|
I will admit that if I explicitly try to look for what part of the rainbow looks "most whiteish" on that chart I'd settle on that area too, but I think that's more a weird artifact of my monitor / optical illusion than anything else. I mean it doesn't look white by itself but it does look the most white of all the other colors, if that makes any sense...
|
# ? Nov 6, 2021 05:14 |
|
|
# ? May 22, 2024 15:12 |
|
Cojawfee posted:Wow Incredible. It seems very possible that actionjackson has a mild case of undiagnosed tritanopia.
|
# ? Nov 6, 2021 05:17 |
|
Holy poo poo. Powerful internet diagnosis if so.
|
# ? Nov 6, 2021 05:51 |
|
Amazing if true. One of my favourite videos on colour science might also be relevant to the broader discussion on colour science: https://www.youtube.com/watch?v=uYbdx4I7STg A key difference between the RGB colour model for an emissive display and a white light source is that just using three LEDs at red green and blue (no matter how precise the wavelengths are) makes a horrible white lamp. I tried making white light using those neopixel RGB LED strips a few years ago and found that light made me effectively red-orange-yellow colourblind, since the reflective properties of whatever the subject is is sampled at three sharp wavelength lines, rather than getting an average over a wider range. There is also the issue of fluorescent materials which might only be activated by certain wavelength ranges that aren't there for trichromatic light, but would be there in the equivalent blackbody spectrum. This is separate from colour temperature, and is often measured at a percentage called CRI, where >90% is getting into acceptable territory. (CRI is not uncontroversial, but it is a measure that is often actually listed for LED lights, unlike the emissions spectrum). To be clear AFAIK CRI has no real relation to colour temperature, it should represent how good the light is at being whatever colour temperature it is designed to be.
|
# ? Nov 6, 2021 08:52 |
|
lmao holy gently caress
|
# ? Nov 6, 2021 09:24 |
|
https://www.youtube.com/watch?v=AO8WEDjn3pc
|
# ? Nov 6, 2021 09:26 |
|
petit choux posted:All my experience in this is just trying to make my alarm clock less bright. YMMV. BTW, have you looked at the documentation for your fridge to see if there are any ways of adjusting that? How about getting some of that window tinting film teenagers and ricers like to put on their cars and putting that over your display window?
|
# ? Nov 6, 2021 09:41 |
|
Sagebrush posted:Incredible. It seems very possible that actionjackson has a mild case of undiagnosed tritanopia. Hell of a way to find out if so. "What do you mean there's no such thing as monochromatic white light?!"
|
# ? Nov 6, 2021 10:15 |
|
longview posted:There is also the issue of fluorescent materials which might only be activated by certain wavelength ranges that aren't there for trichromatic light, but would be there in the equivalent blackbody spectrum. I have a chunk of Nd:YAG laser material sitting on a shelf, it was always a light purple color until I installed some LED lights and then it turned clear like glass. You can pick it up and walk into the next room where I have a different type of bulb and it turns purple again. Apparently the Nd is absorbing some wavelength and re-emitting into the visible range and that excitation wavelength isn't emitted by the LED bulbs.
|
# ? Nov 6, 2021 14:55 |
|
BattleMaster posted:Hell of a way to find out if so. "What do you mean there's no such thing as monochromatic white light?!" sorry to be a buzzkill, but it looks completely yellow to me now. I use f.lux, which now is at 5000K because it's morning. You'll see my original post was at 8:20 pm last night, when the color temperature is set to be much lower. So the whitish appearance is related to these filters. tonight I'll post how it looks both now and then in the evening so you can see the difference.
|
# ? Nov 6, 2021 15:18 |
|
Dang, I was that close to getting my online optometry license.
|
# ? Nov 6, 2021 15:30 |
|
actionjackson posted:sorry to be a buzzkill, but it looks completely yellow to me now. I use f.lux, which now is at 5000K because it's morning. You'll see my original post was at 8:20 pm last night, when the color temperature is set to be much lower. So the whitish appearance is related to these filters. tonight I'll post how it looks both now and then in the evening so you can see the difference. *pumps fist in the air and cheers at this post*
|
# ? Nov 6, 2021 15:33 |
|
https://i.imgur.com/oyppvom.mp4
|
# ? Nov 6, 2021 16:30 |
|
only in my dreams would my fridge be piss colored
|
# ? Nov 6, 2021 16:53 |
|
It would probably be hard to get he piss mist reliably as well.
|
# ? Nov 6, 2021 16:59 |
|
as promised, here's a screenshot taken at 9:10 PM, where f.lux is at 1900K
|
# ? Nov 7, 2021 03:10 |
|
I think it bypasses f.lux. You could take a picture, but we believe you.
|
# ? Nov 7, 2021 03:13 |
You mean that white patch in the middle there?
|
|
# ? Nov 7, 2021 03:48 |
|
ante posted:I think it bypasses f.lux. You could take a picture, but we believe you. ah that's a good point yes the area from about 550-575 appears whitish when viewed with a filter at such a low color temperature
|
# ? Nov 7, 2021 03:50 |
|
actionjackson posted:ah that's a good point actually what is happening is f.lux has changed the white background of the graphic to yellow, and the specific setting you're using has white objects transformed so they match that part of the spectrum. The spectrum is still objectively yellow, but you're reading it as a "white" stripe in the middle because you know the background is meant to be white and your brain is compensating.
|
# ? Nov 7, 2021 04:45 |
|
Sagebrush posted:actually what is happening is f.lux has changed the white background of the graphic to yellow, and the specific setting you're using has white objects transformed so they match that part of the spectrum. The spectrum is still objectively yellow, but you're reading it as a "white" stripe in the middle because you know the background is meant to be white and your brain is compensating. I think there's at least some nugget of truth in here Still really* sad to hear we didn't internet diagnose actionjackson with colorblindness *not really
|
# ? Nov 7, 2021 07:35 |
|
ive become an actionjackson truther, that mofo isn't just colorblind they are using a screenreader.
|
# ? Nov 7, 2021 12:24 |
|
I still don't understand why actionjackson thought the spectrum of visible light would have white on it.
|
# ? Nov 7, 2021 17:05 |
|
Cojawfee posted:I still don't understand why actionjackson thought the spectrum of visible light would have white on it. Stuck in color temperature brain space? That's a spectrum with white in the middle. Color temperature can make a lot of this stuff more confusing by being just barely different enough.
|
# ? Nov 7, 2021 17:24 |
|
Fun fact, you can make white light with only two wavelengths, aka dichromatic white light sources! https://ocw.snu.ac.kr/sites/default/files/NOTE/795.pdf Basically just draw a line through two points on the 1931 chart to intersect whatever kind of white you like and there's your primaries: So somewhere around blue/cyan and green/orange can be mixed to make the worlds shittiest white light. This can be useful if you're making a white/blue/yellow palette emissive display I guess.
|
# ? Nov 7, 2021 20:48 |
|
longview posted:Fun fact, you can make white light with only two wavelengths, aka dichromatic white light sources! Ikea 'white' LEDs are like this.
|
# ? Nov 7, 2021 21:27 |
|
Pretty much all white LEDs use a blue or violet emitter that pumps a yellow-orange phosphor coating, and the combination produces a reasonably good white light. You can see the yellow phosphor when you look directly into a white LED (when it is turned off, obviously). Better color rendering is achieved with better phosphors that produce a wider spectrum.
|
# ? Nov 7, 2021 21:36 |
|
Most tunable white smartbulbs will have a cool white and a warm white LED. I mentioned earlier that white is complicated. This line on the 1931 XY colormetry graph is called the black body* curve: Our eyes / brain interpret the entire thing as "white". Towards the left, you get the cool white (usually 5000-7000k is a good value) and the right is warm white (usually about 3000k). Phosphor coating is the preferred way to get a white LED, because the other way is a mixture of individual colours, and that is awful. The spectral distribution I posted earlier is an example of that, it is a high quality light, high CRI: Here's an example of a colour mixture: It has a low CRI. It'll also look white if you're just staring at it, but that's not really how you experience light. Every object has a similar graph describing how they reflect light. To borrow early graphs, here is what an amber object would look like: That means the object acts kinda like a filter. It only bounces back the amber portion of whatever source light is illuminating it. However, imagine shining that second white light on this object. It only bounces back amber light, but hey, there is nothing on the spectrum in that range. The object will appear black. The first illuminant we looked at? Plenty of light in that spectrum. The option will render accurately, it'll still be amber. The measure of how well a light does this across a series of (made up, imaginary) samples is the Colour Rendering Index, or CRI. * everything I've talked about in the last few pages have a lot more depth to them, I know about the D65 illuminant, don't @ me, gently caress colour theory is an insanely complicated topic
|
# ? Nov 7, 2021 21:57 |
|
Crankit posted:ive become an actionjackson truther, that mofo isn't just colorblind they are using a screenreader.
|
# ? Nov 8, 2021 03:48 |
|
why is the text purple
|
# ? Nov 8, 2021 04:14 |
|
Bit of an obscure one here: How do you reconcile Scipy.signal's IIR design with CMSIS-DISP's API? Scipy.signal outputs in one of 3 forms: - Numerator/denominator - Pole Zero - Second-Order-Sections. CMSIS requires an array of length a multiple of five. Each 5 values are coefficients b0, b1, b2, a1, and a2 for a filter state: "*Coefficients b0, b1 and b2 multiply the input signal x[n] and are referred to as the feedforward coefficients. Coefficients a1 and a2 multiply the output signal y[n] and are referred to as the feedback coefficients. Pay careful attention to the sign of the feedback coefficients. Some design tools use the difference equation*" Scipy's formats seem incompatible: Numerator/Denominator uses "b" and "a" terminology, but returns 2 arrays: A numerator array of lengh 6, and denominator array of len 6. SOS format also returns arrays of length 6. This is in contrast to FIR, where there's a 1-to-1 mapping. Ie both use an array of coefficients corresponding to a convolution kernel. IIR seems more diverse by comparison. scipy.signal.iirdesign CMSIS-DSP Biquad Cascade
|
# ? Nov 8, 2021 05:55 |
|
Generally, you scale the world so that a0 (0th feedback coefficient) is 1
|
# ? Nov 8, 2021 06:24 |
|
Thank you! That makes sense.
|
# ? Nov 8, 2021 06:26 |
|
Ambrose Burnside posted:Anybody know of any good hands-on pneumatic logic resources? i havent worked with this stuff in years and it isn't coming back to me as quick as i'd like. neither basic digital logic nor PLC-type ladder logic are that tricky for me but i forgot that pure pneumatic logic is "fucky to implement" and "not quite the same as normal digital logic", you kind of get steered into working with a limited set of valve-circuit-equivalent Logic Units and not working problems the way you would if high pressure air wasn't involved. for those playing at home, i figured out what i was looking for, i didn't offer much to go off of; i was looking for which logic functions get used and how they're physically implemented, because i was looking to recreate them from scratch as a part of some exploratory 3d prints. found a nice cheat-sheet with the highlights all in one place: i was kind of hoping to use simple ball check valve variants as much as possible because they're relatively easy to get working acceptably at small scales, but i probably wont be able to avoid printing tiny valve spools and such, alas
|
# ? Nov 8, 2021 06:34 |
|
I restored this quite old alarm clock radio (brand DUX). Meguiars Plast-X works quite well for cleaning up these old things. I didn't manage to get the alarm function to work properly though, but I don't use it. I wanted to get rid of my modern clock radio because I too had the problem of a display being too loving bright. And the radio quality sucked poo poo through a straw. This one gets incredible reception on FM. Nobody here sends on MW anymore so that dates it a bit. But all I found inside was an old ferrite wound antenna, I didn't know those could get FM reception too, but it's all I found. (I cleaned the volume control afterwards) Looks a lot better irl than the photos indicate.
|
# ? Nov 8, 2021 07:20 |
|
His Divine Shadow posted:Nobody here sends on MW anymore so that dates it a bit. But all I found inside was an old ferrite wound antenna, I didn't know those could get FM reception too, but it's all I found. I've never seen a radio where the AM band was labeled MW. Is this a European thing?
|
# ? Nov 8, 2021 14:14 |
|
speaking of alarm clocks, I'd love to get an analog one, but how do you deal with there being no am/pm setting for the alarm? do you have to turn it off and then turn on the alarm again every day so it doesn't go off every 12 hours instead of 24? https://us.braun-clocks.com/collections/analogue-clocks/products/bc02x-braun-classic-analogue-travel-alarm-clock-grey
|
# ? Nov 8, 2021 15:04 |
His Divine Shadow posted:I restored this quite old alarm clock radio (brand DUX). Meguiars Plast-X works quite well for cleaning up these old things. I didn't manage to get the alarm function to work properly though, but I don't use it. No, it looks great. Those little boxes on the right side with the screwdriver slot, what are those guys? Pots?
|
|
# ? Nov 8, 2021 15:11 |
|
SkunkDuster posted:I've never seen a radio where the AM band was labeled MW. Is this a European thing? It's a European and old-American thing, yeah. On real old radios (especially ones that also have shortwave) it'd be labeled MW or Broadcast. I think calling it "AM" only really became a thing once FM was commonplace.
|
# ? Nov 8, 2021 15:13 |
|
|
# ? May 22, 2024 15:12 |
|
SkunkDuster posted:I've never seen a radio where the AM band was labeled MW. Is this a European thing? MW means Medium Wave, there's also Long Wave (LW) and Short Wave (SW). I'm actually sad about the decline of MW and SW radio in europe
|
# ? Nov 8, 2021 15:56 |