WEBVTT Kind: captions Language: en 00:00:00.080 --> 00:00:02.699 The year is 1954. 00:00:02.699 --> 00:00:08.320 After much deliberation, confusion, and fussing about, the FCC had, for the second time, settled 00:00:08.320 --> 00:00:12.599 on the way color TV would be transmitted in the United States, and Westinghouse would 00:00:12.599 --> 00:00:17.090 release the first commercially produced color television set using the new standard, the 00:00:17.090 --> 00:00:19.550 Westinghouse H840CK15. 00:00:19.550 --> 00:00:22.790 Ah, the H840CK15. 00:00:22.790 --> 00:00:24.150 Really rolls off the tongue. 00:00:24.150 --> 00:00:28.310 RCA would follow with the more famous CT-100 weeks later. 00:00:28.310 --> 00:00:32.020 The reason for the fussing about had to do with compatibility. 00:00:32.020 --> 00:00:36.600 Most experimental color televisions used complicated schemes to create a color image which would 00:00:36.600 --> 00:00:40.290 not be compatible with existing black and white televisions sets. 00:00:40.290 --> 00:00:45.090 It seemed like this was an inevitability, and the FCC did briefly approve simultaneous 00:00:45.090 --> 00:00:49.730 transmission of both color and black and white signals, creating a fragmented television 00:00:49.730 --> 00:00:50.860 landscape. 00:00:50.860 --> 00:00:56.010 But the National Television System Committee, NTSC, having backed the work of RCA, would 00:00:56.010 --> 00:00:59.290 save the day by introducing a new color system. 00:00:59.290 --> 00:01:04.750 Their compatible color managed to, in a sense, hide the color signal within the black-and-white 00:01:04.750 --> 00:01:05.750 transmission. 00:01:05.750 --> 00:01:09.261 The benefits of a single transmission for both color and black and white television 00:01:09.261 --> 00:01:14.270 sets were obvious, and though it took some convincing, the FCC would later decide that 00:01:14.270 --> 00:01:16.670 new NTSC color was the way to go. 00:01:16.670 --> 00:01:20.740 To understand this story, we need to look at how color images are made. 00:01:20.740 --> 00:01:25.430 Our eyes actually only see three colors of light--Red, green, and blue. 00:01:25.430 --> 00:01:29.620 Our brains interpret other colors by comparing the amount of each of the three colors we 00:01:29.620 --> 00:01:34.470 see, for example yellow light will stimulate both the red and green cells in your eyes 00:01:34.470 --> 00:01:39.730 in close to equal amounts as yellow lies between green and red on the color spectrum. 00:01:39.730 --> 00:01:44.000 We knew this for a long time before television was around, in fact this knowledge can be 00:01:44.000 --> 00:01:49.110 traced back to 1802 with Thomas Young’s pretty correct postulation. 00:01:49.110 --> 00:01:53.860 This was further refined in 1850 by Hermann von Helmholtz and thenceforth was known as 00:01:53.860 --> 00:01:56.360 the Young-Helmholtz theory. 00:01:56.360 --> 00:02:00.800 Experiments in color photography were done almost as soon as we figured out photography, 00:02:00.800 --> 00:02:05.170 and the three-color method, as suggested by Scottish physicist James Clerk Maxwell based 00:02:05.170 --> 00:02:07.970 on the Young-Helmholtz theory, worked perfectly. 00:02:07.970 --> 00:02:12.820 Plenty of very old color photographs exist, and this one is perhaps my favorite. 00:02:12.820 --> 00:02:17.020 The only clue that this photograph is from 1912 is the moustache. 00:02:17.020 --> 00:02:21.020 Anyway, with humans possessing the knowledge of how to recreate an image in color using 00:02:21.020 --> 00:02:25.650 an RGB system, it should be no surprise that we threw our hats into the ring for color 00:02:25.650 --> 00:02:27.920 television as soon as we figured out television. 00:02:27.920 --> 00:02:32.450 The first color television demonstration was done by none other than our friend John Logie 00:02:32.450 --> 00:02:35.300 Baird, who adapted his mechanical television system 00:02:35.300 --> 00:02:39.190 to produce color all the way back in 1928. 00:02:39.190 --> 00:02:42.920 But as we know, mechanical television was not meant for this world. 00:02:42.920 --> 00:02:45.920 CRT based electronic television was far superior. 00:02:45.920 --> 00:02:50.610 So, we got to work figuring on out how to adapt a CRT into a color display. 00:02:50.610 --> 00:02:55.860 The most obvious thing would be to simultaneously transmit three separate television images, 00:02:55.860 --> 00:02:58.900 with each representing one of the RGB channels. 00:02:58.900 --> 00:03:03.470 A special camera with a beam-splitting arrangement behind the lens could send the image to three 00:03:03.470 --> 00:03:08.010 separate iconoscope tubes, each with a colored filter in front of it. 00:03:08.010 --> 00:03:13.130 Later cameras would use dichroic glass to simultaneously split the beam and filter color. 00:03:13.130 --> 00:03:17.120 This would cause each tube to only detect light of that color. 00:03:17.120 --> 00:03:21.090 On the receiving end, three separate television picture tubes could receive each of these 00:03:21.090 --> 00:03:22.590 three signals. 00:03:22.590 --> 00:03:25.420 RCA experimented with just such a system. 00:03:25.420 --> 00:03:29.980 The three CRT displays were tinted either red, green, or blue, to match the colors detected 00:03:29.980 --> 00:03:31.780 by the tubes in the camera. 00:03:31.780 --> 00:03:34.470 The trick was combining the images together. 00:03:34.470 --> 00:03:39.459 An optical system similar to that of the camera could re-combine the output of the three CRTs, 00:03:39.459 --> 00:03:41.910 but this didn’t work too well. 00:03:41.910 --> 00:03:46.819 Though CRT projection systems would use this approach years later, at the time it wasn’t 00:03:46.819 --> 00:03:47.950 a great option. 00:03:47.950 --> 00:03:51.840 Besides the fact that each television set using this system would cost at least three 00:03:51.840 --> 00:03:56.740 times as much as an ordinary television, given the fact it essentially is three TVs, the 00:03:56.740 --> 00:04:01.379 resulting image had to be recombined onto a frosted screen, and with the comparatively 00:04:01.379 --> 00:04:04.640 dim tubes of the time, it just wasn’t great. 00:04:04.640 --> 00:04:08.709 The real trick would be to figure out how to use just one picture tube. 00:04:08.709 --> 00:04:11.300 Once again, John Logie Baird appeared. 00:04:11.300 --> 00:04:15.740 In 1940, he demonstrated a sequential color system using CRT technology. 00:04:15.740 --> 00:04:20.479 But, I’ve also found a source indicating this was in 1939. 00:04:20.479 --> 00:04:24.809 Also in 1940, CBS demonstrated their sequential color system. 00:04:24.809 --> 00:04:27.900 Regardless of who was first, both of these systems are similar. 00:04:27.900 --> 00:04:31.939 These work by placing a large disk in front of a black and white picture tube. 00:04:31.939 --> 00:04:36.539 This disc contained alternating sections tinted red, green, and blue. 00:04:36.539 --> 00:04:40.309 This disc was quite a bit larger than the tube, so that only one color was in front 00:04:40.309 --> 00:04:41.900 of the tube at one time. 00:04:41.900 --> 00:04:45.900 The disc would tint the apparent color of the picture tube, and by spinning the disc 00:04:45.900 --> 00:04:50.680 quickly in front of it, it would rapidly produce a red, then green, then blue image. 00:04:50.680 --> 00:04:55.180 If you do this fast enough, persistence of vision will kick in, and you won’t notice it. 00:04:55.180 --> 00:04:59.020 A similar disc was placed in front of the camera tube in the studio, so that it would 00:04:59.020 --> 00:05:01.680 only see red, then green, the blue light. 00:05:01.680 --> 00:05:06.879 So long as you can synchronize the camera and TV, you could transmit a full color image. 00:05:06.879 --> 00:05:09.939 But now is where we run into problems of compatibility. 00:05:09.939 --> 00:05:13.659 You could in theory just slap one of these discs in front of both a conventional camera 00:05:13.659 --> 00:05:18.819 and conventional TV, and it would work, but it would be painful to view. 00:05:18.819 --> 00:05:22.900 With a framerate of 30 frames per second, each color would only appear in front of the 00:05:22.900 --> 00:05:25.029 tube 10 times per second. 00:05:25.029 --> 00:05:26.050 This would be obvious. 00:05:26.050 --> 00:05:30.930 In fact, avert your eyes if you suffer from epilepsy, it would look like this. 00:05:30.930 --> 00:05:32.520 This isn’t great, now is it? 00:05:32.520 --> 00:05:37.520 Even if you spun the disc twice as fast and tinted each consecutive field and not frame, 00:05:37.520 --> 00:05:39.750 it would still be very visible. 00:05:39.750 --> 00:05:40.750 And now, 00:05:40.750 --> 00:05:44.979 I was pleasantly surprised to learn of Guillermo González Camarena, a Mexican inventor who 00:05:44.979 --> 00:05:49.879 applied for a patent detailing a system much like we’ve just discussed in 1940. 00:05:49.879 --> 00:05:52.849 Thanks to multiple commenters for letting me know about him. 00:05:52.849 --> 00:05:54.990 Information about him is very spotty. 00:05:54.990 --> 00:05:59.279 For example, there is a Wikipedia entry on him, but it’s pretty paltry and also somewhat 00:05:59.279 --> 00:06:00.839 contradicts itself. 00:06:00.839 --> 00:06:04.479 The actual patent is easily accessible, though, and it’s worth taking a look at. 00:06:04.479 --> 00:06:08.360 We’ll explore Camarena’s work in more detail in the next video, including his work 00:06:08.360 --> 00:06:12.530 on two-color TV, but first I need to address a small issue. 00:06:12.530 --> 00:06:16.689 His patent was to adapt an existing black-and-white set to color. 00:06:16.689 --> 00:06:21.160 This wouldn’t look too good, as we just discussed, as the color wouldn’t change fast enough. 00:06:21.240 --> 00:06:25.720 It would technically work, but its practical viewability would be questionable. 00:06:25.729 --> 00:06:30.529 He had the idea fundamentally correct, but the CBS system produced far and away better 00:06:30.529 --> 00:06:31.529 results. 00:06:31.529 --> 00:06:36.490 Also, CBS’s demonstration to the press of their field-sequential color-system happened 00:06:36.490 --> 00:06:40.020 just 9 days after Guillermo filed his patent. 00:06:40.020 --> 00:06:43.610 So here we are again, with multiple people who could be given credit as the inventor 00:06:43.610 --> 00:06:45.219 of color TV. 00:06:45.219 --> 00:06:47.409 So we’ll just say it was a group effort. 00:06:47.409 --> 00:06:50.330 CBS actually brought their system to commercial use. 00:06:50.330 --> 00:06:55.360 However, the CBS system radically altered the way television transmissions were done. 00:06:55.360 --> 00:06:59.199 To get around the high flicker caused by the color wheel, they elected to increase the 00:06:59.199 --> 00:07:05.419 field rate from 60 hz to 144 hz, but in order for each two-field frame to be completely 00:07:05.419 --> 00:07:10.020 colored, the wheel needed to cycle through the RGB pattern twice with each frame, reducing 00:07:10.020 --> 00:07:13.569 the effective frame rate to 24 frames per second. 00:07:13.569 --> 00:07:17.529 This was very effective at making the color wheel hard to notice since it changed colors 00:07:17.529 --> 00:07:23.539 144 times per second, but it came at the huge expense of wiping out any compatibility with 00:07:23.539 --> 00:07:25.499 existing black and white televisions. 00:07:25.499 --> 00:07:27.819 But, the color wheel was simple. 00:07:27.819 --> 00:07:31.780 Aside from the extra circuitry required to synchronize the wheel with the correct fields, 00:07:31.780 --> 00:07:35.710 it was really a run-of-the-mill black and white TV and camera, both modified with a 00:07:35.710 --> 00:07:40.370 much higher scan frequency, and with a spinning color wheel in front of each of them. 00:07:40.370 --> 00:07:44.849 It was easy to produce and worked reasonably well, and so the FCC decided they would allow 00:07:44.849 --> 00:07:47.569 broadcasting of the CBS color system. 00:07:47.569 --> 00:07:53.080 On June 25th, 1951, the first network color television broadcast occurred. 00:07:53.080 --> 00:07:59.069 But this whole time, RCA was trying to convince the FCC of their “compatible color” system. 00:07:59.069 --> 00:08:02.949 They understood that if would be really great if you could broadcast a color transmission 00:08:02.949 --> 00:08:06.520 that could still be viewed with the black and white televisions already in service. 00:08:06.520 --> 00:08:11.069 And to make that possible, they needed a whole new type of picture tube. 00:08:11.069 --> 00:08:12.999 Enter the shadow mask. 00:08:12.999 --> 00:08:16.939 One simple way to make a color picture tube would be to create a pattern of alternating 00:08:16.939 --> 00:08:20.319 red, green, and blue dots on the inside surface. 00:08:20.319 --> 00:08:24.259 From far enough away these dots would blend into each other and wouldn’t be noticeable. 00:08:24.259 --> 00:08:29.559 So, RCA used picture tubes which contained just an arrangement, with each dot being filled 00:08:29.559 --> 00:08:32.760 with either a red, green, or blue phosphor. 00:08:32.760 --> 00:08:36.840 This tube could be scanned at the same field and frame rate as an ordinary black and white 00:08:36.840 --> 00:08:39.630 tube, but could produce a full color image. 00:08:39.630 --> 00:08:42.640 But now you need a way to control which dots are lit up. 00:08:42.640 --> 00:08:45.470 If they all worked together, it would simply appear as black and white. 00:08:45.470 --> 00:08:48.910 You need a way to control where the electron beam lands. 00:08:48.910 --> 00:08:51.160 The shadow mask is just the solution. 00:08:51.160 --> 00:08:56.560 The 1938 invention of German man Werner Flechsig is a sheet of metal with a bunch of tiny holes 00:08:56.560 --> 00:08:57.820 punched through it. 00:08:57.820 --> 00:09:01.540 The shadow mask sits just behind the grid of phosphors. 00:09:01.540 --> 00:09:07.200 The holes work in conjunction with three separate electron guns, one for each color of phosphor, 00:09:07.200 --> 00:09:08.960 in the neck of the picture tube. 00:09:08.960 --> 00:09:13.310 These guns are arranged in a triangular pattern, and their beams converge right at the shadow 00:09:13.310 --> 00:09:14.370 mask. 00:09:14.370 --> 00:09:18.700 The mask prevents the beams from landing on the wrong color, as the beam can only pass 00:09:18.700 --> 00:09:23.560 through the mask at a certain angle, thus ensuring there’s no accidental cross-over. 00:09:23.560 --> 00:09:28.470 This is why placing a magnet near the face of a color CRT makes such far-out patterns 00:09:28.470 --> 00:09:29.590 appear. 00:09:29.590 --> 00:09:37.120 The magnet bends the beam after the shadow mask, and thus the electron beams land where they shouldn't. 00:09:38.430 --> 00:09:41.440 Now comes the time to explain the thumbnail of this video. 00:09:41.440 --> 00:09:45.700 It’s very important to understand that the individual groupings of red, green, and blue 00:09:45.700 --> 00:09:48.540 phosphors are NOT pixels. 00:09:48.540 --> 00:09:53.481 The TV set isn’t even trying to line the beam up with these triads, if you will, and 00:09:53.481 --> 00:09:56.390 it has no way to address them individually. 00:09:56.390 --> 00:10:01.180 In fact, the pattern of triads in this picture tube doesn’t even form a grid, as each adjacent 00:10:01.180 --> 00:10:04.290 column of triads is shifted up half way. 00:10:04.290 --> 00:10:09.330 But that makes sense when you keep in mind that the electron guns are arranged in a triangle--logically 00:10:09.330 --> 00:10:12.250 the targets they aim to hit would be, too. 00:10:12.250 --> 00:10:14.970 And before you bring up Trinitron, I’ll be addressing that in another video. 00:10:14.970 --> 00:10:16.510 Hold your commenting horses. 00:10:16.510 --> 00:10:20.560 To form an image, the face of the tube is scanned in horizontal lines just like a black 00:10:20.560 --> 00:10:21.990 and white television set. 00:10:21.990 --> 00:10:26.170 It’s these lines that make up the image, not the dots on the screen. 00:10:26.170 --> 00:10:32.020 This is the precise reason why I used a black and white TV in my video on how analog television works. 00:10:32.020 --> 00:10:35.620 The lines are obvious on a black and white set, but a color set makes them less so. 00:10:35.620 --> 00:10:39.830 It’s tempting to imagine these groupings as pixels, but in reality they are simply 00:10:39.830 --> 00:10:44.360 a regular pattern of dots which, when combined with the shadow mask, force each individual 00:10:44.360 --> 00:10:47.570 electron beam into its respective color. 00:10:47.570 --> 00:10:51.380 This is most easily demonstrated with white text on a black background, so let’s pull 00:10:51.380 --> 00:10:53.160 out the old PlayStation. 00:11:05.840 --> 00:11:10.030 The text here doesn’t fall nicely in line with the individual phosphor groupings. 00:11:10.030 --> 00:11:13.850 All around the edges of the text, the phosphor groupings are only partially lit. 00:11:13.850 --> 00:11:18.351 That’s because the scan line isn’t landing nicely within the center of the triads, and 00:11:18.351 --> 00:11:21.060 it just barely grazes the bottom of these triads. 00:11:21.060 --> 00:11:25.030 But it doesn’t matter, as the position of the shadow mask and phosphor triads is irrelevant 00:11:25.030 --> 00:11:28.170 to the scanning beam--the beam can land anywhere it wants. 00:11:28.170 --> 00:11:32.560 But the shadow mask will always prevent each individual color component of the beam from 00:11:32.560 --> 00:11:34.360 hitting the incorrect phosphor. 00:11:34.360 --> 00:11:39.000 The shadow mask worked really well, but it required very powerful electron beams. 00:11:39.000 --> 00:11:44.920 About 85 percent of the beam energy is lost just in the shadow mask, so only 15 percent 00:11:44.920 --> 00:11:46.940 gets through the tiny holes. 00:11:46.940 --> 00:11:51.230 Without really powerful electron guns, a dim image would result. 00:11:51.230 --> 00:11:55.850 Even with suitable electron guns, early color CRT displays were often less bright than their 00:11:55.850 --> 00:11:57.610 black and white counterparts. 00:11:57.610 --> 00:12:02.660 Nevertheless, it meant that the picture tube could on its own produce a full color image. 00:12:02.660 --> 00:12:06.810 The real challenge then was to find a way to transmit color television in a way that 00:12:06.810 --> 00:12:11.860 a color set could interpret but that would still work for existing black and white TVs. 00:12:11.860 --> 00:12:17.100 You could simultaneously transmit three separate monochrome images and assign one to the green 00:12:17.100 --> 00:12:21.130 electron gun, another to the red, and the final to the blue. 00:12:21.130 --> 00:12:24.500 But which one would you have the black and white television receive? 00:12:24.500 --> 00:12:28.990 Picking just one color would produce a very unnatural image on the black and white set. 00:12:28.990 --> 00:12:34.230 Also, this would triple the bandwidth needed, which wasn’t really gonna fly with the FCC. 00:12:34.230 --> 00:12:39.210 Stay tuned as in the next video, we’ll explore how RCA managed the seemingly impossible task 00:12:39.210 --> 00:12:45.880 of sending three times the data without needing more bandwidth--by hiding the color in plain sight. 00:12:46.000 --> 00:12:48.880 Thanks so much for watching, I really hope you enjoyed the video! 00:12:48.880 --> 00:12:52.160 If you’re new to this channel and like what you see, please consider subscribing. 00:12:52.160 --> 00:12:55.920 I’d also like to thank all of my Patreon supporters, with a special thanks to these 00:12:55.920 --> 00:12:57.930 people who get their name in lights! 00:12:57.930 --> 00:13:02.660 Your can support me and this channel by visiting the link on your screen or in the description, 00:13:02.660 --> 00:13:04.700 and I thank you for your consideration. 00:13:04.700 --> 00:13:06.180 I’ll see you next time!