1 00:00:00,089 --> 00:00:03,690 In this video, I’d like to revisit a concept from my television series. 2 00:00:03,690 --> 00:00:07,720 I created this thumbnail for my video explaining analog color TV, 3 00:00:07,720 --> 00:00:10,230 and it’s been causing debate ever since. 4 00:00:10,230 --> 00:00:14,120 Though I had hoped my video on Trinitron would help illustrate my point and put the debate 5 00:00:14,120 --> 00:00:18,039 to rest, there was and is still much debate in the comments. 6 00:00:18,039 --> 00:00:23,289 It seems this debate comes mostly from semantics, and I’ll admit I see a gap in my explanation. 7 00:00:23,289 --> 00:00:24,410 So let’s try again. 8 00:00:24,410 --> 00:00:26,180 This is the TV in my kitchen. 9 00:00:26,180 --> 00:00:29,949 Like any TV on sale today, it produces an image by manipulating the brightness of many 10 00:00:29,949 --> 00:00:35,300 thousands (and these days millions) of individual picture elements called pixels, which is actually 11 00:00:35,300 --> 00:00:37,480 just short for picture element. 12 00:00:37,480 --> 00:00:41,130 There are a few different technologies in use these days, but at their core their job 13 00:00:41,130 --> 00:00:46,620 is to produce a set brightness value for the red, green, and blue components of each pixel. 14 00:00:46,620 --> 00:00:51,680 These are called subpixels, and in many LCD panels each subpixel is actually further divided 15 00:00:51,680 --> 00:00:53,820 into sub-sub pixels. 16 00:00:53,820 --> 00:00:58,210 This probably increases the total number of discrete brightnesses each color can make, 17 00:00:58,210 --> 00:01:03,379 and thus allows for more precise control over the panel and a larger number of possible colors. 18 00:01:03,379 --> 00:01:06,179 Someone please correct me if that’s not what the subdivisions do. 19 00:01:06,180 --> 00:01:10,320 The combination of red, green, and blue can create what appears to our eyes to be any color, 20 00:01:10,320 --> 00:01:13,860 because the way we perceive color (for those of us with normal trichromatic color 21 00:01:13,869 --> 00:01:18,240 vision, anyway) is through the ratio of stimulation between the three different cone cells in 22 00:01:18,240 --> 00:01:19,549 our eyes. 23 00:01:19,549 --> 00:01:23,900 Their primary sensitivities are red, green, and blue, so by using just these three colors, 24 00:01:23,900 --> 00:01:29,090 we can activate the cone cells in any given ratio and thus produce any apparent color. 25 00:01:29,090 --> 00:01:33,679 This biology hack is the result of the overlapping sensitivities of each cone cell. 26 00:01:33,679 --> 00:01:37,450 For example, yellow light stimulates both the red and green cone cells in your eyes 27 00:01:37,450 --> 00:01:42,109 roughly equally, as both of these cells can detect this wavelength of light. 28 00:01:42,109 --> 00:01:45,720 This means that to recreate what we see as yellow light, we don’t need to actually 29 00:01:45,720 --> 00:01:47,869 reproduce the same wavelength of light. 30 00:01:47,869 --> 00:01:52,229 Instead, we can artificially stimulate the red and green cone cells with just red and 31 00:01:52,229 --> 00:01:56,549 green light, and so long as the red and green cells receive the same relative stimulation 32 00:01:56,549 --> 00:02:00,200 as they did with honest-to-goodness yellow light, the brain can’t tell the difference 33 00:02:00,200 --> 00:02:02,020 and thinks it’s yellow. 34 00:02:02,020 --> 00:02:06,140 Simply outputting red, green, and blue light can produce any color to our eyes because 35 00:02:06,140 --> 00:02:10,310 when combined, it can produce the same ratios of stimulation between the three cone cells 36 00:02:10,310 --> 00:02:11,739 that any real color would. 37 00:02:11,739 --> 00:02:17,069 Anyway, the microprocessors inside this television are working together to make it all...happen, 38 00:02:17,069 --> 00:02:21,060 and the main image processor can tell the panel exactly what to do. 39 00:02:21,060 --> 00:02:25,700 The image on screen is coming from a Chromecast, and through the HDMI port on the television, 40 00:02:25,700 --> 00:02:31,019 the Chromecast can tell it exactly what each pixel needs to do to make this image, and 41 00:02:31,019 --> 00:02:33,599 the drivers inside the TV will make that happen. 42 00:02:33,599 --> 00:02:37,719 We can define the resolution of this display by counting how many pixels there are along 43 00:02:37,719 --> 00:02:38,719 each edge. 44 00:02:38,719 --> 00:02:42,480 I’d rather not actually do that, so I’ll just recite the specs here and tell you that 45 00:02:42,480 --> 00:02:49,049 there are 1,366 pixels along the bottom and 768 pixels along the sides, yes I know that’s 46 00:02:49,049 --> 00:02:54,750 not 720P but that’s the panel that’s in here, and that means that there are 1,049,088 47 00:02:54,750 --> 00:02:56,180 pixels on this screen. 48 00:02:56,180 --> 00:03:00,709 Generally, resolution is defined as X by Y, so we’d say this panel has a resolution 49 00:03:00,709 --> 00:03:03,319 of 1366 by 768. 50 00:03:03,319 --> 00:03:06,079 Now take a look at an old school CRT television. 51 00:03:06,079 --> 00:03:08,709 Get nice and close to it and you’ll find what appear to be pixels. 52 00:03:08,709 --> 00:03:11,249 There’s a neat division between red, green, and blue. 53 00:03:11,249 --> 00:03:14,189 The borders are defined, and it’s forming a grid, almost. 54 00:03:14,189 --> 00:03:18,400 But, you would be running a fool’s errand if you attempted to count the number of these 55 00:03:18,400 --> 00:03:21,540 “pixels” along the edges to determine this TV’s resolution. 56 00:03:21,540 --> 00:03:26,040 That’s because these aren’t pixels, and they don’t define its absolute resolution. 57 00:03:26,040 --> 00:03:28,950 To understand why, you need to look at a black and white television. 58 00:03:28,950 --> 00:03:31,700 Oh how convenient, a black a white television. 59 00:03:31,700 --> 00:03:35,719 Now with the set turned off, you can’t see any structure to this screen. 60 00:03:35,719 --> 00:03:40,180 Going back to the LCD TV, even when it’s off, that grid of pixels is still there. 61 00:03:40,180 --> 00:03:44,349 You need to shine a bright light onto it to see them, but the pixels are there as physical 62 00:03:44,349 --> 00:03:45,939 parts of the screen. 63 00:03:45,939 --> 00:03:48,550 But on this little CRT, there’s no grid to be seen. 64 00:03:48,550 --> 00:03:49,889 Let’s switch it on. 65 00:03:49,889 --> 00:03:53,430 With an image now on screen, you should be able to see a series of lines. 66 00:03:53,430 --> 00:03:55,900 In analog video, this is how the image is drawn. 67 00:03:55,900 --> 00:03:59,709 See, the CRT only has one “pixel” to deal with. 68 00:03:59,709 --> 00:04:03,269 At the rear of the picture tube is an electron gun which is projecting a single point of 69 00:04:03,269 --> 00:04:04,560 light at the screen. 70 00:04:04,560 --> 00:04:09,040 Then, electromagnets in the deflection yoke move this point around the screen very rapidly 71 00:04:09,040 --> 00:04:10,819 in a pattern called a raster. 72 00:04:10,819 --> 00:04:14,889 By varying the brightness of the point of light as it moves around the screen, an image 73 00:04:14,889 --> 00:04:15,969 can be made. 74 00:04:15,969 --> 00:04:18,949 The image is drawn as a series of stacked horizontal lines. 75 00:04:18,949 --> 00:04:23,870 In the US, roughly 480 lines are visible on the screen at once, drawn as two fields of 76 00:04:23,870 --> 00:04:28,060 240 lines 60 times per second as interlaced video. 77 00:04:28,060 --> 00:04:31,780 This is why standard definition is defined as 480i here in the States. 78 00:04:31,780 --> 00:04:35,669 I’ve made a video explaining how analog television works in greater detail, which 79 00:04:35,669 --> 00:04:37,770 can you find up above now or down below later. 80 00:04:37,770 --> 00:04:40,870 Now, this TV has no idea what it’s doing. 81 00:04:40,870 --> 00:04:42,340 It doesn’t have a microprocessor. 82 00:04:42,340 --> 00:04:43,669 It doesn’t have an HDMI port. 83 00:04:43,669 --> 00:04:47,300 It doesn’t have any digital circuitry of any kind. 84 00:04:47,300 --> 00:04:51,050 All it’s doing is looking for two pulses in the video signal, the horizontal blanking 85 00:04:51,050 --> 00:04:54,550 interval and the vertical blanking interval, in order to draw the image in the same place 86 00:04:54,550 --> 00:04:58,009 on the screen and not have it roll around like this. 87 00:04:58,009 --> 00:05:02,919 The nature of this signal is analog, and really all the signal does is tell the TV how bright 88 00:05:02,919 --> 00:05:03,919 to make the image. 89 00:05:03,919 --> 00:05:07,990 It’s just timed really really well so that each individual part of the screen is drawn 90 00:05:07,990 --> 00:05:11,590 with the correct brightness, as the position of the point of light is determined by the 91 00:05:11,590 --> 00:05:14,960 length of time that has elapsed from the start of the frame. 92 00:05:14,960 --> 00:05:18,669 And to be clear, we’re dealing with tiny fractions of a second since the beam moves 93 00:05:18,669 --> 00:05:20,000 incredibly quickly. 94 00:05:20,000 --> 00:05:21,669 So then, here’s the challenge. 95 00:05:21,669 --> 00:05:22,960 Where are the pixels? 96 00:05:22,960 --> 00:05:24,280 Well, there aren’t any! 97 00:05:24,280 --> 00:05:28,130 If I change the channel and we take a look at snow, you’ll see that there is no regularity 98 00:05:28,130 --> 00:05:30,139 whatsoever in this noise. 99 00:05:30,139 --> 00:05:34,590 If this image was defined by a grid of discrete picture elements, the borders between white 100 00:05:34,590 --> 00:05:38,830 and black sections should form columns of some sort, or at the very least there should 101 00:05:38,830 --> 00:05:41,069 be some clear vertical structure visible. 102 00:05:41,069 --> 00:05:42,069 But there isn’t. 103 00:05:42,069 --> 00:05:44,580 They appear completely randomly within the line. 104 00:05:44,580 --> 00:05:48,280 I can tell you exactly where the line is, but I can’t define any separation within 105 00:05:48,280 --> 00:05:49,539 the line itself. 106 00:05:49,539 --> 00:05:51,480 That’s completely arbitrary. 107 00:05:51,480 --> 00:05:54,260 Now here’s where the color CRT comes into play. 108 00:05:54,260 --> 00:05:56,060 Specifically one like this. 109 00:05:56,060 --> 00:05:59,470 This is a GE television, using a slot-mask CRT. 110 00:05:59,470 --> 00:06:03,970 Up close, it appears to have a similar grid structure to the LCD TV in my kitchen. 111 00:06:03,970 --> 00:06:05,920 So then, why aren’t these pixels? 112 00:06:05,920 --> 00:06:07,680 They’re what make up the image, right? 113 00:06:07,680 --> 00:06:09,669 Well, no, they aren’t. 114 00:06:09,669 --> 00:06:11,690 These are actually called phosphor dots. 115 00:06:11,690 --> 00:06:16,690 What they do is create specific targets for the red, green, and blue electron beams to hit. 116 00:06:16,690 --> 00:06:20,560 See to make a color image, we need to make a red, green, and blue image, and they need 117 00:06:20,560 --> 00:06:23,780 to be merged together somehow to appear as one. 118 00:06:23,780 --> 00:06:28,070 In the early days of color TV, there were all sorts of ideas being explored on how to 119 00:06:28,070 --> 00:06:29,500 produce an RGB image. 120 00:06:29,500 --> 00:06:33,110 I’ll throw another card up on my playlist on Television, because if this is the sort 121 00:06:33,110 --> 00:06:35,530 of thing that interests you you can take quite the nerdy deep dive. 122 00:06:35,530 --> 00:06:40,360 A color CRT is functionally identical to a black and white CRT, but it draws three separate 123 00:06:40,360 --> 00:06:42,009 images at once. 124 00:06:42,009 --> 00:06:46,310 Think of it like three picture tubes, one red, one green, and one blue, combined into 125 00:06:46,310 --> 00:06:48,110 a single picture tube. 126 00:06:48,110 --> 00:06:52,289 This combined tube has an electron gun for each color, but of course we also need a way 127 00:06:52,289 --> 00:06:55,560 to separate the colors in order to drive each one on its own. 128 00:06:55,560 --> 00:06:58,200 That’s what the phosphor dots do. 129 00:06:58,200 --> 00:07:02,639 They separate the face of the tube into a mosaic of red, green, and blue dots. 130 00:07:02,639 --> 00:07:06,660 The earliest color TVs used a pattern of phosphor dots that looked like this. 131 00:07:06,660 --> 00:07:09,800 These dots line up with a simple metal sheet just behind them. 132 00:07:09,800 --> 00:07:10,990 Let me show you what it does. 133 00:07:10,990 --> 00:07:12,849 Here I have a green flashlight. 134 00:07:12,849 --> 00:07:16,650 If I shine it at this poster board, it creates a flood of green light. 135 00:07:16,650 --> 00:07:20,860 But if I place a mask in front of it with a single hole, now the light can only make 136 00:07:20,860 --> 00:07:25,099 it through in a straight line between the flashlight and the hole, which produces just 137 00:07:25,099 --> 00:07:26,610 a dot on the poster board. 138 00:07:26,610 --> 00:07:28,939 Now, here’s a red flashlight. 139 00:07:28,939 --> 00:07:32,280 Watch what happens when I put it next to the green flashlight. 140 00:07:32,280 --> 00:07:35,789 Because the red flashlight is in a slightly different position from the green one, the 141 00:07:35,789 --> 00:07:38,669 light it makes can’t take the same path as the green light. 142 00:07:38,669 --> 00:07:43,310 It will go through the hole at a different angle, so the dot it produces appears next 143 00:07:43,310 --> 00:07:44,719 to the green one. 144 00:07:44,719 --> 00:07:48,720 Now if I add a third, blue flashlight and put it in between the red and the green just 145 00:07:48,720 --> 00:07:52,960 above them, a blue dot appears below the red and green dots. 146 00:07:52,960 --> 00:07:56,280 If I take the mask away, it creates just a wash of white light. 147 00:07:56,280 --> 00:08:00,150 But with the mask in place, it produces three small dots of light in the same arrangement 148 00:08:00,150 --> 00:08:03,580 as the flashlights themselves, though it’s mirrored and upside down. 149 00:08:03,580 --> 00:08:07,330 If I add a second hole the to mask, the same pattern appears right next to the first set 150 00:08:07,330 --> 00:08:08,430 of dots. 151 00:08:08,430 --> 00:08:12,370 If I keep going and make a bunch of small holes in this offset pattern, what we get 152 00:08:12,370 --> 00:08:15,490 is a mosaic pattern of red, green, and blue dots. 153 00:08:15,490 --> 00:08:19,539 This is happening because the flashlights are arranged in a triangle, and at every hole 154 00:08:19,539 --> 00:08:25,090 in the mask the beams converge and cross over to project the opposite image on the screen. 155 00:08:25,090 --> 00:08:28,169 Notice how similar this pattern is to this color CRT. 156 00:08:28,169 --> 00:08:32,220 See, if I aim these three flashlights together at the poster board, their beams just blend 157 00:08:32,220 --> 00:08:34,470 together and make what appears to be white. 158 00:08:34,470 --> 00:08:38,220 This is what would happen if we used a color CRT without a mask. 159 00:08:38,220 --> 00:08:41,600 But if I place the mask in front of it, which is just a piece of aluminum foil with some 160 00:08:41,600 --> 00:08:45,890 holes punched in it, suddenly a pattern just like the phosphor dots appears. 161 00:08:45,890 --> 00:08:51,020 Now, the red beam can only hit specific parts of the screen, and the blue and green beams 162 00:08:51,020 --> 00:08:52,970 can’t hit those points. 163 00:08:52,970 --> 00:08:57,110 Because the light sources are physically separated, they can only make their way through the holes 164 00:08:57,110 --> 00:08:58,960 at specific angles. 165 00:08:58,960 --> 00:09:03,440 The mask puts the red targets in the shadow of the blue and green beams. 166 00:09:03,440 --> 00:09:06,350 The mask casts a shadow on the targets. 167 00:09:06,350 --> 00:09:07,400 Wait a minute, 168 00:09:07,400 --> 00:09:08,740 shadow mask! 169 00:09:08,740 --> 00:09:11,780 What’s important to realize here is that the mask is what’s creating the pattern 170 00:09:11,780 --> 00:09:12,780 of dots. 171 00:09:12,780 --> 00:09:16,970 The flashlights are firing indiscriminately at the mask, but the mask will always force 172 00:09:16,970 --> 00:09:20,200 each beam into the correct location on the other side. 173 00:09:20,200 --> 00:09:24,800 This means that no matter what sort of pattern of light the beams or flashlights are creating, 174 00:09:24,800 --> 00:09:28,230 it will always appear as a series of dots on the other side. 175 00:09:28,230 --> 00:09:32,090 So if we go back to our picture tube, what you see as a viewer are the phosphor dots 176 00:09:32,090 --> 00:09:36,030 which are the targets for each individual color electron beam. 177 00:09:36,030 --> 00:09:39,670 Inside the tube, directly behind them, is the metal sheet with holes in it, which always 178 00:09:39,670 --> 00:09:43,900 ensures the color components stay separated and project onto the phosphor dots in the 179 00:09:43,900 --> 00:09:45,440 correct orientation. 180 00:09:45,440 --> 00:09:49,260 But the key here is that they do not change how the image is drawn. 181 00:09:49,260 --> 00:09:53,680 Just like the black and white television, this TV is stacking horizontal lines. 182 00:09:53,680 --> 00:09:57,330 In fact, these two televisions are receiving the same exact signal. 183 00:09:57,330 --> 00:10:00,860 The difference is that the color TV can recover the color information that’s superimposed 184 00:10:00,860 --> 00:10:04,080 in the signal through quadrature amplitude modulation-- 185 00:10:04,080 --> 00:10:06,800 Don’t worry too much about the specifics of that-- 186 00:10:06,920 --> 00:10:10,920 and it can then adjust the relative intensities of the three color components. 187 00:10:10,920 --> 00:10:15,030 But since that means it’s effectively drawing drawing three different sets of lines at once, 188 00:10:15,030 --> 00:10:17,730 it needs a way to keep the colors from crossing over. 189 00:10:17,730 --> 00:10:19,840 That’s what the shadow mask does. 190 00:10:19,840 --> 00:10:24,210 Remember, even though the flashlights were just blasting away at the mask, the mask made 191 00:10:24,210 --> 00:10:28,260 sure each part was separated into little dots on the other side. 192 00:10:28,260 --> 00:10:32,800 From this side of the picture tube, it’s just like taking a black and white CRT, then 193 00:10:32,800 --> 00:10:37,640 drawing a grid on top of it, and then coloring each little cell in with red green or blue. 194 00:10:37,640 --> 00:10:42,150 The only functional difference between a true color CRT and a black and white CRT with lines 195 00:10:42,150 --> 00:10:46,720 and colors drawn on top is that the mask behind the phosphor dots of the color picture tube 196 00:10:46,720 --> 00:10:51,630 ensures the colors stay separated, and thus allows for individual control of each color. 197 00:10:51,630 --> 00:10:56,100 Now, this style of shadow mask makes it hard to even define what could be a pixel. 198 00:10:56,100 --> 00:11:01,400 Assuming each pixel contains one red, one blue, and one green dot, well first of all 199 00:11:01,400 --> 00:11:05,970 they’re triangular, but then each one changes orientation as you move on and really it’s 200 00:11:05,970 --> 00:11:10,380 just a mess but this style of CRT, which uses a slot-mask display, does make a pattern that 201 00:11:10,380 --> 00:11:12,900 really looks like there are pixels. 202 00:11:12,900 --> 00:11:17,020 These CRTs arrange the electron guns in a line, and rather than use a mask with round 203 00:11:17,020 --> 00:11:19,230 holes they use a mask with small slots. 204 00:11:19,230 --> 00:11:24,300 This allows more of the beam energy to pass through the mask and makes a brighter image. 205 00:11:24,300 --> 00:11:26,070 And here’s where the semantics comes in. 206 00:11:26,070 --> 00:11:30,910 I’ll grant you that the picture is “made up of” these groupings of phosphors. 207 00:11:30,910 --> 00:11:34,720 You could say that they are elements of the picture, and thus are pixels. 208 00:11:34,720 --> 00:11:39,580 But this ignores the fact that they are only there as a side-effect of the need for color 209 00:11:39,580 --> 00:11:40,580 separation. 210 00:11:40,580 --> 00:11:44,320 They are in front of what makes the image, and are not the actual building blocks of 211 00:11:44,320 --> 00:11:45,320 the image. 212 00:11:45,320 --> 00:11:47,410 To put it another way, here’s a window screen. 213 00:11:47,410 --> 00:11:51,310 If I put it in front of this album cover, does that become a pixel? 214 00:11:51,310 --> 00:11:54,580 Have I pixellated the image by placing a grid in front of it? 215 00:11:54,580 --> 00:11:58,890 Or have I simply compartmentalized parts of the image into little square cells? 216 00:11:58,890 --> 00:12:01,620 And here’s the part that I think is hardest to understand. 217 00:12:01,620 --> 00:12:06,510 The phosphor dots do not in any way define the maximum amount of detail that can be displayed 218 00:12:06,510 --> 00:12:08,010 on the screen. 219 00:12:08,010 --> 00:12:10,050 That may sound silly, but hear me out. 220 00:12:10,050 --> 00:12:14,310 All they really do is define the maximum color resolution of the display. 221 00:12:14,310 --> 00:12:16,090 Let’s go back to this CRT. 222 00:12:16,090 --> 00:12:20,470 I’ve only shown it in close-up because this is a laughable little 5 inch color TV boombox 223 00:12:20,470 --> 00:12:22,460 from some point in the 1980’s. 224 00:12:22,460 --> 00:12:27,320 The dot pitch, that’s the fineness of the dots, is very poor on this TV. 225 00:12:27,320 --> 00:12:30,100 I mean, you can’t really blame it, as it’s only got 5 inches to work with. 226 00:12:30,100 --> 00:12:34,620 This means it can’t display much color detail, but it can display as much brightness detail 227 00:12:34,620 --> 00:12:36,320 as any television. 228 00:12:36,320 --> 00:12:37,520 Let me boot up Kingdom Hearts. 229 00:12:38,240 --> 00:12:39,020 [PS2 Game Start noise] 230 00:12:39,720 --> 00:12:42,760 OK, so take a look at the menu in the bottom left corner. 231 00:12:42,770 --> 00:12:46,920 If we look at the black and white TV, we can see that there are about 10 or 12 lines defining 232 00:12:46,920 --> 00:12:49,590 the height of the letter M in Magic. 233 00:12:49,590 --> 00:12:54,470 If we count the number of dots along the height of the M, we also get about 10, maybe 11. 234 00:12:54,470 --> 00:12:58,950 But, each cluster of three color dots spans the height of two lines. 235 00:12:58,950 --> 00:13:04,170 We appear to have only half the color resolution as we do brightness resolution in this CRT. 236 00:13:04,170 --> 00:13:07,070 Look at how infrequently a red dot appears among blue and green. 237 00:13:07,070 --> 00:13:11,550 That’s they key here, we’re not getting a lot of complete RGB clusters among the word 238 00:13:11,550 --> 00:13:15,390 Magic, but we can still clearly see the shape of the word Magic. 239 00:13:15,390 --> 00:13:19,440 You can even see the how the center of the A is darker than the rest, but only this one 240 00:13:19,440 --> 00:13:21,740 red dot is actually darker. 241 00:13:21,740 --> 00:13:25,200 But the thing is, from a normal viewing distance, you can’t really tell how poor the color 242 00:13:25,200 --> 00:13:26,200 resolution is. 243 00:13:26,200 --> 00:13:30,520 Once you’re far enough away that you can’t discern the individual phosphor dots, the 244 00:13:30,520 --> 00:13:32,320 image appears more or less normally. 245 00:13:32,320 --> 00:13:36,890 This is in contrast to a digital LCD panel, where the pixels themselves define the shape 246 00:13:36,890 --> 00:13:38,070 of an image. 247 00:13:38,070 --> 00:13:42,920 If I want to draw a letter M using a grid of 10 X 10 pixels, well I can say how bright 248 00:13:42,920 --> 00:13:44,680 I want each pixel to be. 249 00:13:44,680 --> 00:13:49,480 Then I can tell the display what to do with each of these 100 pixels to make an M. But 250 00:13:49,480 --> 00:13:52,960 in the case of a CRT, it’s drawing the M like this, in Lines. 251 00:13:52,960 --> 00:13:57,120 That shape is then forced into the grid of phosphor dots, and wherever it lands will 252 00:13:57,120 --> 00:13:59,300 tell you which dots get lit up. 253 00:13:59,300 --> 00:14:00,500 And that’s the key difference. 254 00:14:00,500 --> 00:14:04,690 In digital video, the pixels define the shape of the image, logically. 255 00:14:04,690 --> 00:14:09,330 In analog video, the shape of the image defines which phosphor dots are lit. 256 00:14:09,330 --> 00:14:11,410 You can see this effect with the small TV. 257 00:14:11,410 --> 00:14:15,560 This screen really suffers where very small, colored elements appear. 258 00:14:15,560 --> 00:14:20,170 If I open the pause menu and look at these stats, some of it is very hard to read. 259 00:14:20,170 --> 00:14:23,820 That’s because this text is colored green, so the blue and red guns pretty much 260 00:14:23,820 --> 00:14:26,300 don’t fire when drawing it. 261 00:14:26,310 --> 00:14:30,890 Since the green dots are so far apart and this text is so small, if the text to be drawn 262 00:14:30,890 --> 00:14:34,060 lies between the green dots, it just won’t get drawn. 263 00:14:34,080 --> 00:14:37,950 It’s not like the gun isn’t firing, it’s just that for the entire section here, the 264 00:14:37,950 --> 00:14:42,760 text is in the shadow of the green gun, so none of its energy is able to light up the screen. 265 00:14:42,760 --> 00:14:46,620 And that’s the point I’m trying to make when I say “these are not pixels”. 266 00:14:46,620 --> 00:14:52,080 These two TVs are displaying the same image and they both have 480 lines of resolution. 267 00:14:52,080 --> 00:14:57,250 But this little CRT has fewer phosphor dots, so it can’t recreate color as precisely 268 00:14:57,250 --> 00:14:59,030 as the larger TV. 269 00:14:59,030 --> 00:15:03,030 But that doesn’t mean it’s not conveying the same 480 lines of resolution. 270 00:15:03,030 --> 00:15:05,730 It is, just in brightness only. 271 00:15:05,730 --> 00:15:09,660 It loses detail in the color department, and as a side-effect it can’t reproduce some 272 00:15:09,660 --> 00:15:11,030 fine color details. 273 00:15:11,030 --> 00:15:16,250 A CRT television has no control over how the three electron beams interact with the mask. 274 00:15:16,250 --> 00:15:20,570 The combined beam can land and will land wherever it wants, and it’s then up to the mask to 275 00:15:20,570 --> 00:15:22,590 separate the color components. 276 00:15:22,590 --> 00:15:26,030 The clusters of phosphor dots are there just because they need to be. 277 00:15:26,030 --> 00:15:29,390 They can be a different size, a different shape, and some TV’s don’t even split 278 00:15:29,390 --> 00:15:30,640 them up beyond vertical stripes. 279 00:15:30,640 --> 00:15:31,880 Trinitron. 280 00:15:31,880 --> 00:15:36,200 As a final point, which I think is at the crux of the issue, in an LCD, OLED, Plasma, 281 00:15:36,200 --> 00:15:40,680 or any sort of digital display, the grid of pixels is an active matrix. 282 00:15:40,680 --> 00:15:44,990 The display has an electrical connection to each one of them, and can talk to it. 283 00:15:44,990 --> 00:15:48,040 The shadow mask and phopshor dots are a passive component of the CRT. 284 00:15:48,040 --> 00:15:49,300 They don’t get addressed. 285 00:15:49,300 --> 00:15:51,070 They don’t have an electrical connection. 286 00:15:51,070 --> 00:15:52,690 They’re just there. 287 00:15:52,690 --> 00:15:57,360 Sure, the TV does “control” which ones get lit up, but it’s not done with logical 288 00:15:57,360 --> 00:16:00,120 control or any precision whatsoever. 289 00:16:00,120 --> 00:16:04,440 Just like the black and white CRT, wherever the beam lands is what part gets lit up. 290 00:16:04,440 --> 00:16:08,570 Now this isn’t to say that an analog TV can’t produce an image made of pixels. 291 00:16:08,570 --> 00:16:11,300 Surely it can, it's just making small squares. 292 00:16:11,300 --> 00:16:14,040 And in fact that’s what it's been doing throughout all of this video. 293 00:16:14,040 --> 00:16:19,040 A PlayStation 2, DVD player, Roku box, or any digital source with a composite output 294 00:16:19,040 --> 00:16:24,740 will take its logical 640X480 digital grid and convert that to the 480 horizontal lines 295 00:16:24,740 --> 00:16:26,340 to drive the TV. 296 00:16:26,340 --> 00:16:29,920 But I guarantee you those ethereal pixels in the logic circuits of the digital source 297 00:16:29,920 --> 00:16:33,230 won’t be lining up nicely with these phosphor dots. 298 00:16:33,230 --> 00:16:34,850 They simply don’t need to. 299 00:16:34,850 --> 00:16:36,980 Thanks for watching, I hope you enjoyed the video! 300 00:16:36,980 --> 00:16:40,240 If this is your first time coming across the channel and you liked what you saw, please 301 00:16:40,240 --> 00:16:41,630 consider subscribing. 302 00:16:41,630 --> 00:16:44,550 As always, thank you to everyone who supports this channel on Patreon! 303 00:16:44,550 --> 00:16:48,490 If you’re interested in making a voluntary contribution to the channel as well, please 304 00:16:48,490 --> 00:16:49,730 check out my Patreon page. 305 00:16:49,730 --> 00:16:53,790 There’s a link on your screen or you can find one down below in the description. 306 00:16:53,790 --> 00:16:56,120 Thanks for your consideration, and I’ll see you next time!