1 00:00:00,000 --> 00:00:05,000 So we're right here in your home lab. Is that what it is? 2 00:00:05,000 --> 00:00:07,000 Slash laundry room. 3 00:00:07,000 --> 00:00:08,000 Yeah. 4 00:00:08,000 --> 00:00:11,000 Yeah, as we were switching offices, we just moved offices, 5 00:00:11,000 --> 00:00:18,000 I set up a little space to test the angular performance of our liquid crystal mode 6 00:00:18,000 --> 00:00:20,000 put back onto the OLPC screen. 7 00:00:20,000 --> 00:00:24,000 So this is a split screen with the usual liquid crystal mode, 8 00:00:24,000 --> 00:00:31,000 and we hacked some of the films and things in the OLPC laptop to improve its performance. 9 00:00:31,000 --> 00:00:37,000 And we basically more than doubled the throughput with that. 10 00:00:37,000 --> 00:00:41,000 So before it took about a watt to get to about 80 nits, 11 00:00:41,000 --> 00:00:45,000 and now with that one watt we're almost at 200 nits. 12 00:00:45,000 --> 00:00:47,000 Nits is a unit of brightness. 13 00:00:47,000 --> 00:00:53,000 Sorry, it's not the best name for a unit, but it refers to how bright something is. 14 00:00:53,000 --> 00:00:56,000 So we've more than doubled the brightness of the OLPC screen. 15 00:00:56,000 --> 00:01:04,000 So we've shown that to Nicholas and John Wadlington and a bunch of the other team that was here last week. 16 00:01:04,000 --> 00:01:06,000 And they totally wanted an X01.5, no? 17 00:01:06,000 --> 00:01:08,000 Maybe. They're figuring it out. 18 00:01:08,000 --> 00:01:13,000 As you know, one of them was in the hospital last week, so they're sort of working through it. 19 00:01:13,000 --> 00:01:17,000 But we're happy. I think we're going to be there in the middle of June to show them some more stuff 20 00:01:17,000 --> 00:01:21,000 to figure out what we do to support OLPC from Pixel 2. 21 00:01:22,000 --> 00:01:25,000 So what are you doing right there? Is that a laser? 22 00:01:25,000 --> 00:01:29,000 Yeah, that's a laser. I was just measuring the angular performance of it, 23 00:01:29,000 --> 00:01:34,000 and then I sort of got this really bright light here. 24 00:01:34,000 --> 00:01:37,000 I'm having trouble getting this through. 25 00:01:37,000 --> 00:01:43,000 This is like 2 million nits, but this is off right now, so I'm not so useful. 26 00:01:43,000 --> 00:01:46,000 I guess we can plug it in, but I should. 27 00:01:47,000 --> 00:01:51,000 I don't know. This is just a very simple lab. We've got a much better lab. 28 00:01:51,000 --> 00:01:54,000 We can invite you to our lab in California sometime. 29 00:01:54,000 --> 00:01:59,000 Or better, the lab of our manufacturing partner, which is unbelievable. 30 00:01:59,000 --> 00:02:04,000 But they let people in with machines that can record stuff? 31 00:02:04,000 --> 00:02:13,000 Not usually. I spent my Memorial Day weekend in one of the best LCD labs in the world, 32 00:02:13,000 --> 00:02:18,000 basically fine-tuning our screen. It was fantastic, and all dressed up in a bunny suit. 33 00:02:18,000 --> 00:02:22,000 Is it better than Google? They have better canteens? 34 00:02:22,000 --> 00:02:30,000 Google has pretty good food, but I think that there's better toys at some of the LCD manufacturing houses 35 00:02:30,000 --> 00:02:33,000 because they concentrate on hardware. 36 00:02:33,000 --> 00:02:41,000 They have a machine that just peels films off of huge LCD screens. 37 00:02:42,000 --> 00:02:45,000 Lots and lots of screens, just an example. 38 00:02:45,000 --> 00:02:49,000 I never knew there was such a thing as a peeling machine, but there is, and I've used it. 39 00:02:49,000 --> 00:02:54,000 What are you planning for XO2? 40 00:02:54,000 --> 00:03:00,000 XO2? I don't know. We are playing a support role to One Laptop per Child. 41 00:03:00,000 --> 00:03:04,000 We're trying to figure out how we can help them with what we're doing. 42 00:03:04,000 --> 00:03:12,000 The idea that we had is basically OLPC was not in the business of supplying components to the IT industry, 43 00:03:12,000 --> 00:03:18,000 but as the netbook revolution was happening, it occurred to me that the best way I could help OLPC 44 00:03:18,000 --> 00:03:24,000 was by going out and making components to supply to the IT industry. 45 00:03:24,000 --> 00:03:28,000 Better screens that everybody could use, because if you can make more of something, 46 00:03:28,000 --> 00:03:30,000 you can make it cheaper and everybody can benefit. 47 00:03:30,000 --> 00:03:33,000 So not just the bottom of the pyramid, but the whole pyramid. 48 00:03:33,000 --> 00:03:39,000 The cool thing about the Pixel-T technology is poor kids in Africa got it first. 49 00:03:39,000 --> 00:03:41,000 It's a classic innovator's dilemma. 50 00:03:41,000 --> 00:03:45,000 Clayton Christensen, a professor at Harvard, wrote a book about a decade ago saying, 51 00:03:45,000 --> 00:03:51,000 really, it's amazing, the quality level of a technology increases over time 52 00:03:51,000 --> 00:03:56,000 to the point where people can't even tell, see, or perceive that quality level. 53 00:03:56,000 --> 00:04:02,000 And then a new technology comes along with theoretically lower quality, but new functionality. 54 00:04:02,000 --> 00:04:08,000 Like being able to see it outside, like higher resolution for reading, like what we did for OLPC. 55 00:04:08,000 --> 00:04:16,000 So the screens that we now have, these 3G screens, actually have totally gone beyond the OLPC technology. 56 00:04:16,000 --> 00:04:19,000 So there's none of the OLPC technology actually left in the screens. 57 00:04:19,000 --> 00:04:24,000 But nonetheless, that was sort of the first step. 58 00:04:24,000 --> 00:04:29,000 And it seems that now there's sort of this commercial-grade quality. 59 00:04:29,000 --> 00:04:35,000 People want them, and do you really need a screen with a million-to-one contrast ratio 60 00:04:35,000 --> 00:04:40,000 or 120% color saturation? 61 00:04:40,000 --> 00:04:43,000 Shouldn't 100% color saturation be enough? 62 00:04:43,000 --> 00:04:45,000 Things like that. 63 00:04:45,000 --> 00:04:48,000 Maybe there's this collective madness, especially in a laptop screen. 64 00:04:48,000 --> 00:04:52,000 You could even see for a large area HDTV screen. 65 00:04:52,000 --> 00:04:56,000 But what we've done as an industry is essentially taken these big screens 66 00:04:56,000 --> 00:05:00,000 and squished them back down and put them on our laptop and say, wow, you can watch movies on them. 67 00:05:00,000 --> 00:05:05,000 Well, that's great, and we can watch movies using the 3G screen too, as we showed. 68 00:05:05,000 --> 00:05:10,000 But what a lot of us do on our laptop screens is read. 69 00:05:10,000 --> 00:05:16,000 We read email, we respond to email, we read blogs, we read websites, we read. 70 00:05:16,000 --> 00:05:23,000 Why not make a screen that's also good for reading, that you can use indoors and out in any lighting condition? 71 00:05:23,000 --> 00:05:27,000 And so we focused on that, and that really started. 72 00:05:27,000 --> 00:05:30,000 It is, we hope, this innovator's dilemma. 73 00:05:30,000 --> 00:05:36,000 And do you think that the screen should be the main part of the laptop 74 00:05:36,000 --> 00:05:39,000 and perhaps the processor is getting a smaller role? 75 00:05:39,000 --> 00:05:41,000 Well, yeah, that's the thing. 76 00:05:41,000 --> 00:05:47,000 Do we really need a gazillion gigahertz and a gazillion gigs of RAM? 77 00:05:48,000 --> 00:05:54,000 Or is it okay just to have a simple machine where you can read and respond to your email, 78 00:05:54,000 --> 00:05:57,000 download videos, watch videos, go to the website? 79 00:05:57,000 --> 00:06:01,000 I mean, if you're crunching scientific calculations, that's one thing. 80 00:06:01,000 --> 00:06:06,000 But most of us want something that's light, that we can carry around, that's rugged, 81 00:06:06,000 --> 00:06:10,000 the batteries last a long time, we can see indoors and out and things like that. 82 00:06:10,000 --> 00:06:12,000 So we've focused on that. 83 00:06:12,000 --> 00:06:16,000 By not fighting the sun, you save a lot of power, right? 84 00:06:16,000 --> 00:06:22,000 And in addition, one of our goals is to try to get laptop and device makers 85 00:06:22,000 --> 00:06:27,000 to undertake the kind of architecture we undertook at One Laptop per Child, 86 00:06:27,000 --> 00:06:33,000 which was leaving the screen on and the Wi-Fi on with the motherboard off momentarily, 87 00:06:33,000 --> 00:06:37,000 so the sort of fast suspend-resume architecture. 88 00:06:37,000 --> 00:06:39,000 We know how to do that. 89 00:06:39,000 --> 00:06:43,000 We shipped about a million of them in very rough numbers from One Laptop per Child, 90 00:06:43,000 --> 00:06:46,000 but still getting the industry to sort of embrace that. 91 00:06:46,000 --> 00:06:48,000 Hopefully next year. 92 00:06:48,000 --> 00:06:54,000 I mean, this year, it's really, given the carnage of the economic crisis, 93 00:06:54,000 --> 00:06:58,000 some of the industry is a bit, understandably, risk-averse. 94 00:06:58,000 --> 00:07:01,000 And so we'll sort of be able to fold that in, we hope, next year. 95 00:07:01,000 --> 00:07:05,000 But for this year, at least we can get the screens into existing, 96 00:07:05,000 --> 00:07:12,000 essentially existing notebook architectures and get them to people this fall. 97 00:07:12,000 --> 00:07:14,000 So we thought speed was important. 98 00:07:14,000 --> 00:07:17,000 So to do, is that what you call DECON, the DECON process? 99 00:07:17,000 --> 00:07:20,000 Oh, it's similar to the DECON, yeah, where you've got some memory. 100 00:07:20,000 --> 00:07:23,000 Also, we're slowing down the liquid crystal mode, hopefully next year. 101 00:07:23,000 --> 00:07:26,000 There's a lot of work that shows that you can slow it down to 15 hertz. 102 00:07:26,000 --> 00:07:31,000 Even there were some papers at this SID conference showing 1 hertz. 103 00:07:31,000 --> 00:07:35,000 See, there's an issue usually with the lifetime of the liquid crystal molecules, 104 00:07:35,000 --> 00:07:37,000 if you don't flip them. 105 00:07:37,000 --> 00:07:42,000 One of the problems with an LCD molecule is, I'm looking for a, sorry, 106 00:07:42,000 --> 00:07:44,000 this is a piece of normal PC thing. 107 00:07:44,000 --> 00:07:47,000 One is a slight anode, the other is a slight cathode. 108 00:07:47,000 --> 00:07:49,000 And so they reorient in an applied field. 109 00:07:49,000 --> 00:07:53,000 But they need a constant AC charge across them. 110 00:07:53,000 --> 00:07:59,000 And if you don't do that, basically little bits of the liquid crystal molecule 111 00:07:59,000 --> 00:08:04,000 fall off into the alignment layers, and the liquid crystal becomes liquid, 112 00:08:04,000 --> 00:08:07,000 a much stronger liquid crystal, and it goes black, 113 00:08:07,000 --> 00:08:10,000 if you don't flip them quickly enough over time. 114 00:08:10,000 --> 00:08:16,000 And so that's why most liquid crystal displays are updated at 30 or 60 hertz, 115 00:08:16,000 --> 00:08:18,000 which we can do with ours. 116 00:08:18,000 --> 00:08:22,000 But we're working on research to lower that to 15 hertz and even lower. 117 00:08:22,000 --> 00:08:25,000 And there's some research suggesting much lower. 118 00:08:25,000 --> 00:08:29,000 Indeed, there are so-called bistable liquid crystal modes right now 119 00:08:29,000 --> 00:08:32,000 that is also something in our roadmap over time. 120 00:08:32,000 --> 00:08:36,000 So if you lower the hertz by half, it saves the battery half? 121 00:08:36,000 --> 00:08:37,000 That's right. 122 00:08:37,000 --> 00:08:42,000 And if you lower the voltage by half, it saves four times the energy 123 00:08:42,000 --> 00:08:50,000 because power is voltage, sorry, power is capacitance times frequency 124 00:08:50,000 --> 00:08:52,000 times voltage squared. 125 00:08:52,000 --> 00:08:57,000 And so if you lower the thing in half, you save four times the power. 126 00:08:57,000 --> 00:09:01,000 So to do the decon, you have to talk with Intel and Arm and all these people? 127 00:09:01,000 --> 00:09:04,000 Or is it other kind of people, the screen people who need to get that working, 128 00:09:04,000 --> 00:09:06,000 the motherboard people? 129 00:09:06,000 --> 00:09:12,000 It's a trivial thing to do in the timing controller, relatively, 130 00:09:12,000 --> 00:09:16,000 so there's mosquitoes sometimes. 131 00:09:16,000 --> 00:09:24,000 To do that, it's really working primarily with the ODM to do the architecture. 132 00:09:24,000 --> 00:09:32,000 That said, various CPU makers are working on alternative low-power architectures 133 00:09:32,000 --> 00:09:35,000 for their CPUs that maybe do the same thing inside of them. 134 00:09:35,000 --> 00:09:37,000 I must go with you. 135 00:09:37,000 --> 00:09:38,000 John, you can't leave. 136 00:09:38,000 --> 00:09:39,000 We have a meeting. 137 00:09:39,000 --> 00:09:41,000 We're going. 138 00:09:41,000 --> 00:09:45,000 Last thing, so you're shipping, the first batch will be without all these decon things, right? 139 00:09:45,000 --> 00:09:46,000 That's right. 140 00:09:46,000 --> 00:09:49,000 So the first batch is a run-in change, and by that, basically, 141 00:09:49,000 --> 00:09:53,000 we're asking the manufacturers to plug in our screen instead of their standard screen 142 00:09:53,000 --> 00:09:55,000 to existing electronics. 143 00:09:55,000 --> 00:09:59,000 We'd like them to mod the electronics to move the power consumption much lower, 144 00:09:59,000 --> 00:10:01,000 but for the first batch this year. 145 00:10:01,000 --> 00:10:07,000 And so next year, we believe that we can really get the huge battery life increases. 146 00:10:07,000 --> 00:10:10,000 And for this year, you can save two, three watts. 147 00:10:10,000 --> 00:10:14,000 Because for this batch, they have nine-cell batteries for Intel processors, 148 00:10:14,000 --> 00:10:16,000 and there's Arm processors that have eight hours. 149 00:10:16,000 --> 00:10:20,000 So for now, it might be okay until we do the decon that will improve much more next year. 150 00:10:20,000 --> 00:10:22,000 John, you really can't leave. 151 00:10:22,000 --> 00:10:23,000 Okay, sorry.