated keywords, I sent an additional database query.

The response JSON data was useless. It just contained various references to John/Sherlock explicit fanfic hidden very subtly in the code.

And an almost unfathomable reference to GLaDOS.
Last edited by peterdavidcarter on Sun Mar 26, 2017 3:47 pm UTC, edited 3 times in total.

User avatar
Jorpho
Posts: 6203
Joined: Wed Dec 12, 2007 5:31 am UTC
Location: Canada

Re: 1814: "Color Pattern"

Postby Jorpho » Sun Mar 26, 2017 3:39 pm UTC

racemaniac wrote:Anybody got any more verses for the it's a moiré?
The communicators from the original Star Trek used moire patterns assembled from a specific kit.
http://www.herocomm.com/Details/MoireStory.htm

Unfortunately "communicator" is a darn difficult word to work into the verse. But If you recall the original lyrics, the rhythm does change for the third verse, enabling more elaborate variations.

If you stick your hand
deep inside a crack
and you do not get it back
that's a moray...

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3566
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1814: "Color Pattern"

Postby Soupspoon » Sun Mar 26, 2017 4:35 pm UTC

Jorpho wrote:The communicators from the original Star Trek used moire patterns assembled from a specific kit.
http://www.herocomm.com/Details/MoireStory.htm
Some people are far too obsessed. I shall have to add the details of this person to my master catalogue of people who are far too obsessed over irrelevant details.

Unfortunately "communicator" is a darn difficult word to work into the verse.

When lines catch your eyes
As they call "Energize!",
..."?

(Yeah, that's a Transporter Room phrase. Tried to use "Beam me up", but given the common "Scotty" version is a mis-quote, perhaps I'd have annoyed a pedant even if I had found a rhyme that scanned.)

User avatar
PinkShinyRose
Posts: 832
Joined: Mon Nov 05, 2012 6:54 pm UTC
Location: the Netherlands

Re: 1814: "Color Pattern"

Postby PinkShinyRose » Sun Mar 26, 2017 8:53 pm UTC

Couldn't you just fix this by unevenly distributing the pixels on the chip.

EDIT: and shouldn't you get this when you rotate a photo less than a 90° interval in paint/gimp/photoshop?

User avatar
peterdavidcarter
Posts: 42
Joined: Wed Dec 14, 2016 6:36 pm UTC

Re: 1814: "Color Pattern"

Postby peterdavidcarter » Sun Mar 26, 2017 8:55 pm UTC

PinkShinyRose wrote:Couldn't you just fix this by unevenly distributing the pixels on the chip.


Are you talking about the source code, or pure binary, or something else...

User avatar
Flumble
Yes Man
Posts: 2051
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1814: "Color Pattern"

Postby Flumble » Sun Mar 26, 2017 9:26 pm UTC

peterdavidcarter wrote:
PinkShinyRose wrote:Couldn't you just fix this by unevenly distributing the pixels on the chip.


Are you talking about the source code, or pure binary, or something else...

PinkyShinyRose is talking about randomly sprinkling pixels on crisps, like they do with salt and spices, instead of rasterising them.


PinkShinyRose wrote:EDIT: and shouldn't you get this when you rotate a photo less than a 90° interval in paint/gimp/photoshop?

If you take 1 sample per pixel (at the centre or on a corner or whatever, as long as it's not random), yes. But the image editor by default takes a weighted average of the 'original pixels' (which, for rotation, usually means bilinearly interpolating the 4 nearest source pixels).

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 3566
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 1814: "Color Pattern"

Postby Soupspoon » Sun Mar 26, 2017 10:03 pm UTC

(Ninjaed by a reply more succinct, in part... Ah well.)

PinkShinyRose wrote:Couldn't you just fix this by unevenly distributing the pixels on the chip.

It'd take some designing to do that. An efficient packing like squares-in-a-strict-grid, with minimal boundaries between them, works best with something that you want to produce squares-in-a-strict-grid in both storage and output devices (hexagons/triangles, packed, would give an interesting alternate tendency to moire... once you got over the data handling and display differences). Attempting to jiggle this means juggling each photosensitive zone (randomly/pseudorandomly) within a subset of the larger grid (perhaps reducing the sensitivity by the same percentage, say 25%, as you decrease the collection area per tile.

"Analogue" photography works with an irregularly packed mess of photosensitive zones, edge-to-irregular-edge. Differently sized crystal-grains (echoing their significance nicely enough to not matter). How about a precisely photolithographed Penrose Tiling? Hmm, might still 'enjoy' a rotational symmetry (or sufficiently so) unless it's a particularly devious one. And then it needs matching to a counterpart display (without conversion, or the smearing of data in the intermediate that we're desperately trying to avoid in our attempt to retain the full original resolution without these awkward patternations), which is an interesting standard to maintain.

Without that, we're getting advice such as http://www.picturecorrect.com/tips/moir ... remove-it/ that boils down to trying not to invoke it in the first place. Or, if you can't sufficiently do so, anti-alias (i.e. blur) it, by just the right amount to squelch it without squelching 'real' details.


EDIT: and shouldn't you get this when you rotate a photo less than a 90° interval in paint/gimp/photoshop?
The interpolation/averaging used (e.g. Cubic) probably smears the hell out of source, most of the time.

User avatar
peterdavidcarter
Posts: 42
Joined: Wed Dec 14, 2016 6:36 pm UTC

Re: 1814: "Color Pattern"

Postby peterdavidcarter » Sun Mar 26, 2017 10:48 pm UTC

Flumble wrote:
peterdavidcarter wrote:
PinkShinyRose wrote:Couldn't you just fix this by unevenly distributing the pixels on the chip.


Flumble wrote:Are you talking about the source code, or pure binary, or something else...
PinkyShinyRose is talking about randomly sprinkling pixels on crisps, like they do with salt and spices, instead of rasterising them.


There in no rand(int) only salt and spice as the base needs. And requires.

User avatar
Jorpho
Posts: 6203
Joined: Wed Dec 12, 2007 5:31 am UTC
Location: Canada

Re: 1814: "Color Pattern"

Postby Jorpho » Thu May 17, 2018 11:29 pm UTC

I stumbled upon this and felt it vital to preserve it somewhere for future reference.
QWaJ7PM.jpg


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: No registered users and 43 guests