The Color Temperature Paradox
HTML-код
- Опубликовано: 17 окт 2024
- Go to brilliant.org/... to try everything Brilliant has to offer for free for a full 30 days, and also get 20% off your annual premium subscription!
If you take a piece of white paper into different lighting conditions, it will be an objectively different color in each situation, but our brains are clever enough to make us feel like it's still white - it's still the same piece of paper, after all. To match our experience, cameras have to do this, too, "balancing" the colors of an image so that a white object looks white under a given light, rather than some other color. And the typical unit to measure the color of a light is the Kelvin. Which is weird, because Kelvin is a unit for measuring temperature, not color. What temperature and the color in a photograph have to do with each other comes down to history and physics.
Support MinutePhysics on Patreon! / minutephysics
Link to Patreon Supporters: www.minutephysi...
MinutePhysics is on twitter - @minutephysics
And facebook - / minutephysics
Minute Physics provides an energetic and entertaining view of old and new problems in physics -- all in a minute!
Created by Henry Reich
I'm an astrophysicist and we deal with the idea that 'red = cold, blue = hot' all the time. I've (jokingly) suggested that we should swap labels on the department's taps so that the red tap is cold and the blue is hot.
Bah! What do you know? You call oxygen and carbon metals!
@@rtg_onefourtwoeightfiveseven That’d be hilariously mean for anyone visiting lmao
I love this
Haha, when i was a kid and found out that hotter things were blue, i asked my dad "why is the hot tap red then?"
Stop joking and DO IT!
During the Australian forest fires of 2020, one day everything went orange, like going outside was like being in a movie set in Mexico. But you couldn't take photos of it with your phone because the camera would auto colour correct.
Unless of course you went into manual mode with your phone and stopped the auto color balance from happening.
And if you couldn't do that with your phone, then you got the wrong phone.
I went to a place with a very nice yellow lit tram, ut my cam autocorrected it. Its annoying
Mexico 😭
Next time this happens just switch your cellphone camera to "Pro" mode or equiv, and change the WB to 4800-5000K (daylight). It will show sunsets in a nice deep orange/red and white surfaces not lit directly by the sun during a normal sunny day, as blue lit.
One of the most insane experiences during that time, flying into MEL, from the smoke we couldn’t see the ground until we were under tower height. Smoke, smoke, smoke, runway.
Another point for confusion for people who have experienced the many different lighting technologies we have had over the years, incandescent lights which produce a "warm" orange yellow light, also give off a lot of heat, while some of the alternatives to incandescent, which often gave off "colder" blue or white lights, often gave off much less heat. This adds to the image people have of orange/yellow being warm, and blue/white being cold in our minds.
Its't that just because "warmer light" (which comes from cooler things) comes from multichromatic things that emit more infra-red light that makes them _WARMER_ !
@@stevevernon1978 Pretty much, yeah. It's because orangeish light often comes from things that glow because they're hot (and these emit infrared). Bluish light could in theory come from something glowing because it's even hotter (and it would emit even more infrared), but in practice, it's really hard to heat something to 20,000 K. So bluish light sources typically use other mechanisms, like LEDs, or indirect sunlight (which uses the scattering of the atmosphere to increase the relative amount of blue even though the sun itself has a lower color temperature), and these don't follow the traditional formula of temperature vs. heat, so they don't feel hot.
Another point of confusion might be the color of normal fire, and hot iron, and the color we usually see in snow and ice.
Yeah, I have this aunt who always being near a white-yellow light bulb(LED) says "It's so hot in here, I'm gonna move to a colder place", but then is near a white-bluish light bulb(LED) and doesn't say anything, then I always tell her "it's not hot, they are LED and touch them with my hand and she is like " I don't know but..." and just continue doing the same everytime hahaha it's just fun for me now, years ago it used to bother me xD
There's also the fact that large bodies of water are blue, and cold to the touch
As a Home Depot employee that constantly has to explain color temperature of light bulbs, this was quite informative. Also I'm really thankful we have a "selectabulb" demo unit to show customers what different color temps look like, plus color-coded packaging so once they know what color they want, I can just point them to the right shelf and bay and they're good to go.
Be careful with the demo unit. Unless you're seeing it it isolation and take some time to adapt, a warm lamp for the evening will look way warmer in the bright neons of the store, since your eyes adjust for the "average" light.
Supermarket research discovered that a Silver silk bed sheet sold well under halide bulbs giving it a golden shimmer
But under standard Cool white leds in other stores,
it looked gray and didnt sell well.
Similar to the Viral dress
good stuff! never occurred to me that "warm light" and "cool light" are the wrong way around
The way I think of it is historically, warm light comes from things like fire. We observe warm light from warm sources. Cool light is light often reflected or filtered such as an overcast sky or light bouncing off icebergs. Cool light has it's lower reddish frequencies absorbed by cold things causing it to shift more blue.
It's just rare for humans to be seeing things so hot that they went beyond the orange / yellow spectrum, so these colors get associated with hot. Meanwhile ice and water tends towards blue.
Big fan of your channel btw.
It is not the wrong way around. It is just more useful for the description to match our experience rather than what it objectively is. Our experience tells us that yellow gives us a warmer, cozy environment that's not harsh, while blue reminds us of colder seasons, where there is a notable natural absence of colorful things contrasting against the clear blue sky. Also, the blue tends to keep us more alert much like a cold wind does. It's mostly only people dealing with kilns, forges, or similar devices that actually care if a yellow glow is actually colder than a blue one, and that's not a regular experience for most people.
At the very least, it is helpful that light bulbs at least have an objective measure of their color rather than simple qualitative names, because you can at least be confident at what the camera is actually trying to adjust for when you see the color temperature setting. It gives you a standard way to describe lighting that is not quite amber, warm white, soft white, cool white, and daylight.
I thought of this ages ago and it's always annoyed me
Cool person
he remembered his youtube password
Part 2
ruclips.net/video/aQeaztXRhIM/видео.html
should have used lastpass - now 10% off with code minuteselloutdatabreach
😓😂
Y'all need to get a new line for when somebody hasn't uploaded it in a while. Beyond cliche. 🙄
So he had a 3 month hiatus. Probably well deserved. 😂
This concept really screwed with me badly when I was young artist. Learning blue=hot before color temperature in painting confused me for the longest time even though it really shouldn’t have.
I know, that is just silly that blue is called "cold" and red "hot".
@@wawrzynieckorzen78yeah, but it makes sense when you think that blue is meant to refer to ice and red is meant to refer to fire. The colours are not based on the temperature of light.
???
@@Anonymous-df8it !!!
@@RanOutOfSpac ???
3:15 no one calls it color temperature, we all call it white balance, which accurately reflects what you explained here lol
Yeah, white balance is a very common and well known term. Idk how he missed that
Ya know, you could’ve waited a few more seconds. At 3:19 there is a note on the top addressing white balance.
Speak for yourself.
It doesn't need to be called color temperature compensation... "white balance" is a shorter and more intuitive term which is already in widespread use and covers both CCT and duv and other types of adjustments.
Yes, second half of this video is just people being wrong. Hardly a paradox.
I mean he addressed that in an on-screen comment but said it was unintuitive I think
Henry's jab at "white balance" not eluding to compensation is non-sensical. White balance isn't compensating for anything, it's actively warping the output of its sensors to conform to human bias.
White balance is also the thing that ruins all my photos on a new camera until I find the setting that turns it off
@@Bruh-zx2mc Alluding.
One thing you missed was how the camera “knows” what “white” is in the first place which is done by looking for Middle Grey or 18% grey!
But 50% is the middle, not 18% 🎭
it might be more 'chose not to include' than 'missed'
@@billcox6791 Our eyes do not perceive brightness linearly. We are much more sensitive to shifts in low brightness than high brightness, and it takes an exponentially higher amount of light to keep increasing things by the same increment of perceived brightness. So, to our eyes, 18% middle gray seems like the middle point between totally black and totally white.
@@KaiBurley Yeah, it just amused me as one more layer on top of how we tell the camera it’s hot so it makes the image colder so it looks warmer to our eyes when compared to middle grey which is 18%.
@@billcox6791 Although what you say is in some sense false, because, as others have said, logarithms, 50% vs 18% doesn't much matter to the camera's white balance; to the camera white balance (the _ratios_ of R, G and B) is basically orthogonal to exposure (the average of R, G and B).
I mean, that's a lie, it's all a lie, the eye is sensitive to spatial gradients more than it is to absolute values; it switches to a different colour model entirely in low light; it's not equally sensitive to R, G and B; R, G and B components correspond to weighted sums over ranges of wavelengths and not wavelengths at all; and the nerves behind the eyeball are so heavily multiplexed that motion, colour and spatial frequency impinge on each others' codespace. And that's _before_ considering the eye's white balance processing which relies not merely on what you can see, but also your experience and expectations for what you see, what else you've seen recently, and (this is true) the colour of your own nose, which you can always see and never notice. And then there's the fact that (local) overexposure and underexposure are going to be processed very differently by a camera and the eye; it's not merely logarithmic vs linear, because the camera is roughly logarithmic globally and linear locally, while the eye has these … response curves. Multi-axis response hypersurfaces, really.
But whether your colour reference is pale grey or dark grey really doesn't matter to the white balance. It matters to the exposure, a different subsystem in the camera, since it's possible to handle optomechanically (in large part… I mean there's software gain and mult fusion), unlike white balance, which is handled in firmware.
Unless filters. And chromatic aberration. And lens shading. Actually even barrel distortion technically has an impact on the discussion. Let's not talk about this any more, actually. It's really very very complicated.
Ice and deep water can appear blue because of the way they absorb and scatter light. Fire looks red. Maybe some evolutionary intuition there? Cool video.
As someone who went from a science kid to a filmmaker, this is by far one of the most frustrating things I had to learn. “Cold” colors, expressed in HIGHER temperature (warmer) numbers is achieved by LOWERING the camera temperature setting and forcing it to expect warmer light, which is achieved with cooler temperatures.
It becomes intuitive once you’ve dealt with it long enough, but this single-handedly convinced me to stay away from lighting and just let someone else handle it!
Part 2
ruclips.net/video/aQeaztXRhIM/видео.html
I think it helps if you think of it as the absolute energy of the wavelengths of light. Light from heat just shows an increase in energy, shifting from infrared to red to blue and up
Temperature scales are a little strange, a regular candle fire is already 2000 K, the surface of the Sun is "just" 5800 K
So there are lamps hotter than the surface of the Sun, and if I heat something enough it tends to glow blue?
@@Khether0001 The sun has the same issue as neon lamps and LED's. Their color is influenced more by nuclear emission spectra than just their temperature.
>Oaisus : I don't think so. The nuclear reactions in a star don't occur at the star's surface. They occur at the core, where the pressure is much higher. It can take thousands of years for the energy emitted by the fusion of two nuclei deep within the star to reach the surface, and during that long, winding journey from core to surface the photons are absorbed & re-emitted countless times. That changes their wavelength as some of their energy is converted to kinetic energy (higher temperature) of the atoms & charged particles that absorb/re-emit them. The sunlight we see is emitted by the hot gases at the sun's surface.
I MISSED THIS CHANNEL SOOOOO MUCH.
Part 2
ruclips.net/video/aQeaztXRhIM/видео.html
He is busy writing the scripts for the xkcd channel
I was hoping this video would be about why we colloquially think red is warm and blue is cold.
I wished this was int the video too, but I guess it's just because cold objects are often blue or white and white reflects the blue sky
I always assumed that was because of the way temperature affected skin (Caucasian).
ice and water are blue and fire is red, its probably something simple as that
Winter is very blue, sommer is red. Literally regarding the light outside
@@moartems5076 I like that, in combination with the fire/water idea. The sky should be more blue at shallower sun angles. Also true for noon vs dusk, I think.
i thought the paradox was that ice looks blue so we call that cold light and fire is obviously red~orange, so we call that warm
That's what I always thought, too
I never heard what the paradox even was. I assumed it was the thumbnail as well with the colors. I guess the solution to this paradox is that the blue ice is actually channeling the light from the very hot sun (and blue sky), while red in fire is because the fire isn't so extremely hot.
3:22 we call it whitebalance. Because you're balancin the white according to the surrounding light. So if you're stting the camera to 2500k you're balancing the white to a 2500k light. Makes perfect sens if you phrase it like that imo.
Just wanted to write that, but there is no need. Your comment is spot on. I also use the term "white balance" in its equivalent translation. Not "color temperature".
Yea he makes a white lie by omitting the most obvious information. No one says "adjust the colour temperature" everyone says white balance, because you usually also need to adjust green and magenta, and white balance encompasses both.
That's literally what the asterisk note is saying
@@Heulerado I kid you not, i did not see that
@@Heulerado Nope.
Thank you for not simply keeping on talking in an attempt to bamboozle your viewers into watching an ad. That's how respectful content creators do it
as a photographer who has to deal with color temperature all the time: at first it was disconcerting.. and after more than 20 years.. it still is!
The fact that we still call a warm tone, a tone that is much lower than the blue tone.. *brain pops out*
But this video helped at least
Just think of it this way, the Sun is yellow and warm, the Moon is White and cool. Which are whole other contributing reasons why we also associate the colours differently.
@@Targe0the moon is actually the same color temperatur as the sun!
The sun is actually only "warm" during golden hour
@@FilmmakerIQ wtf is a golden hour?
Pretty simple really. Blue leans towards light with higher energy and red towards lower. When we see a blue hot object really it's emitting mostly UV, but since 'blue' is the cut off point for our eyes it just stops visibly changing after that. Therefore higher energy -> "blue". Same goes for red, since it's mostly emitting IR we see it as 'red' because red is the cut off point for lower energy light. Though we actually can sense (some) parts of the IR spectrum, just not with our eyes but our skin. We sense it as heat. Hold a warm cup of tea against your face and you'll feel it. Though in this case the temperature of the teacup is so far in the IR that it doesn't emit any visible light at all, which is why we don't see it as glowing red.
I wanted to make this video so much. It's the sun and Rayleigh scattering that leaves red and green light from the sun which is seen as warm and the shadows look blue despite being a higher wavelength and being hotter.
idk how common this is, but my eyes see white pieces of paper as slightly different colors when viewed separately, like the orange-ish and blue-ish hues at the beginning of this video. With both of my eyes open, the hues cancel each other out to be true white (as far as my perception goes anyway).
i feel like the word "paradox" is thrown around a bit liberally here.
In the mid 2000’s Daylight white bulbs were used for outdoor lighting quite a bit. They still are today but they mess up my internal clock if I wake up at the middle of the night and see it reflected in my bathroom window
More on color please!
One thing to mention is that while balance is not just to align the image to what we see. Otherwise, our eyes should be able to adjust to the image just as we adjust to the light -- at least viewing the screen in a dark room. White balance is also there because the whitest part of the image is fixed color temperature. For example, the overexposed part of the screen (i.e all RGB channels are 100%) should have a fixed color temp (say 6300k). This will differ from the part that is not over exposed. This is why white balance is less of an issue in HDR photos/videos.
@@davidli9887 interesting
I wouldnt call black body radiation a paradox, but we just accept the fact that entities are capable of emitting black body radiation. Some things in physics make sense but arent reducable to any underlying laws, they are just as they are. It is more like strong emergence not a mystery. (strong emergence isnt a mystery either nor magic, it is ontological irreducability, the thing that allows you to be a conscious being that emerged by non conscious matter)
That was one CONFUSING way to express that humans are used to reds and oranges feeling "hot" (mostly because of sunshine), and violets and blues feeling "cold" because of darkness and night. It's as much a visual anomaly as it is a sense of touch or whatever.
That’s not at all what the video was about.
They only briefly mention color theory warm and cool colors.
The video was largely about black body radiation and color constancy.
To me, blues (and especially blues bordering on white) are cold because that's the color of ice and snow, not nighttime. And let's not forget the color coding of almost every single water tap everywhere.
@@MasterHigure Precisely. I mean, sure, light and fire are unique regarding the "color swapped" feel, but it's like... come on. The normal fires MOST humans see are red and orange with yellow wisps. Water appears blue due to reflection and refraction, and we associate it with "cold". As to the Kelvin + cameras thing, sure, there's a lot of miscommunication and misconceptions. But ultimately, the visible spectrum light is pretty tiny, so, whatever; it's complicated. What might one say about UV vs. IR? Is one "hotter" and one "colder"? Sort of, but either could be used to screw you up--mostly UV, obviously.
@@Novastar.SaberCombat What do you mean, sort of? IR is colder than UV, obviously. There isn't any ambiguity about it, it's just true.
Can you make a general video on blackbody radiation? That's the key behind these definitions and I think it'd lend an important piece of context to this explanation for anyone who wants to understand it fully.
super useful, thank you so much! that four square layout showing the lights creating colour and the camera compensating / subtracting colour was a great image
I very much appreciate the bit about kelvin being the full term. I've definitely commented on videos which have said it the old way (the way it was before 1967 I believe). I'm not sure if any of my comments were on older videos of this channel, but definitely on some otherwise accurate videos of some channels.
the way cameras store color in images is also interesting. in order to sample more dark colors than bright colors to imitate human vision (in the early days of digital imaging when saving disk space was essential), the cameras store the square roots of the brightness values. this leads to unintended effects when you try to modify the image file without accounting for this fact.
when i first picked up a camera, i wish someone had said "color temperature is opposite of what you think"
There.
Then you realize it's just metal being heated like by a blacksmith. low= orange Hot= yellow very hot = white
The term you were trying to think of when you wanted a better term than 'white balance', is 'chromatic adaptation'.
To perform chromatic adaptation, you first convert an image into the XYZ colorspace. From there you perform what's known as a 'Von Kries transformation', where you convert to LMS, scale the three color channels so that a given white point becomes a different white point, and then convert back to XYZ from LMS.
Because the conversion to LMS, the scaling of the color channels, and the conversion back to XYZ can all be expressed in a single 3-by-3 transformation matrix, it's considered its own single step in the calculation - assuming you already have it computed, of course.
Additionally, there are many different XYZ to LMS conversion matrices, and the most commonly used ones (the 'Bradford' and 'CIECAM02' ones) are actually somewhat 'wrong'. They are calculated from a 'spectrally sharpened' version of the LMS color space, because it seems that doing so leads to more perceptually accurate results when performing chromatic adaptation.
When the LMS color space is used for other purposes, it's recommended to use the Hunt-Pointer-Estevez transformation matrix (sometimes called the 'Von Kries' transformation matrix, because it was the matrix used in conjunction with Von Kries transformations).
@0:43 makes so many things make more sense. Thanks for the video!
Wait, that's not a paradox... that's just mildly confusing if you don't understand.
Maybe that's the paradox
Paradox doesn't always mean something that is contradictory; it just refers to something that goes against your expectation.
Because you'd expect the high temperature to be blue, but the screen shows you yellow, it's against your expectation.
Para = beyond, distinct from
Doxa = expectation, belief, judgement
(See also: orthodox, meaning 'true belief' or 'true to expectation')
@sp00ky_guy oh wow. I just looked it up, and you are completely right.
I had no idea!
@@TopLieutenant to add to that, this would be a veridical paradox. A paradox that seems counterintuitive, but is actually true.
In contrast to the typical version people mean: antimony paradox. A paradox that is truly logically contradictory.
And for completions sake: a falsidical paradox is one that seems wrong and actually is, because it makes a logical error somewhere in its reasoning.
There are other methods of categorizing paradoxes, but this one is a common one.
Wow. Such a wholesome thread on the internet. Two people accepting and understanding enough to have a decent dialogue instead of just trying to prove each other wrong. ❤
Yay, more fourminutephysics!
appreciate that henry had to find out the colour-opposite of the colour pens he used ro represent the temperatures because it's filmed in negative colour...
Nice video! Color temperature can get real confusing sometimes. As someone who does both stage lighting the occasional short film, there's a reason some people prefer to say amber and blue. Definitely going to show this when someone asks me about CT though...
I always tell my students that they shouldn't say "degrees Kelvin" but people know what you mean if you say "degrees Kelvin." Points to "A Brief History of Everything, feat. Neil deGrasse Tyson" video on this channel. But thanks for another good explanation video.
I love the video. I know all of this having an interest in photography, and having worked calibrating monitors for professional photographers, but it still hurt my head to watch. Bravo!
the colors in the drawn segments were very pleasing to me
I like "Temperature Offset" more. It succinctly captures the purpose of this feature
i like to call the green-purple color axis the "flavor"--green being tart, and purple being sweet.
(it's sometimes labeled "tint", but i find that to be rather unspecific, and doesn't specify any fun names for the directions. hence, my way)
THANK YOU SO MUCH for inverting the surface of the sun that you write on. ♥♥♥
Having to do it myself with hotkeys was tedious and error prone.
1:08 so incandescent lightbulb filaments are about half as hot as the surface of the sun?😮
Nahhh black body radiation isn’t directly related to the temperature of the light source (tl;dr since it’s not a black body), and “colour temperatures” are about a curve of a set of colours, whereas a specific colour (I mean hue) is better to specify by nanometres or terahertz. Or zettajoules, that’s another option.
I never deeply understood white balance until I came across this video. thanks mate this was an explanation of a lifetime!
You didn't mention the Black Body Locus which is the path that the color of an incandescent theoretical black body would take in a chromaticity diagram as the blackbody temperature changes. That is where "color temperature" came from.
Thank you for the Kelvin explanation. I had a feeling kelvin was its own scale and it not necessary to say degrees, even though i do hear it a lot.
One source of frustration for me is that when light bulbs are labeled by 'color temperature' in Kelvin the green-magenta dimension is omitted, despite its contribution to the 'feel' of the light. I remember appreciating the way the older daylight CFLs showed a slight green bias suggesting sunlight filtered through a leafy forest canopy. The newer LEDs with the same Kelvin labeling, while superior in terms of color reproduction, seem biased toward magenta giving them an interstellar starlight quality.
I swear it's been ages since I saw you upload. happy to see you again
So this is why my phone likes to make everything blue shortly after sundown.
I think the reason we call blue light "cold light" and red light "warm light" is because blue fire wasn't known back then (or was not common) in ancient times.
Usually, blue things relate to cold things. Ice is blue, water is blue, snow is white but kinda blue, the sky is blue, the clouds are kinda blue.
Red light is common in dangerous things. Fire, lava, and the sky turns red when there's a wildfire, molten metal. The sun's also turning from yellow to orange to red in sunrise and sunset.
I do photography and literally had this realization with color temperature and blackbody radiation. Good to see an explanation
in its brevity this video is so well done. Thanks!
Wow! It's this time of the year. Minute physics uploaded.
Holy crap, dude. Thank you.
hey im a big fan and youve been one of the first channels ive ever watched. I have a thermodynamics exam in two days and this video made me realize I dont know what a kelvin is and I looked it up and it being related to the caesium standard and the boltzman constant is not understood by me yet. Thing is I was thinking why isnt Kelvin defined as the average speed of atoms since the difference between cold and warm is how fast particles move?
great video anyways! as always!
Looking back at chem lab classes this makes sense. The ‚flame burner‘ (Bunsenbrenner in German) was orange/yellow but the warmer and hotter flame (over 1000°C) was always blue(name-rauschende Flamme). The hottest area is over the small flame inside the whole blue flame.
FYI, the english word for Bunsenbrenner is Bunsen burner.
Interesting that the sun's surface temperature is only double that of a simple light bulb filament.
Seems surprisingly cold, but considering how enormous the size difference is, fair enough, that's a lot of energy.
nice brief tutorial, I tend to adjust white balance (or color compensation) of my JVC 750 video camera to where it looks good. But then I haven't dove into studying color charts, gamma, etc. I remember some years ago a newspaper reporter had a stack of cards of various shades so he can white balance his DSLR camera to whatever lighting condition.
I use ProCam (if that's what that is) too, it's wonderfully feature-full and somehow wonderfully minimal at the same time.
3:41 yeeen again ❤
Nicely done (as usual)! If you haven't already, maybe you could explain how a black object, which should absorb all visible wavelengths, can be shiny. It seems counterintuitive, like color temp.
This Channel is our family favorite.
For some reason I expected this video to be about the ultraviolet catastrophe! Black-body radiation is weird.
I worked in the lighting industry for nearly 20 years. White balance is even more complicated. You could easily do a series of hour long videos on light, lighting and the eye
There is another quirk with today's modern light sources: When light is created by thermal radiation the emitted spectrum is continuous, for example an incandescent light-bulb not only emits visible radiation, but also infrared. LED light sources especially are very different, to create a "white" light, red, green and blue LEDs with individual very narrow spectrums are combined to create something that looks "white" to our eyes but might be perceived by a camera in a very different way. It would be nice if you could make a video about that aspect.
Was just wondering about this concept the other day! Great vid
1. take photo of color swatch
2. match final colors to reference
3. awe at the white balance color temperature compensation correction de-tinted in K(without degrees) you achieved
We can dive even deeper into the quality of the light source too, while the light might appear X Kelvin, it’s missing significant portions of the spectrum behind that and looks weird. Think LED headlights, you’re missing so much of the spectrum, you can miss objects of certain colours or they camouflage.
when the uploads come in slow you know the youtube money is coming in fast
Had a hell of a time discussing with my chemistry classmates that Blue = higher number in K = Hotter
but also when discussing about lighting and multimedia that Higher K = Cooler color = Daylight
I like the fact that the colors that were shown after inverting is the actual color being talked about
That orthogonality of green/magenta is not something i really thought about before, and is never mentioned in lighting. But I've definitely noticed some 4000k LED spotlights have a magenta tint compared to natural light of a similar colour temperature
1:38 wait, how can sunlight be 10,000k when the sun's surface is only 6,000k?
The color is changed by interacting with the atmosphere.
@@jomialsipi I thought he was talking about the actual temperature (and how that affected the colour), but that makes a lot more sense. In hindsight those numbers did seem rather high if they were the real temperatures😅
I'll just put a comment floating here that we think of light blue as "cool/cold" because that's the color that sunlight ends up looking after it's passed through a glacier of ice thick enough to absorb (and/or scatter) the longer, weaker wavelengths of redder light out of the original white light, same for the deep-ish ocean looking blue before even the high-intensity bluer light is absorbed and it all just goes black. The More You Know.
I think it's also that in shadow, or at dusk or dawn, things that are out of direct sunlight are mostly illuminated by scattered light from the sky, which is blue. And the first fires that humans were able to make to warm themselves burned at fairly low temperatures and produced orange light.
thanks for putting the *notes in the upper part of the frame 👍
1:15 the subtle prod towards Portal there.
Honestly wish he would film these inverted because I like watching these at night and the white background kills my eyes
I thought this was going to be about how we think of red things as hot and cool things as blue because our everyday perception of temperature includes things like fire and water, but actually as things grow hotter they go from red to blue.
Still an awesome video, welcome back!
The whole camera compensating reminds me of additive versus subtractive coloring.
The warm/cold thing comes from experience with things that are red-hot. A snow covered ground and ice may appear blueish. We don't usually experience things that are blue-hot.
Yes, I was always confused with how a camera temperature setting, or the post-processing's, is inverse to a lightbulb temperature.
great, i could follow but still am confused anyway. Especially the part with color compensation. lol. great video
You absolutely say "Degrees Kelvin"
When talking about Portal.
Wow! ...what a brilliant Video!
Thanks! This has been messing my head. Explains a lot of cold electricity.
Thus, thanks to this double reversal, setting the white balance of a camera to lower K 'cools' the image and setting it to higher K 'warms' it, which is intuitive.
Also, your eyes (or rather your brain) do color correction naturally. Even outside the usual color temperature range. If you've ever worn one of those red-cyan 3D glasses, you might have noticed that after a while of wearing, the eye that saw through the red filter now sees everything in cyan, and vice versa.
I knew this, but I still learned something new!
_shaking my fist at my camera_ *"...Damn you, history and physics!!"*
god, it used to bother me so much as a child (and still somewhat does) when people talked about "warm and cold colors". it's a bloody color, mate, the room is still gonna be room temperature whether you paint the walls blue or yellow.
Is not a paradox, the way we experience the world is just an exception to the overall rules of the universe, as we can’t experience about 100% of the temperature range
‘paradox’ has multiple meanings. in this context, it is used to mean counter intuitive
Octaves of light. Double the frequency; get the same colour.
honesly, i don't consider orange light warm or cozy at all. the color of the light doesn't awakens any feelings in me.
that's why i mostly shoot black and white or very desaturated colors. however i still use saturated colors sometimes for utilizing "color contrast".
I think we measure colour in kelvins for the same reason we measure _temperature_ in kelvins*, we've got the convention established and it's quite costly and annoying for the generation that would have to scrap what they've already learned and replace it with something more elegant.
*i.e. units of 13.80649/(0.919263177 × 6.62607015) times the caesium-133 hyperfine transition energy, a quantity that has absolutely no significance in basic physics beyond being about 0.01 times the change in molecular energy (kT) produced by going from the melting to the boiling point of water under the approximate average pressure at Earth sea level
It's an even bigger quirk, that many people still calibrate their monitors to 5000K because that used to be the default color temperature in print media and surely all modern media will look the best this way. However, ITU Rec. 601 defines the whitepoint to be 6500K, so does Rec. 709, and that's what all modern media use. Also a reference monitors should have a gamma of 2.4, despite a display monitor to be defined to have a gamma of 2.2. Also many modern displays have a whitepoint that is up to 9300K. When you create content, you should review that content on a monitor with D6500 and gamma 2.4 to match the desired aesthetic and that means if you calibrate your monitor that way, you get a display result closes to the one that the creator of that content probably intended when viewing the content. Feel free to go with other settings but anything below 6500K is probably way too "warm".
Do cameras usually allow white balance also on the axis green-magenta? And how is it done if it can't be expressed by temperature?
The confusion comes from old conventions meeting new information and knowledge. Kinda like the conventional current flow and the real current flow in electric circuits. Before we knew the charge of the electron we thought that electrical current flowed from the highest potential (positive terminal) to the lowest (negative terminal), like water down a mountain, and this is still a source of confusion today. In this case, we defined color temperature due to our associations with reddish color for warmth and heat, and blueish colors with cold, and only centuries after did we find out about wavelengths of light, their energy, black body radiation and temperature dependent light emission spectra.
1:17 Y'know, it would be nice of you to explain WHY we don't say degrees for Kelvin.
The reason BTW, is because Kelvin is an absolute measurement like length, mass, etc. 0 Kelvin means literally an object has no warmth whatsoever, whereas objects at 0 degrees Fahrenheit or Celsius still have warmth and you can have objects be below 0 degrees.
Anyways, the whole point about cameras compensating for lighting doesn't really feel like a paradox. If anything the biggest paradox mentioned in this video to me is why red/orange light is referred to as 'warm' while white and blue are 'cool'.