it's almost ridiculous how good you are at explaining things. Well done indeed. You should seriously consider a job doing that, if it's not your job already.
Had the same exact thought. I don't think I'd have application for most of the other videos on your channel, but I feel like I want to watch every single one of them just cause I'm sure I'd understand anything explained like this
You have a talent at explaining things. Never seen anybody expressing thoughts so well structured and consistent. I understood everything without referencing to external resources. Thank you very much from Russia.
This really is the best video on this subject on RU-vid. Well done! I understand the difference between these two formats way better now. Props for using real world examples to explain concepts too. That's a sign of a great teacher right there!
Thank you mate. You have a great skill in explaining things. Your voice was so relaxing and your explanation were so confident and bold that I could literally feel the RGB vs YUV differences though I was stuck understanding the core differences while searching.
As I was watching this I was hoping you'd touch on the old B&W and colour TV problem. Well done! I remember at the age of 5-ish when my parents bought our first colour TV, I asked them to make sure I could still watch all the same shows, specifically Sesame Street! It seems even at that age I could at some level understand the potential problems with changing a broadcast system and maintaining compatibility!
I was just checking out the NUMA video and realized that out of all the channels which cover Computer Science Topics. You have the best methodology of teaching. The audio quality is really good and the visualizations are enough and not daunting for newbies. I think you can do something big out of this. Maybe start with an Operating System playlist or Algorithms. Seriously dude, If you bring the simplicity out of complex topics in advanced subjects like AI and ML I'd stop watching 3Blue1Brown haha.
You are perfectly right. I have a Skyworth TV panel connected to a Geforce gtx1070 ti, and used to set RGB full range. Always with a default wrong tonality tending to blue. To have a more warm white I've tried to combine a slitghtly higher red and green, but the result was a loss of saturation. When I've switched to YCbCr the colors went back to the right saturation keeping the mix previously choosen, and it definitely looks like all the colors have the same saturation. The fact itself is weird to me because at the end the conversion should be rgb, but somehow it worked, somehow the colors processing is better.
Fantastic video! I would suspect that it's not that our eyes are being bad at seeing color. It's our eyes/brain being great at compensating for shifts in color, reconstructing the image for our perception.
fabulously explained. I searched for a video explaining this, because googling only threw two kinds of results: either too trivial, with gamers forums talking about "which is better: rgb or yuv", without really penetrating the subject, or with too abstract literature such as Wikipedia's entry on yuv, very cryptic. Your explanation is remarkably better to all previous findings. Cheers and you just earned a new suscriber. Regards from Colombia, South America.
Your delivery/explanations reminded me of a couple of favourite teachers from my old high school days (looooong ago, heh) and I mean that in a good way. Beautifully done, and subbed!
Well I’ll be damned. Someone finally explained it in depth. Most videos just explain it in a way that is straight from the dictionary. Even a dictionary way to help “show” it. Anyways thanks!
YCbCr 4.4.4 format output seems also have anti-aliasing effect. When I'm change this setting on Nvidia control panel all round icon in desktop and round object in CSGO is anti-aliased, this effect is maybe just the combination of my monitor and GPU but if this work you just got free performance anti-aliasing.
RGB is a three-dementional colour space. Colour systems would be sRGB, Abode RGB and so on, they are a part of the same space, but differ in coordinates.
Yes indeed. I wasn't denying that right? Edit: Oh well just noticed I did call RGB a "system" a bunch of times, the thing is, these terms are loosely defined and a lot people use different terms to describe these things. In fact, "color system" is usually used to describe systems like RAL and Pantone. Things like sRGB and Adobe RGB are often simply referred to as "color space" or, if you want to be fancy "color subspace".
Well, there is “people” approach and strictly technical one. If we are trying to be mathematicaly correct then we have a bunch of vector spaces with different axis. RGB space is a “cube” of sorts, CMYK is a four-dimensional “tesseract”. Inside that space we define coordinates to describe a colour. Those coordinates are strictly mathematical and are defined by CIA RGB 1931 if I'm not mistaken. We then use those mathematical definitions to agree upon the colour of roadsigns between countries, for example. And only upon that space different consortiums of tech companies, TV broadcasting firms and other entities try to come up with industrial standards like Rec. 709 or the new one for HLG/HDR. The same goes for sRGB, Adobe RGB and others, they are temporary and in constant change. While CIA RGB is constant, it doesn't change for almost a century and many people confuse RGB space with sRGB system (and it's outdated and being dropped in favour of Adobe RGB). There also is a bunch of formulas to convert image from one system to another we have to use for different devices and means of image transfer. So to comprehend all that complexity I would have used terms that are used by International Union for Electrocommunication (I don't know how it's called in English, in Russian that's Международный Союз Электросвязи). But surely you haven't made any mistakes, that's just my suggestion, your content is great as always!
