PodSearch

Hypercritical

94: Shrink, Shrink, Shrink!

 

00:00:00   you can't thread this thing through [TS]

00:00:01   until the needle comes down but you got [TS]

00:00:02   to make sure the needle is all the way [TS]

00:00:04   down before you grab it then you what [TS]

00:00:05   you have to start and stop a piece of [TS]

00:00:06   machinery from moving it's there is a [TS]

00:00:09   limit how fast a sewing machine can [TS]

00:00:11   operate this is hypercritical weekly [TS]

00:00:16   talk show ruminating on exactly what's [TS]

00:00:18   wrong we'll have apple related [TS]

00:00:20   technologies and businesses nothing is [TS]

00:00:21   so perfect that it can't be complained [TS]

00:00:24   about by my friend and yours John [TS]

00:00:25   siracusa I'm Dan Benjamin today is [TS]

00:00:28   Friday November 16 2012 this is episode [TS]

00:00:31   number 94 of our beloved hypercritical [TS]

00:00:33   show we would like to say thank you very [TS]

00:00:36   much to our sponsors today squarespace [TS]

00:00:38   com hover com and lynda.com who are [TS]

00:00:42   collaborating to make this show possible [TS]

00:00:43   we also want to thank mutual mobile [TS]

00:00:46   these guys are super passionate about [TS]

00:00:48   iOS as a platform they have a large [TS]

00:00:49   dedicated team all these guys do is [TS]

00:00:52   conceived deliver and support mobile [TS]

00:00:54   solutions for global brands and you know [TS]

00:00:55   what they want some help they want some [TS]

00:00:57   help they trust your engineers freedom [TS]

00:01:00   to make technical decisions by hiring [TS]

00:01:01   smart talented people to their team and [TS]

00:01:03   that's you they want you you need to [TS]

00:01:06   come in and once you do iOS work with [TS]

00:01:08   them they're a great company they [TS]

00:01:11   support the iOS community they do [TS]

00:01:13   meetups they do hackathons to do all the [TS]

00:01:15   cool stuff and just by going and looking [TS]

00:01:18   at this job you can support this show [TS]

00:01:20   careers dot mutual mobile com such [TS]

00:01:24   five-by-five go there to learn more and [TS]

00:01:25   of course we'd be irresponsible and [TS]

00:01:28   potentially liable if we did not mention [TS]

00:01:31   the bandwidth for November's brought to [TS]

00:01:32   you by mind node mind mapping app for [TS]

00:01:35   mac and iOS and brainstorm your project [TS]

00:01:37   you can organize your life you plan your [TS]

00:01:38   vacation don't matter they late collect [TS]

00:01:40   structure expand your ideas they have [TS]

00:01:42   icloud sharing keep your mind maps with [TS]

00:01:44   you very very cool app beloved mind node [TS]

00:01:47   dot-com how are you doing today John [TS]

00:01:51   siracusa I'm just fine you still the [TS]

00:01:54   same end time we got to be out of here [TS]

00:01:56   by noon well what time is our thing over [TS]

00:01:59   to Evernote 1145 and we need to prep for [TS]

00:02:02   it a little bit and I'd say fifteen [TS]

00:02:04   minutes to get over there yeah I mean if [TS]

00:02:05   we go if we go 15 minutes past it's fine [TS]

00:02:07   so 1215 all right well we gotta get [TS]

00:02:09   started then let's do it boom yeah we [TS]

00:02:13   got [TS]

00:02:14   one major topic of follow-up and then [TS]

00:02:15   one main topic for the show so maybe [TS]

00:02:18   we'll fit it in we'll see first let's [TS]

00:02:20   start with some little tiny follow-up [TS]

00:02:22   first one is from one canepa CA and the [TS]

00:02:28   EPA okay I did on that while back we [TS]

00:02:31   talked about the microsoft surface and [TS]

00:02:33   how much free space was available on it [TS]

00:02:35   right up because the OS and bail office [TS]

00:02:38   took up a lot of room yes and they don't [TS]

00:02:41   have a 16 gig model daily up 32 and the [TS]

00:02:44   last show I think I mentioned that one [TS]

00:02:46   of the big outs that you get with [TS]

00:02:47   Microsoft services that it's got or the [TS]

00:02:49   surface RT anyways you got a place to [TS]

00:02:51   plug in an SD card and SD cards are [TS]

00:02:53   cheap by third-party SD card shoved in [TS]

00:02:55   there by 64 gig SD card and suddenly [TS]

00:02:58   your surface has tons of space well one [TS]

00:03:00   writes in to tell me and the world that [TS]

00:03:04   that's all well and good but apparently [TS]

00:03:06   you can't install applications on the SD [TS]

00:03:09   card so you can put photos and videos [TS]

00:03:11   and things that are actually big but the [TS]

00:03:13   applications have to be on the main [TS]

00:03:14   system I don't think that's a big [TS]

00:03:16   limiting factor but it does explain how [TS]

00:03:18   they manage to deal with like what [TS]

00:03:20   happens if I yank up the SD card do a [TS]

00:03:21   bunch of my apps disappear well no [TS]

00:03:23   because you can't put any apps on uh-huh [TS]

00:03:25   so that's interesting compromise and i [TS]

00:03:28   believe is only applies to the RT when [TS]

00:03:30   the surface pro comes that I think you [TS]

00:03:31   will be able to put apps there but we'll [TS]

00:03:32   see alright another little tidbit from [TS]

00:03:37   Ali Hawkins last week he talked about [TS]

00:03:40   putting music behind ads you actually [TS]

00:03:44   talked about this on build and analyzed [TS]

00:03:45   to how you had put the music track [TS]

00:03:48   behind that ad read you did when Marco [TS]

00:03:50   had the ad on Howard Stern Show and I [TS]

00:03:52   asked you what the name of that thing [TS]

00:03:54   was called you didn't quite know a [TS]

00:03:55   couple people wrote in including Ali to [TS]

00:03:58   say that that's called a bed apparently [TS]

00:04:00   yeah I that's just something that I [TS]

00:04:04   should have been able to respond to I've [TS]

00:04:06   heard it referred to as a couple [TS]

00:04:07   different things bed is the correct term [TS]

00:04:09   music bed if you want to get specific [TS]

00:04:11   about it there you go but there you go [TS]

00:04:15   now the world knows all right now on to [TS]

00:04:18   the main topic of follow-up which [TS]

00:04:19   believe it or not actually shouldn't be [TS]

00:04:21   surprising as one of the small things we [TS]

00:04:23   talked about last week was a voting [TS]

00:04:25   technology [TS]

00:04:27   this is not a political show John that's [TS]

00:04:30   right I was happy to see that we didn't [TS]

00:04:32   get any feedback as far as I saw from [TS]

00:04:35   people who are angry about politics [TS]

00:04:37   because we didn't talk about politics we [TS]

00:04:40   just talked about voting technology but [TS]

00:04:42   there were lots of people who are not [TS]

00:04:43   angry but engaged on the topic of voting [TS]

00:04:49   technology I'm going to read a couple of [TS]

00:04:52   snippets of feedback here and you tell [TS]

00:04:54   me if you would notice a theme mostly [TS]

00:04:56   because I cherry picked these and it [TS]

00:04:57   makes it look like this a theme but now [TS]

00:04:59   this really was the theme of a lot of [TS]

00:05:01   feedback to the first is from andreas [TS]

00:05:03   hartl he says paper-based method without [TS]

00:05:07   anything more high-tech than a ballpoint [TS]

00:05:08   pen does work vote counting can be [TS]

00:05:10   verified by anybody without [TS]

00:05:11   sophisticated technical or mathematical [TS]

00:05:13   skills it's only downside is that you [TS]

00:05:15   have to wait a little bit longer for the [TS]

00:05:16   result and he gave me a link to the 25th [TS]

00:05:19   chaos communication congress website [TS]

00:05:22   where they had a paper and a [TS]

00:05:24   presentation on a voting technology i [TS]

00:05:26   put them both in the show notes so you [TS]

00:05:27   can read the paper and also watch the [TS]

00:05:29   video which is kind of long and boring [TS]

00:05:31   but it's all there and the the paper is [TS]

00:05:34   avoiding after any da p which is a [TS]

00:05:38   german voting system and digital pin [TS]

00:05:40   which is another kind of electronic [TS]

00:05:42   voting and subtitle is why cryptography [TS]

00:05:44   might not fix the issue of transparent [TS]

00:05:46   elections and is a little bit from the [TS]

00:05:47   abstract these methods introduce a level [TS]

00:05:50   of complexity into elections which [TS]

00:05:52   prevents most voters from understanding [TS]

00:05:53   the election process and its [TS]

00:05:54   verification where elections are [TS]

00:05:57   currently controlled by the people trust [TS]

00:05:58   in the ability of experts is required [TS]

00:06:00   when cryptographic methods are [TS]

00:06:02   introduced the next bit of feedback is [TS]

00:06:05   from Leonard if I'm pronouncing isn't [TS]

00:06:07   incorrectly because it's felt a little [TS]

00:06:09   bit differently there's one thing that [TS]

00:06:11   paper voting system has going for that [TS]

00:06:13   no computer system will ever get it's [TS]

00:06:15   entirely comprehensible observable and [TS]

00:06:17   provable by the simplest educated [TS]

00:06:19   persons this is essential for trust in [TS]

00:06:21   the voting process what are you saying [TS]

00:06:23   is that that anybody with just the basic [TS]

00:06:25   understanding of making a mark on a [TS]

00:06:28   paper in a certain place can verify this [TS]

00:06:31   right versus the electronic and [TS]

00:06:33   cryptographic ones require master's [TS]

00:06:35   level computer science degree of you'll [TS]

00:06:37   figure it out all right he says contrast [TS]

00:06:38   this to making a cross in a circle [TS]

00:06:39   putting it in a box and they [TS]

00:06:41   counting the votes and two piles it's so [TS]

00:06:42   simple that even a six-year-old can get [TS]

00:06:44   it a six-year-old could observe me doing [TS]

00:06:45   it and testified cheatham a six-year-old [TS]

00:06:47   can verify my result he does not need to [TS]

00:06:49   trust me and he said any computer system [TS]

00:06:53   has the problem that is basically magic [TS]

00:06:55   to normal people he also says they have [TS]

00:06:58   a word in German about this but I'm not [TS]

00:06:59   going to try to pronounce it I'll come [TS]

00:07:00   on no come on out but apparently is it [TS]

00:07:04   means roughly that a ruling group has [TS]

00:07:07   now it has a knowledge advantage that [TS]

00:07:08   they could use to control the uneducated [TS]

00:07:10   let's see next one from john berg mayor [TS]

00:07:15   any system needs to be capable of paper [TS]

00:07:18   recounts a lot of the theoretically very [TS]

00:07:21   secure systems you mentioned doing do [TS]

00:07:22   you have audible papal trails but the [TS]

00:07:24   point is this but this point is so [TS]

00:07:26   amazingly important that many of the [TS]

00:07:27   tech nerds talk about voting misted it [TS]

00:07:28   needs to be possible to conduct a [TS]

00:07:30   complete hand recount without any [TS]

00:07:32   computers without electricity even that [TS]

00:07:34   no system ever to the end of time no [TS]

00:07:36   matter what technology is ever invented [TS]

00:07:38   that does not have the capability the [TS]

00:07:40   capability doing a recount that can be [TS]

00:07:41   conducted by candlelight without [TS]

00:07:43   computers is a suitable choice next one [TS]

00:07:48   is from Mike Musel I talks about what [TS]

00:07:51   they have in Germany is that in Germany [TS]

00:07:52   in 2009 our highest court ruled that [TS]

00:07:54   voting computers we had up until then [TS]

00:07:55   did violate the principles of our [TS]

00:07:57   constitutions elections have certain [TS]

00:08:00   requirements that must be fulfilled in [TS]

00:08:01   order to be called democratic the [TS]

00:08:02   manganese is the need for the election [TS]

00:08:04   to be transparent to the voters and [TS]

00:08:05   verifiable with our good old crosses on [TS]

00:08:08   paper ballots even a ninety-year-old [TS]

00:08:09   non-tech lady can stay at the polling [TS]

00:08:11   station after the election is closed [TS]

00:08:12   watch the ballots to be correctly [TS]

00:08:14   counted and note the sum of the votes [TS]

00:08:16   for each party even if you had a [TS]

00:08:18   cryptographically totally secure [TS]

00:08:20   voting's computer setup how could that [TS]

00:08:21   nine-year-old non-tech lady know that [TS]

00:08:23   everything is correct she can't she [TS]

00:08:24   would have to trust the tech people how [TS]

00:08:26   can she know that those people one [TS]

00:08:27   hundred percent trustworthy and [TS]

00:08:28   competent again she can't there's no way [TS]

00:08:30   let's see if there's anything more on [TS]

00:08:35   this one I just about up anyone can go [TS]

00:08:38   and count the votes she has a good point [TS]

00:08:42   about smartphone voting remotely or [TS]

00:08:45   whatever it says even for technically [TS]

00:08:46   feasible it would pose a risk that for [TS]

00:08:48   example husband trusts the control house [TS]

00:08:49   wife boat pilot the security print [TS]

00:08:52   principal Democrat elections [TS]

00:08:53   if you're not doing it in a controlled [TS]

00:08:55   pulling place anonymously people who try [TS]

00:08:58   to control across your boat so we did [TS]

00:08:59   remote voting Omri our Beave rights in [TS]

00:09:05   with a link to NPR on the media podcast [TS]

00:09:10   segment about online voting and it's a [TS]

00:09:14   short segment like five minutes long and [TS]

00:09:15   it's about how is the country of Estonia [TS]

00:09:17   has been has been doing online voting [TS]

00:09:20   since 2005 they have government issued [TS]

00:09:22   identification cards with the password [TS]

00:09:23   and that's the only real security [TS]

00:09:25   measure taken when they just basically [TS]

00:09:26   you know over the internet do their [TS]

00:09:27   votes so you want to hear how that's [TS]

00:09:30   working out in a country that's actually [TS]

00:09:31   doing it and the pros and cons of that [TS]

00:09:33   approach take a look at that also in the [TS]

00:09:37   show notes so show notes five by five [TS]

00:09:39   dot TV such hypercritical / 94 yes many [TS]

00:09:43   many links we've already just burned [TS]

00:09:44   through in this quick follow-up so the [TS]

00:09:46   theme here as you can see is the idea [TS]

00:09:47   that you need to have a you need to have [TS]

00:09:51   a way to both understand like you know [TS]

00:09:55   non-technical birth needs to be able to [TS]

00:09:57   understand what's going on the voting [TS]

00:09:58   system and also the boring system still [TS]

00:10:01   needs to work ah by candlelight with no [TS]

00:10:04   electricity hmm now the first thing imma [TS]

00:10:07   say on this topic is that a secure you [TS]

00:10:10   know and inaudible electronic voting [TS]

00:10:12   system doesn't preclude any of these [TS]

00:10:14   things in fact some of these systems [TS]

00:10:16   that were linked in that in the past [TS]

00:10:17   shows show notes showed that yes the you [TS]

00:10:20   know just because you use cryptography [TS]

00:10:21   and electronic things to make the thing [TS]

00:10:24   auditable and secure and all that stuff [TS]

00:10:26   doesn't mean that you don't also have [TS]

00:10:28   okay so say the power was out no one has [TS]

00:10:30   any computers because it's like the [TS]

00:10:31   terrible premise of the show revolution [TS]

00:10:33   you still end up with a bunch of paper [TS]

00:10:36   artifacts that you could then hand count [TS]

00:10:38   manually right so it's not you know [TS]

00:10:42   various voting systems passed or failed [TS]

00:10:44   this thing but there's nothing inherent [TS]

00:10:45   about electronic voting or [TS]

00:10:46   cryptographically secure voting that [TS]

00:10:48   doesn't say also you have you know it [TS]

00:10:50   doesn't preclude you having this [TS]

00:10:51   auditable paper trail right ah so most [TS]

00:10:56   of these feedback you know acknowledge [TS]

00:10:58   that but the other thing is to said you [TS]

00:10:59   know all right so you can make it secure [TS]

00:11:00   an electronic or whatever better you [TS]

00:11:01   also have to sort of degrade gracefully [TS]

00:11:02   there has to be a paper trail an [TS]

00:11:05   artifact in the ability to do a Henry [TS]

00:11:07   count but that aside I don't think any [TS]

00:11:10   of these requirements you know eliminate [TS]

00:11:12   electronic voting another thing in [TS]

00:11:13   Germany Bailey had to eliminate the [TS]

00:11:14   cruddy electronic voting machines they [TS]

00:11:15   had and by the way several people from [TS]

00:11:17   different countries wrote in to say that [TS]

00:11:18   they have similar problems the US where [TS]

00:11:20   when electronic voting machines were [TS]

00:11:21   rolled out they were worse in all [TS]

00:11:23   possible ways than the things that they [TS]

00:11:25   were replacing we have the same problem [TS]

00:11:26   the United States pointed last showbiz [TS]

00:11:28   that that doesn't need to be the case [TS]

00:11:29   that we have the smarts to make this [TS]

00:11:30   better but it is important to recognize [TS]

00:11:32   it just making electronic usually seems [TS]

00:11:34   to go terribly wrong because of just you [TS]

00:11:36   know corruption and competence and all [TS]

00:11:38   the usual things that happen in [TS]

00:11:40   government but I want to think about [TS]

00:11:43   this idea that the the election both [TS]

00:11:49   these idea is that that an election [TS]

00:11:51   system needs to work by candlelight and [TS]

00:11:53   the people need to understand so first [TS]

00:11:57   the candlelight one people I understand [TS]

00:12:00   people fearing comforted by the idea [TS]

00:12:03   that if a giant EMP goes often destroys [TS]

00:12:06   all electricity producing things or [TS]

00:12:08   whatever crazy scenario you want to come [TS]

00:12:09   out with that we can you know you a [TS]

00:12:12   bunch of old ladies could do the recount [TS]

00:12:14   like you don't need to understand [TS]

00:12:15   anything you just need to look at pieces [TS]

00:12:17   of paper and put them into piles and [TS]

00:12:18   then count the things in the pile and [TS]

00:12:20   you have 20 different people do this [TS]

00:12:21   it's completely comprehensible the [TS]

00:12:24   comfort of that is obvious like you [TS]

00:12:27   don't like something that you don't [TS]

00:12:29   understand but the comfort of that I [TS]

00:12:32   think is mostly theoretical because [TS]

00:12:35   there's no way that any single person is [TS]

00:12:38   going to be able to hand count all the [TS]

00:12:40   votes for any country or a significant [TS]

00:12:42   size all right a group of people can [TS]

00:12:44   hand count them in a distributed manner [TS]

00:12:46   and then they add up their sums and you [TS]

00:12:48   have people repeated or whatever but in [TS]

00:12:49   general it you know you can never have [TS]

00:12:53   every single person in the country get [TS]

00:12:55   their hands on all the paper ballots [TS]

00:12:56   like say you took all the paper ballots [TS]

00:12:58   and the country the size of the US and [TS]

00:12:59   each person the entire country got a [TS]

00:13:01   chance to hand count them to verify for [TS]

00:13:02   themselves well that's just obviously [TS]

00:13:03   not going to work right so in all cases [TS]

00:13:06   you're trusting that some group of [TS]

00:13:08   people that is not you is trustworthy in [TS]

00:13:10   some way or at least it there that their [TS]

00:13:12   biases are counter to each other and [TS]

00:13:14   balance each other out or whatever [TS]

00:13:15   but people like to just think that you [TS]

00:13:18   know okay well I'm not going to hang out [TS]

00:13:20   the ballots and no 10 is going to Hank [TS]

00:13:22   out the balance but I do though that 17 [TS]

00:13:24   people hand counted the ballots and half [TS]

00:13:26   of them were for party a and half of [TS]

00:13:28   them from a party B and if they keep [TS]

00:13:30   coming up with the same numbers i trust [TS]

00:13:33   those number right I don't quite see the [TS]

00:13:37   connection between that distributed [TS]

00:13:39   trust where you didn't do it your friend [TS]

00:13:41   didn't do it but eventually there's some [TS]

00:13:42   group of people who actually had [TS]

00:13:43   encountered the things and anyone can [TS]

00:13:44   volunteer to count them liking all [TS]

00:13:45   countries like Jeremy and stuff anyone [TS]

00:13:46   can go down there and count them and [TS]

00:13:47   satisfy themselves but everybody can't [TS]

00:13:49   go down account so there is some remote [TS]

00:13:51   trust and the idea that okay but I [TS]

00:13:55   understand what they're doing they're [TS]

00:13:56   just looking at pieces of paper and [TS]

00:13:58   counting them sorting them into piles I [TS]

00:13:59   understand that process I don't think [TS]

00:14:02   those things are as connected as people [TS]

00:14:04   seem to think they are ah that because [TS]

00:14:08   you understand the system therefore this [TS]

00:14:10   indirect trust is more trustworthy right [TS]

00:14:14   because indirect trust for set work for [TS]

00:14:17   when you're saying I don't you know [TS]

00:14:19   these mathematicians assure me that it's [TS]

00:14:20   correct but I don't know these [TS]

00:14:21   mathematicians right that indirect dress [TS]

00:14:23   is no good but the indirect trust of [TS]

00:14:25   seven people away from you a bunch of [TS]

00:14:27   people counted a piece of paper that's [TS]

00:14:28   better because you're like okay not only [TS]

00:14:29   am I trusting these people who are not [TS]

00:14:32   me but I understand what they're doing [TS]

00:14:34   therefore I feel like I'm less likely to [TS]

00:14:35   be cheated my argument would be that [TS]

00:14:38   it's much easier to like if those people [TS]

00:14:42   were cheating or there was some sort of [TS]

00:14:43   organized cheating conspiracy there [TS]

00:14:45   there is nothing that anyone else can do [TS]

00:14:48   to verify that they're not cheating [TS]

00:14:51   except for count them themselves whereas [TS]

00:14:53   and this is I think you know this is may [TS]

00:14:54   come back to people not understanding [TS]

00:14:55   the the mathematical details of [TS]

00:14:58   cryptography and stuff like that or I [TS]

00:14:59   mean I don't understand the mathematical [TS]

00:15:00   details but just like the consequences [TS]

00:15:01   of it if you can imagine a system where [TS]

00:15:06   uh you add up a bunch of numbers and you [TS]

00:15:10   come up with a total like 12 or 13 right [TS]

00:15:12   and what you want to know is did my vote [TS]

00:15:15   contribute to that number 12 it seems [TS]

00:15:17   impossible because you just thought up [TS]

00:15:18   with the number 12 you have no idea if [TS]

00:15:20   your vote was part of that 12 right how [TS]

00:15:22   can how can you ever know that you know [TS]

00:15:24   it was just 12 was made up of a bunch of [TS]

00:15:26   votes maybe one of those was yours maybe [TS]

00:15:27   it wasn't maybe your [TS]

00:15:28   was counted maybe it wasn't you just [TS]

00:15:30   know number 12 will imagine the system [TS]

00:15:31   where you could come up with a total [TS]

00:15:33   number of votes and you could be [TS]

00:15:36   mathematically assured that your vote [TS]

00:15:38   contributed to that total this is the [TS]

00:15:41   getting into the magic thing it's like [TS]

00:15:42   what do you mean that makes no sense you [TS]

00:15:43   can't tell if your number contribute at [TS]

00:15:44   all that's impossible well there are [TS]

00:15:46   things that you can do mathematically to [TS]

00:15:49   make that so that you can confirm that [TS]

00:15:51   your vote really did contribute to the [TS]

00:15:53   total and that if your vote wasn't [TS]

00:15:55   counted towards a total in the correct [TS]

00:15:57   manner the total would not validate in a [TS]

00:15:59   particular way uh this is going to help [TS]

00:16:01   anybody because this is even more [TS]

00:16:02   complicated but if you ever go read up [TS]

00:16:03   on Bitcoin which is this distributed [TS]

00:16:06   fiat currency / scam / whatever you know [TS]

00:16:11   you know I don't know how you want to [TS]

00:16:13   talk about wait why are you calling a [TS]

00:16:14   scam that's not a scam but like it's [TS]

00:16:16   it's an economic thing where it's kind [TS]

00:16:20   of like gambling but not really but the [TS]

00:16:24   currency and the imaginary nature of [TS]

00:16:26   money is a topic that's beyond the scope [TS]

00:16:28   of this podcast we're talking right now [TS]

00:16:30   but the one thing Bitcoin does have [TS]

00:16:32   going for it is that it has the system [TS]

00:16:33   whereby everybody who's part of the [TS]

00:16:36   Bitcoin system can tell whether a [TS]

00:16:39   transaction is valid or not like you [TS]

00:16:42   can't sneak in an invalid transaction [TS]

00:16:45   that you know puts more money into one [TS]

00:16:47   account that it took then it took out of [TS]

00:16:48   the other or whatever because this is [TS]

00:16:50   consensus type of algorithm and those [TS]

00:16:52   things will be invalidated and it's [TS]

00:16:53   basically possible to tell whether the [TS]

00:16:56   money that you put it in correctly [TS]

00:16:58   transferred to over there ah [TS]

00:16:59   mathematically and so that's that you [TS]

00:17:01   know that ignores all the whether it's a [TS]

00:17:03   good idea to get into Bitcoin and then [TS]

00:17:04   how much it's valued and stuff like that [TS]

00:17:06   just simply the foundation of being able [TS]

00:17:08   to tell that these inputs produce this [TS]

00:17:11   output in this particular way and that [TS]

00:17:13   everybody in the system has to agree on [TS]

00:17:14   it for the system to move forward agree [TS]

00:17:17   sounds like it's something anyway it's [TS]

00:17:18   it's basically a much more complicated [TS]

00:17:22   version of the thing that I was saying [TS]

00:17:23   where if you have a total number you [TS]

00:17:25   want to know that your vote counted [TS]

00:17:26   towards it and that's something you can [TS]

00:17:28   do with math there's always going to be [TS]

00:17:30   a secret somewhere of some secret key or [TS]

00:17:31   some other things being corrupted or [TS]

00:17:33   stolen but trusting that there's a [TS]

00:17:37   mathematical foundation to proving [TS]

00:17:39   whether something happened or not [TS]

00:17:41   I i think is better than trusting that [TS]

00:17:44   people you don't know and never going to [TS]

00:17:46   meet have hand counted these things [TS]

00:17:47   simply because you think they're you [TS]

00:17:49   know balancing each other off and if [TS]

00:17:50   they were cheating they would catch each [TS]

00:17:51   other whatever like there's nothing [TS]

00:17:52   there's nothing to hang your hat on [TS]

00:17:54   mathematically that you can use to to [TS]

00:17:57   prove that those guys did the right [TS]

00:17:58   thing you basically just have to trust [TS]

00:17:59   them whereas if you have a [TS]

00:18:00   cryptographically secure assistant [TS]

00:18:02   whether you understand it or not uh [TS]

00:18:03   there should be enough smart people in [TS]

00:18:06   the world assuring you on both sides of [TS]

00:18:09   the coin that like look if you do this [TS]

00:18:10   you'll be able to tell whether your vote [TS]

00:18:11   was counted and I do get the wrong ansel [TS]

00:18:13   your note you know your vote wasn't [TS]

00:18:14   counted and the things required cheat on [TS]

00:18:16   this you know would be you know that [TS]

00:18:20   they can lay out the exact parameters [TS]

00:18:21   there's no there's no spy craft you [TS]

00:18:24   could do to corrupt the system in a way [TS]

00:18:27   that wouldn't be detectable except for [TS]

00:18:29   you know stealing secret keys or [TS]

00:18:31   something uh now just because I I I have [TS]

00:18:36   more trust in mathematics combined with [TS]

00:18:39   this remote trucks than people doing the [TS]

00:18:40   other things like they'd rather have the [TS]

00:18:41   comfort doesn't mean that this is that [TS]

00:18:42   these people all don't have points that [TS]

00:18:44   this is all well and good and you can [TS]

00:18:46   try to like and put it on a blackboard [TS]

00:18:49   and explain to people how it works and [TS]

00:18:50   stuff but the bottom line is if they [TS]

00:18:51   don't trust it it's not good because you [TS]

00:18:53   know you want a system that everybody [TS]

00:18:55   trusts whether people trust it is not [TS]

00:18:58   necessarily connected to whether it is [TS]

00:18:59   more or less trustworthy so I think [TS]

00:19:02   their their point on it being able to be [TS]

00:19:05   counted by candle light stands not so [TS]

00:19:08   much on the merits of the technology but [TS]

00:19:11   on the fact that most of the world [TS]

00:19:12   probably agrees with them and in the end [TS]

00:19:14   this is a system that everyone has to [TS]

00:19:16   agree is fair for things to work well so [TS]

00:19:18   if that's if that is unchangeable and [TS]

00:19:21   you cannot convince the world that [TS]

00:19:23   you're secure voting system actually is [TS]

00:19:25   secure and you cannot convince them that [TS]

00:19:27   you can show them how to secure it and [TS]

00:19:28   that it's more secure than other [TS]

00:19:30   alternatives it doesn't matter if it's [TS]

00:19:31   really more secure not all it matters is [TS]

00:19:33   what they think of course that's the [TS]

00:19:36   ultimate election hack the fact that it [TS]

00:19:39   doesn't really matter what actually [TS]

00:19:40   happens in the election all that matters [TS]

00:19:41   is that everyone involved or almost [TS]

00:19:43   everyone involved agrees that it was [TS]

00:19:45   fair whether it was really fair not [TS]

00:19:47   almost immaterial if you were the best [TS]

00:19:49   way to cheat an election is to do it in [TS]

00:19:51   such a way that everyone agrees that the [TS]

00:19:52   result is fair [TS]

00:19:54   so that's kind of depressing but but [TS]

00:19:57   anyway I think their point stands like I [TS]

00:20:00   said I think any decent [TS]

00:20:02   cryptographically secure voting system [TS]

00:20:03   will also and must also include a paper [TS]

00:20:06   trail and all the things that to be able [TS]

00:20:08   to be hand countered and if the security [TS]

00:20:10   is just extra on top of it oh and I also [TS]

00:20:12   want to get back to one of the piece in [TS]

00:20:13   the feedback of like the only thing you [TS]

00:20:14   get from an electronic auditable roading [TS]

00:20:16   record is that it's uh it's faster [TS]

00:20:20   depart to tabulate the results that is [TS]

00:20:22   not one of the goals of these systems or [TS]

00:20:23   shouldn't be ready shortly speed it [TS]

00:20:25   should be accuracy or not just accuracy [TS]

00:20:27   but like it's an extra level [TS]

00:20:29   verification like you've got all the [TS]

00:20:30   same old verification with pieces of [TS]

00:20:32   paper and people can hand recount them [TS]

00:20:33   do a lot stuff but also this is extra [TS]

00:20:35   layer of security on top of that that [TS]

00:20:38   gives you that you didn't have before [TS]

00:20:40   like for example before you didn't have [TS]

00:20:41   the ability to like go to a website and [TS]

00:20:44   enter in some secret code that you got [TS]

00:20:46   in your little paper receipt and find [TS]

00:20:48   out whether your vote was counted or not [TS]

00:20:50   you may not understand that you may [TS]

00:20:52   think it's all a lie and it's totally [TS]

00:20:53   fabricated and not given you know any [TS]

00:20:56   credence whatsoever that's fine because [TS]

00:20:59   it's just on top of the existing thing [TS]

00:21:02   that you do trust but in general I think [TS]

00:21:04   you know that that's the case where your [TS]

00:21:06   your disbelief the sort of Luddite [TS]

00:21:09   disbelief that electronic scan cannot [TS]

00:21:10   possibly add in the extra security [TS]

00:21:12   doesn't mean that it doesn't actually [TS]

00:21:14   add extra security and make it more [TS]

00:21:15   difficult to to corrupt and stuff like [TS]

00:21:17   that so I think that's like a bonus type [TS]

00:21:19   system yeah I think I basically combine [TS]

00:21:24   the two whether whether you know people [TS]

00:21:26   understanding it and having the paper [TS]

00:21:27   trail or two separate things for one [TS]

00:21:29   they kind of meld into each other uh [TS]

00:21:30   yeah III don't want to call people [TS]

00:21:36   luddites because they don't trust [TS]

00:21:37   electronic voting because if you look at [TS]

00:21:38   the history of electronic voting you [TS]

00:21:40   know like everyone who wrote in [TS]

00:21:42   including you know everything's happened [TS]

00:21:43   in the US we've just done it so terribly [TS]

00:21:45   that how could it not have a bad [TS]

00:21:47   reputation I mean it's got a deserve a [TS]

00:21:49   terrible reputation and especially [TS]

00:21:51   people who don't understand it all they [TS]

00:21:52   know is every time someone tries to [TS]

00:21:54   introduce electronic voting it ends up [TS]

00:21:57   being a giant disaster therefore we [TS]

00:21:59   should never do anything electronically [TS]

00:22:00   and that's it I was also trying to think [TS]

00:22:02   of other things in the world that [TS]

00:22:05   average person doesn't understand but [TS]

00:22:07   the net [TS]

00:22:07   less people trust it's difficult to come [TS]

00:22:10   with straight through analogies but [TS]

00:22:12   there are plenty of them I the problem [TS]

00:22:14   is they're not exactly the same thing [TS]

00:22:15   like people understand how airplanes fly [TS]

00:22:17   but they have faith that they'll be in [TS]

00:22:18   the air and that you know something like [TS]

00:22:21   they have really have no idea like her [TS]

00:22:22   you know and people don't understand how [TS]

00:22:24   government works in general and how you [TS]

00:22:25   know just how the machinery of [TS]

00:22:27   government works in actuality but they [TS]

00:22:29   trust that all this law was passed and I [TS]

00:22:31   trust it was passed in a way that didn't [TS]

00:22:32   involve just a bunch of guys in the back [TS]

00:22:34   room chomping cigars and saying let's do [TS]

00:22:35   this you know there are many things in [TS]

00:22:38   the world that people don't understand [TS]

00:22:40   that nevertheless put their trust in so [TS]

00:22:41   I'm not of the opinion that it would be [TS]

00:22:44   impossible to get human beings to trust [TS]

00:22:46   an end an auditable cryptographically [TS]

00:22:50   secure electronic voting system I'm of [TS]

00:22:52   the opinion that it would be possible to [TS]

00:22:54   it maybe you have to wait for a couple [TS]

00:22:56   generations people die off and gradually [TS]

00:22:58   introduce it and actually do it right [TS]

00:22:59   for once and stuff like that but it [TS]

00:23:01   could happen and I think the people who [TS]

00:23:03   are against it would eventually die off [TS]

00:23:05   and everyone else would think it's crazy [TS]

00:23:07   to do it in any other way I don't think [TS]

00:23:08   voting is unlike every other part of the [TS]

00:23:11   human condition where it's like well [TS]

00:23:12   everything else we can advance [TS]

00:23:13   technology and but voting we have to do [TS]

00:23:15   with rocks until the end of my voting is [TS]

00:23:18   different in some fundamental way than [TS]

00:23:20   anything else human beings do right [TS]

00:23:21   again out it cannot be improved upon [TS]

00:23:23   with modern technology it must be the [TS]

00:23:25   same that it was you know right hundreds [TS]

00:23:27   and hundreds and hundreds of years ago [TS]

00:23:28   and there are many ways you can shoot [TS]

00:23:30   down any analogy flying a plane is not [TS]

00:23:31   like an election don't you understand [TS]

00:23:33   electron is how we govern ourselves it's [TS]

00:23:34   like meta system it's so important ball [TS]

00:23:36   blah but there is you know there's [TS]

00:23:37   nothing in in life that is so special [TS]

00:23:41   that technology cannot touch it ever and [TS]

00:23:43   it almost sounds like the intro TV show [TS]

00:23:45   no nothing is so perfect that it yeah I [TS]

00:23:49   don't know it's it's kind of similar [TS]

00:23:52   thing but anyway uh like I said I'm not [TS]

00:23:55   I'm not arguing against paper trails I [TS]

00:23:57   think they have to be there and I think [TS]

00:23:58   there's no reason they shouldn't be but [TS]

00:24:00   I am arguing for electronic voting [TS]

00:24:02   systems done right and done right means [TS]

00:24:05   you have all of the benefits of the [TS]

00:24:06   existing system plus many extra benefits [TS]

00:24:09   on top of that that people may not care [TS]

00:24:10   about or anything but as long as the [TS]

00:24:12   superset of the existing security [TS]

00:24:14   mechanisms then so what people can [TS]

00:24:16   pretend that's all magic and BS and it [TS]

00:24:18   is equally secure to the best you know [TS]

00:24:20   paper system with [TS]

00:24:21   the hand recounts in all its stuff like [TS]

00:24:23   all right that went a little long I [TS]

00:24:25   think I was rambling there but I think [TS]

00:24:26   it's alright sponsor sure hover com [TS]

00:24:29   simplified domain management you've [TS]

00:24:31   probably registered a domain from a [TS]

00:24:32   company that they tried upsell you on [TS]

00:24:34   stuff you go there you register domain [TS]

00:24:36   name and the next thing you know you've [TS]

00:24:37   signed up for it for five years and [TS]

00:24:38   you've bought all kinds of additional [TS]

00:24:41   you know back ordering things and stuff [TS]

00:24:43   you didn't want or even if you are smart [TS]

00:24:45   and savvy enough to disable all of those [TS]

00:24:48   things as you're going through the [TS]

00:24:49   checkout process they're still there and [TS]

00:24:52   there's still kind of a nuisance well [TS]

00:24:53   hover com does not do that these guys [TS]

00:24:56   their goal is to provide you with a [TS]

00:24:58   straightforward minimal clean honest [TS]

00:25:01   process that makes it easy to go there [TS]

00:25:03   register the domain that you want and go [TS]

00:25:05   away that's what they do there's a [TS]

00:25:08   search box you type in the domain name [TS]

00:25:10   you want if it's available you register [TS]

00:25:12   it if it's not available they'll show [TS]

00:25:13   you some alternatives and that's it you [TS]

00:25:16   get out there's a checkbox do you want [TS]

00:25:18   to have the free who is privacy I [TS]

00:25:20   believe that checked then you can go [TS]

00:25:22   into your settings and make it so that [TS]

00:25:24   there's like a checkbox in there that [TS]

00:25:25   says that never never contact me you can [TS]

00:25:28   do that if you want i don't mind them [TS]

00:25:30   contacting me because they have some [TS]

00:25:31   pretty cool stuff like their domain [TS]

00:25:33   transfer valet service where if your [TS]

00:25:35   transfer your domain from another [TS]

00:25:37   registrar they'll do the entire thing [TS]

00:25:38   like a human being that works there for [TS]

00:25:40   free will do the whole process for you [TS]

00:25:42   they have email hosting if you wanted [TS]

00:25:44   they have built-in free dns which is [TS]

00:25:47   what i'm pretty much using for almost [TS]

00:25:48   everything now go check these guys out [TS]

00:25:50   hover com / dan sent me this will get [TS]

00:25:53   you ten percent off everything that you [TS]

00:25:54   do or you can just use the coupon code [TS]

00:25:56   dan sent me when you're buying something [TS]

00:25:58   like if you're an existing customer if [TS]

00:26:00   you're already there and you're [TS]

00:26:01   transferring more domains there's no [TS]

00:26:02   limit on how many times you can use that [TS]

00:26:04   and i actually use that myself for my [TS]

00:26:07   domain name registration problem my [TS]

00:26:10   addiction and you can go there hover com [TS]

00:26:13   such dance let me get ten percent [TS]

00:26:14   hopefully you liked the experience as [TS]

00:26:16   much as I do thanks very much to them [TS]

00:26:18   for making the show possible what you [TS]

00:26:22   got next John and before we move on one [TS]

00:26:25   point that a lot of these bits of [TS]

00:26:29   feedback oh I got a turned off all right [TS]

00:26:34   a lot of these bits of feedback about [TS]

00:26:36   voting kind of matter of factly state [TS]

00:26:40   that the way they're voting system works [TS]

00:26:42   in their country the voting system do [TS]

00:26:44   they like usually it that involves [TS]

00:26:46   government-issued ID right and from [TS]

00:26:50   coming from a u.s. perspective the rest [TS]

00:26:51   of the world probably thinks we're crazy [TS]

00:26:52   but this just goes to show it like you [TS]

00:26:54   know and in the same way that I don't [TS]

00:26:56   think that the average person needs to [TS]

00:26:58   understand the math involved for a [TS]

00:27:00   cryptographic voting system to [TS]

00:27:03   eventually be considered secure people [TS]

00:27:07   and the rest of world don't think twice [TS]

00:27:09   about having government-issued ID and [TS]

00:27:13   having to show it at the polls to vote [TS]

00:27:15   whereas an America that is a very very [TS]

00:27:17   big issue and some of the same people is [TS]

00:27:21   that you know don't you understand it [TS]

00:27:23   has to be a paper ballot has to be hand [TS]

00:27:25   candid and of course you have your [TS]

00:27:26   government issue date I'm turn around [TS]

00:27:27   said don't you understand you can't [TS]

00:27:28   require government issued ID and say [TS]

00:27:30   well what how can you vote without a [TS]

00:27:31   government issue lady how do we even [TS]

00:27:32   know who's what it again you know [TS]

00:27:35   everyone has their personal hang-ups [TS]

00:27:37   culturally personally then not all of [TS]

00:27:42   them are based on logic so in the same [TS]

00:27:44   in the same way that I'm saying to these [TS]

00:27:46   people that owe you know you you know it [TS]

00:27:48   doesn't need to be done with rocks when [TS]

00:27:52   that they would say to us you know you [TS]

00:27:53   have to have a government ID to vote how [TS]

00:27:55   can you vote without that it's just [TS]

00:27:56   stupid and I can't talk about this too [TS]

00:27:58   much without getting into politics or US [TS]

00:28:00   politics we just go show that the rest [TS]

00:28:01   of the world is different like and I [TS]

00:28:03   don't think most of the feedback even [TS]

00:28:04   acknowledge that they were just like let [TS]

00:28:06   me explain how it works in the civilized [TS]

00:28:08   world we use rocks and pieces of paper [TS]

00:28:09   and a government-issued ID it's like [TS]

00:28:11   that's not gonna fly in the US that's no [TS]

00:28:13   ah people are strange when you're a [TS]

00:28:18   stranger is that it yes that is the [TS]

00:28:20   correct Jim arson doors lyrics okay all [TS]

00:28:24   right so let's move on to the main topic [TS]

00:28:25   today and this could be short how many [TS]

00:28:27   more sponsors do we have we have two [TS]

00:28:29   more so we'll space them out alright so [TS]

00:28:32   don't let me go too long alright gotta [TS]

00:28:34   finish this topic is as teasing the last [TS]

00:28:37   show it's about using ARM processors all [TS]

00:28:41   caps a RM instead of Intel processors on [TS]

00:28:44   max right and it is spawned by this [TS]

00:28:47   business we guard [TS]

00:28:48   from that a week ago titled Apple said [TS]

00:28:51   to be exploring switch from Intel for [TS]

00:28:53   Mac that is bad headline because it is [TS]

00:28:55   awkward right here's a little synopsis [TS]

00:28:58   thing Apple exploring ways to replace [TS]

00:29:00   Intel processors in its Mac personal [TS]

00:29:03   computers with a version of the chip [TS]

00:29:05   technology uses in the iphone or ipad [TS]

00:29:07   according to people familiar with the [TS]

00:29:09   company's research people familiar with [TS]

00:29:11   Apple engineers have grown confident [TS]

00:29:13   that the chip designs used for its [TS]

00:29:14   mobile devices will one day be powerful [TS]

00:29:17   enough to run its desktops and laptops [TS]

00:29:18   said three people knowledgeable of with [TS]

00:29:22   knowledge of the work drank who has to [TS]

00:29:24   remain anonymous because the plans are [TS]

00:29:26   confidential alright so typically Fuzzle [TS]

00:29:29   esource thing and mostly gonna use as [TS]

00:29:31   articles a jumping-off point for a [TS]

00:29:33   discussion about this idea and i'm going [TS]

00:29:38   to give sort of a cliff notes background [TS]

00:29:42   on what i think are the important issues [TS]

00:29:44   behind this if you know anything about [TS]

00:29:46   cpu technology and none of this is going [TS]

00:29:48   to be new information and i apologize [TS]

00:29:51   but most of the time i do a show like [TS]

00:29:53   this i get positive feedback from the [TS]

00:29:54   people who may not be you know nerds in [TS]

00:29:58   this particular section of the topic so [TS]

00:30:00   if you're a tech nerd you can tune out [TS]

00:30:04   for a while and just wait till the end [TS]

00:30:05   when i get to some conclusions and but [TS]

00:30:07   if you're not i hope i will at least [TS]

00:30:09   teach you something in broad strokes [TS]

00:30:12   that you may not have known too much [TS]

00:30:14   about before alright so let's start with [TS]

00:30:18   the risk versus sisk you know those [TS]

00:30:20   stand for damn i do not remember i used [TS]

00:30:23   to know and back in college I there you [TS]

00:30:26   go you'll be one of those people I was [TS]

00:30:28   getting value out of the section that's [TS]

00:30:30   round all right so these are all caps [TS]

00:30:33   are is C and C is C rse is reduced [TS]

00:30:37   instruction set computing on complex [TS]

00:30:39   instruction set computing yeah you go [TS]

00:30:42   pulling stuff up all right no as you as [TS]

00:30:45   soon as you said the first and i [TS]

00:30:46   remembered second yeah this is one of [TS]

00:30:48   those i think it's pretty much like a [TS]

00:30:50   backronym where the the acronym Sisk [TS]

00:30:53   didn't invent didn't exist until the [TS]

00:30:54   acronym risk was invented someone can [TS]

00:30:56   send me reckons it out but it's kind of [TS]

00:30:58   like oh I have an idea and I've given [TS]

00:31:00   name and I'm also going to give a name [TS]

00:31:02   to your idea because my idea is better [TS]

00:31:04   than your idea and I without contrast [TS]

00:31:06   them and yours as complex as my [TS]

00:31:07   introduce anyway this was a hot debate I [TS]

00:31:12   know decade ago longer yeah it's a [TS]

00:31:16   couple other i would say longer because [TS]

00:31:18   i remember the well this came this is [TS]

00:31:20   the whole spark vs intel thing its like [TS]

00:31:24   the 90's yeah mid late nineties even [TS]

00:31:26   before that does there's a bunch of [TS]

00:31:27   articles from ARS technica that are [TS]

00:31:29   great to look at this prism the show [TS]

00:31:31   notes one is called risk versus sisk the [TS]

00:31:33   post risk era by John Stokes from 1999 [TS]

00:31:37   that's a really old one and there's also [TS]

00:31:39   risk versus Sisk in the mobile error [TS]

00:31:40   also by John Stokes in 2008 if you want [TS]

00:31:44   nerdier backgrounds on this stuff I [TS]

00:31:47   suggest reading those articles all right [TS]

00:31:53   but I'm going to give you the cliff [TS]

00:31:54   notes version so first let's start with [TS]

00:31:55   you know reduced instruction set [TS]

00:31:57   computing complex instructions I [TS]

00:31:58   computing what the heck is an [TS]

00:31:59   instruction set instructions that is [TS]

00:32:02   basically the list of things that you [TS]

00:32:03   can ask a processor to do like you build [TS]

00:32:07   this little piece of hardware and you [TS]

00:32:08   have to talk to it and what can you tell [TS]

00:32:10   it to do it is the API for the processor [TS]

00:32:13   using the parlance of our time yeah cuz [TS]

00:32:16   there's only you know what can you tell [TS]

00:32:17   a processor to do this limited set of [TS]

00:32:19   things you can tell it to do and you [TS]

00:32:21   have to define them can you it can you [TS]

00:32:22   tell the process of dad two numbers [TS]

00:32:23   together probably right so how do you do [TS]

00:32:25   that while you tell this instruction and [TS]

00:32:27   you put this number here in this number [TS]

00:32:28   that you know there's a list of things [TS]

00:32:29   that you can tell it to do did you ever [TS]

00:32:31   get assembly I sure did okay surely did [TS]

00:32:34   yes too much assembly so how it works is [TS]

00:32:40   that you know these days is source code [TS]

00:32:43   is compiled the machine code which is [TS]

00:32:45   like a list of instructions for the CPU [TS]

00:32:47   and the CPU reads those instructions [TS]

00:32:48   from memory and execute them right [TS]

00:32:50   machine code is a turd term you don't [TS]

00:32:53   hear thrown around too much today mostly [TS]

00:32:55   because not that relevant but what it [TS]

00:32:56   what it comes down to is just a bunch of [TS]

00:32:58   numbers right it's going to read a bunch [TS]

00:33:00   of numbers from memory and it's going to [TS]

00:33:03   interpret them in a certain way that's [TS]

00:33:04   like oh when I see the number five that [TS]

00:33:06   means the add instruction and then the [TS]

00:33:07   next number is the one thing I'm adding [TS]

00:33:09   in the next number is the other thing [TS]

00:33:10   I'm [TS]

00:33:10   it's not really like that but that's [TS]

00:33:12   that's the idea ah almost nobody writes [TS]

00:33:15   machine code in the end these days uh [TS]

00:33:18   maybe people who would like me and [TS]

00:33:20   undergraduate course had to design their [TS]

00:33:23   own CPUs and then program with machine [TS]

00:33:24   code ended up writing machine code where [TS]

00:33:26   you just literally write a series of [TS]

00:33:27   numbers it is like the least human [TS]

00:33:29   friendly way to program I guess toggle [TS]

00:33:31   switches we worked so if you couldn't [TS]

00:33:33   even type the numbers in you had to do [TS]

00:33:34   them their binary representation with [TS]

00:33:36   toggle switches that would be worse but [TS]

00:33:38   no one does that right assembly which is [TS]

00:33:40   a term you may have heard assembly code [TS]

00:33:42   is kind of like a symbolic version a [TS]

00:33:44   machine code instead of having my number [TS]

00:33:46   five you decide to if you type a DD the [TS]

00:33:49   word ad we're going to tell us the ad [TS]

00:33:51   instructor so you don't have to remember [TS]

00:33:52   that the add instruction is number five [TS]

00:33:53   you just type in a DD and then the next [TS]

00:33:56   thing you type in a bunch of letters it [TS]

00:33:57   says where it's getting one number from [TS]

00:33:58   and you know whatever it's much nicer [TS]

00:34:00   for human beings to not have to type [TS]

00:34:02   just a series of numbers they want words [TS]

00:34:04   or and so they just basically map the [TS]

00:34:05   words two numbers you take that assembly [TS]

00:34:07   code you run it through an assembler it [TS]

00:34:09   reads the little words it converts them [TS]

00:34:11   into the actual numbers and then the [TS]

00:34:12   Machine execute those numbers and those [TS]

00:34:15   little words or those numbers that's [TS]

00:34:16   your instruction set about you've got an [TS]

00:34:18   add obstruction or subtract instruction [TS]

00:34:19   and compare instructional also the stuff [TS]

00:34:21   they have numbers assigned to them you [TS]

00:34:23   write an assembly code the CPU is not [TS]

00:34:25   executing assembly code your assemblers [TS]

00:34:26   looking at the assembly code converting [TS]

00:34:28   them pretty much one to one into their [TS]

00:34:30   equivalent numbers putting them into [TS]

00:34:31   memory than getting the CPU to slurp [TS]

00:34:33   those instructions out of memory and [TS]

00:34:34   then go through each one execute it and [TS]

00:34:36   so and so forth right now people used to [TS]

00:34:39   write programs in assembly a lot more [TS]

00:34:40   than they do today now almost no one was [TS]

00:34:42   writing a machine code in our lifetimes [TS]

00:34:43   button assembly you know people used to [TS]

00:34:47   write programs that huge parts of the [TS]

00:34:49   original mac operating system were [TS]

00:34:51   written in assembly not just the drivers [TS]

00:34:53   or you know anything you would think [TS]

00:34:54   it's like low-level code but just like [TS]

00:34:56   huge parts of the actual operating is [TS]

00:34:57   like parts of the mac toolbox that did [TS]

00:34:59   the GUI like buttons and dialogues and [TS]

00:35:01   menus big parts of that were in an [TS]

00:35:04   assembly a lot of the reason they're [TS]

00:35:05   written assembly is because assembly is [TS]

00:35:07   like you know that's you talking almost [TS]

00:35:09   directly to the cpu granted you're [TS]

00:35:11   writing in a symbolic language thick is [TS]

00:35:12   converted into numbers but it's very [TS]

00:35:14   close to one to one correspondence is [TS]

00:35:16   what's going on inside the CPU so if you [TS]

00:35:18   want to wring every last piece of [TS]

00:35:19   performance out soon as I look CPU I [TS]

00:35:21   gotta tell you do this then you're going [TS]

00:35:23   to do this and [TS]

00:35:24   you're going to put the result there and [TS]

00:35:25   the next instructor not going to take [TS]

00:35:26   that result from there and put it over [TS]

00:35:27   here like very low level direct you know [TS]

00:35:30   not not a big layer of abstraction [TS]

00:35:31   between you and telling the CPU what to [TS]

00:35:33   do and assembly code can be very small [TS]

00:35:36   and very clever and do interesting [TS]

00:35:37   things so and memory was very tight back [TS]

00:35:41   then so large parts of the back-up rings [TS]

00:35:42   is written in assembly and if you're [TS]

00:35:46   writing anything in assembly what you [TS]

00:35:47   want is a a processor that has lots of [TS]

00:35:51   powerful instructions for assembly right [TS]

00:35:53   you wanted to be able to write an [TS]

00:35:57   instruction to this cpu where do [TS]

00:35:59   something useful for you all right so [TS]

00:36:01   here's an example of what later became [TS]

00:36:04   known as complex instruction set [TS]

00:36:06   computer single command that says take a [TS]

00:36:08   piece of memory from location a and move [TS]

00:36:12   that piece of memory into a different [TS]

00:36:13   place in memory location B that seems [TS]

00:36:16   like the simplest possible instruction [TS]

00:36:17   you can do but it's not it's actually [TS]

00:36:19   pretty complicated because you were [TS]

00:36:20   telling the CPU go into Ram fine address [TS]

00:36:24   57 take the stuff that's there take a [TS]

00:36:27   bite take a word dick a bubble or take [TS]

00:36:29   some piece of memory and then move it to [TS]

00:36:32   a different place in memory address 122 [TS]

00:36:35   right that is actually relatively [TS]

00:36:40   complicated because there's no way you [TS]

00:36:43   know it's it has to you have to the CPU [TS]

00:36:44   have to pull the thing out of memory [TS]

00:36:46   hold it for a second and then stick it [TS]

00:36:49   into another place but if you were [TS]

00:36:51   writing if you were had to write [TS]

00:36:52   assembly you would want to be able to [TS]

00:36:54   write that you wouldn't want to have to [TS]

00:36:55   write take the thing that's in memory [TS]

00:36:57   location 57 pull it into the CPU take it [TS]

00:37:01   out of the CPU and stick it into memory [TS]

00:37:03   location 256 you just want to say just [TS]

00:37:04   move it from 56 to 27 or whatever you [TS]

00:37:06   just one instruction right that is an [TS]

00:37:08   example of a complex instruction set it [TS]

00:37:12   it's you can imagine the the individual [TS]

00:37:15   steps that you want to do but you don't [TS]

00:37:17   want to have to write those individual [TS]

00:37:18   step you just want to sell cheap cell is [TS]

00:37:20   super you look let's move this thing [TS]

00:37:21   from one place to the other because if [TS]

00:37:22   you didn't you have to write three [TS]

00:37:23   instructions every time you just want it [TS]

00:37:24   to move something from one thing it so [TS]

00:37:26   the race back then was to make CPUs that [TS]

00:37:28   had really fancy instructions so that if [TS]

00:37:30   you're writing an assembler or even if [TS]

00:37:32   you're writing a compiler or whatever [TS]

00:37:33   you're like I don't need to handhold the [TS]

00:37:35   CPU I can say [TS]

00:37:37   here's what I want you to do you figure [TS]

00:37:39   out how to do it I don't care how you do [TS]

00:37:40   it but I have a very powerful [TS]

00:37:41   complicated instruction and you do that [TS]

00:37:44   the way you want it that's complex [TS]

00:37:45   instruction set computing because that [TS]

00:37:47   they all the CPU makers are trying to [TS]

00:37:49   make you know my CPU has an instruction [TS]

00:37:51   and lets you you know compared to [TS]

00:37:52   strings to see whether they're equal [TS]

00:37:54   just making this up but like like wow I [TS]

00:37:56   don't even have to go through the [TS]

00:37:57   individual steps of that the CPU will [TS]

00:37:58   just do it itself that's a great CPU [TS]

00:37:59   it's powerful it's awesome right who [TS]

00:38:02   wouldn't like that John right alright so [TS]

00:38:05   here's here this kind of thing falls [TS]

00:38:07   down and I'm going to go into a crazy [TS]

00:38:08   analogy that's not even remotely close [TS]

00:38:10   to being correct but hopefully give you [TS]

00:38:12   just what's going on right so I'm going [TS]

00:38:14   to say that this the complex instruction [TS]

00:38:16   set computing having very powerful [TS]

00:38:18   instructions I can tell the computer to [TS]

00:38:19   do this complicated thing is kind of [TS]

00:38:21   like a sewing machine and there's a [TS]

00:38:24   particular sewing machine animated gift [TS]

00:38:27   that I tried to find on the web I found [TS]

00:38:29   like eight thousand versions of it I put [TS]

00:38:30   one in the chat room here I have you [TS]

00:38:32   seen this one you're looking at chat [TS]

00:38:34   room I am looking let me see oh yeah [TS]

00:38:37   this is very essence old one that's [TS]

00:38:41   great there's a glier ones too but for [TS]

00:38:43   people who don't know how sewing [TS]

00:38:44   machines work or have never used a [TS]

00:38:45   sewing machine take a look of his [TS]

00:38:46   animation you see all seen sewing [TS]

00:38:48   machines it's like a big machine with [TS]

00:38:51   like a needle and you put fabric down on [TS]

00:38:53   a table and you slide it into needle [TS]

00:38:54   needles up and down really fast up and [TS]

00:38:56   down up and down up and down you feed [TS]

00:38:57   the material through it and so does [TS]

00:38:58   things together this animation shows you [TS]

00:39:01   what's actually going on it also [TS]

00:39:02   explains the mystery of the bobbin yeah [TS]

00:39:05   like when the needle goes down in a song [TS]

00:39:07   machine it's not just just punching a [TS]

00:39:09   string through and pulling it back up [TS]

00:39:10   because that wouldn't work we'll just [TS]

00:39:11   make a series of holes all right there's [TS]

00:39:12   a thread threaded through the needle and [TS]

00:39:14   when it goes down through the fabric [TS]

00:39:15   into the table below it is a little [TS]

00:39:17   bobbin that basically takes some thread [TS]

00:39:20   that's on the bobbin and weave it [TS]

00:39:21   through the thread that came down with [TS]

00:39:23   the needle and you get stitches so if [TS]

00:39:25   you stare at this gif animation long [TS]

00:39:26   enough you should get an idea of like [TS]

00:39:28   the basics of how a sewing machine works [TS]

00:39:31   and it's actually you see what it's [TS]

00:39:32   doing it's like you know if a human [TS]

00:39:33   being was doing it they wouldn't [TS]

00:39:34   probably do it in this exactly this way [TS]

00:39:35   but you end up getting a bunch of [TS]

00:39:37   interlocked stitches with one set of [TS]

00:39:39   thread going down the needle and the [TS]

00:39:40   other set of thread on the bobbin and [TS]

00:39:42   the interlock with each other and so two [TS]

00:39:43   pieces of fabric together and so this is [TS]

00:39:46   a pretty ingenious machine I also put in [TS]

00:39:48   two links to a very old show called the [TS]

00:39:51   secret [TS]

00:39:51   life of machines oh yeah which was a [TS]

00:39:54   british show most notable to my young [TS]

00:39:56   American self as the program where i [TS]

00:39:58   learned what sticky tape is as opposed [TS]

00:40:01   to non sticky tape i guess that's what [TS]

00:40:03   they call scotch tape and Scotch is the [TS]

00:40:05   name brand anyway put a link to that in [TS]

00:40:08   the show notes it's like The Secret Life [TS]

00:40:09   machine is a sewing machine part 1 and [TS]

00:40:10   part 2 I actually encourage you to watch [TS]

00:40:12   these things yes they look like they're [TS]

00:40:14   on terrible video from the 70s or 80s [TS]

00:40:15   and the people have British accents and [TS]

00:40:18   they're blurry but really watch it and I [TS]

00:40:20   think by the end of watching these [TS]

00:40:21   programs unlike by the end of this [TS]

00:40:22   podcast you will actually understand how [TS]

00:40:25   sewing machines work and I think it's [TS]

00:40:27   important because you can see you know [TS]

00:40:28   how how they figured out how to do this [TS]

00:40:29   how to take this very complex series of [TS]

00:40:31   motions having one piece of thread [TS]

00:40:33   threaded through a hole in fabric and [TS]

00:40:36   take another piece of thread and loop it [TS]

00:40:38   through that and then come back out how [TS]

00:40:39   can you do that like without having a [TS]

00:40:41   human being threading things through a [TS]

00:40:43   little hole and guiding it through and [TS]

00:40:44   the Machine they came up with to do it [TS]

00:40:46   looks almost nothing like it would look [TS]

00:40:47   like if your hands did it but it it is [TS]

00:40:49   you know basically the bobbin is being [TS]

00:40:51   passed through not literally but you [TS]

00:40:54   know effectively the threat on the [TS]

00:40:57   bobbin is being passed through a loop [TS]

00:40:58   brought down by the other needle and [TS]

00:41:00   it's quite an interesting little machine [TS]

00:41:02   right this is very close to [TS]

00:41:05   philosophically what Sisk is like like [TS]

00:41:07   if you had to explain someone how to sew [TS]

00:41:09   two things together uh with a particular [TS]

00:41:12   locking stitch that holds two pieces of [TS]

00:41:14   fabric together you can do it but that's [TS]

00:41:16   a lot of steps and you'll be like [TS]

00:41:18   wouldn't it be better if I could just [TS]

00:41:19   sort of feed a piece of fabric through a [TS]

00:41:21   machine I say it machine look just do [TS]

00:41:24   that stitch thing I don't wanna have to [TS]

00:41:25   tell you all the individual steps so it [TS]

00:41:27   basically comes down to is you gotta [TS]

00:41:28   punch needle through and then something [TS]

00:41:29   grabs little thing and then you gotta [TS]

00:41:30   have a little bobbin it tucks its thread [TS]

00:41:32   through there but I don't have typed I [TS]

00:41:33   just want to have instruction called [TS]

00:41:34   stitch and I hit a pedal on the floor [TS]

00:41:36   anything goes stitch and I'm I just hold [TS]

00:41:38   down that pedal go stitch stitch stitch [TS]

00:41:40   stitch stage so I can feed the fabric [TS]

00:41:42   through and I've made a sewing machine [TS]

00:41:44   is awesome sisk processors were like [TS]

00:41:46   sewing machines and the cooler steps [TS]

00:41:48   they could do the more interesting [TS]

00:41:50   stitches that they could do the more [TS]

00:41:52   powerful those CPUs were but like a [TS]

00:41:55   sewing machine like there's a limit on [TS]

00:41:57   how fast the sewing machine can do what [TS]

00:41:59   it does if you stare at that animation [TS]

00:42:01   you're like okay how fast can this [TS]

00:42:02   animation go can I do like five [TS]

00:42:05   just a second ten stitches a second [TS]

00:42:07   maybe tonight of 20 maybe but now you [TS]

00:42:11   know my 20 stitches a second that's [TS]

00:42:12   pretty darn fast can you make a machine [TS]

00:42:14   that performs a series of complicated [TS]

00:42:16   moves that best can I do 200,000 [TS]

00:42:18   stitches a second probably not someone [TS]

00:42:21   someone from the industrial sewing [TS]

00:42:23   machine world will send me some feedback [TS]

00:42:24   on this show and tell me how fast the [TS]

00:42:26   fastest sewing machine in the world can [TS]

00:42:27   go go but like at a certain point [TS]

00:42:29   pressing the pedal to do one stitch and [TS]

00:42:32   it's like you told me to do one stitch [TS]

00:42:33   and I got into this series of 27 really [TS]

00:42:34   complicated steps and the steps depend [TS]

00:42:37   on each other the kind of interlocked [TS]

00:42:38   and you know you get into issues of like [TS]

00:42:40   physics of things hurtling past each [TS]

00:42:41   other and you can't thread this thing [TS]

00:42:43   through until the needle comes down but [TS]

00:42:44   you got to make sure the needle is all [TS]

00:42:45   the way down before you grab it then you [TS]

00:42:47   but you have to start and stop a piece [TS]

00:42:48   of machinery from moving it's there is a [TS]

00:42:51   limit how fast a sewing machine can [TS]

00:42:53   operate compare that limit of how fast [TS]

00:42:56   can I make this little sewing machine [TS]

00:42:57   thing go compare that limit to how fast [TS]

00:43:00   an electric motor can spin just plain [TS]

00:43:03   old straight barrel electric motor a [TS]

00:43:06   sewing machine you know a thousand rpm [TS]

00:43:09   100 rpm and you know stitches per minute [TS]

00:43:11   rotations from it or whatever how many [TS]

00:43:13   stitches can you make per second doesn't [TS]

00:43:15   come close to an electric motor electric [TS]

00:43:16   motor can do tens of thousands of [TS]

00:43:18   rotations per minute a electric motor [TS]

00:43:21   can spend very very fast you just you [TS]

00:43:24   know it you put it in some really good [TS]

00:43:26   ball bearings you have some magnets in a [TS]

00:43:28   coil of wire and electricity maybe you [TS]

00:43:31   even magnetically levitates spindles so [TS]

00:43:33   there's no friction except for air that [TS]

00:43:34   you could suck the air out of it to make [TS]

00:43:35   it a vacuum you know you can make [TS]

00:43:37   electric motor spin really fast because [TS]

00:43:39   all does is spin there's no oscillating [TS]

00:43:41   parts there's no thread going through a [TS]

00:43:44   piece of fabric and one piece of thread [TS]

00:43:46   being shoved through a nose but it's [TS]

00:43:47   just spinning that's all does it rotates [TS]

00:43:50   around again and again and you can do [TS]

00:43:52   that really really really fast and if [TS]

00:43:54   you want to make that luxury motor go [TS]

00:43:56   really fast you know right where you [TS]

00:43:57   have to concentrate what are the things [TS]

00:43:59   stopping me from going too fast friction [TS]

00:44:01   uh that the force inside of the thing [TS]

00:44:04   I'm making it spin apart if it goes too [TS]

00:44:05   fast you know that isn't the centripetal [TS]

00:44:07   force on the outsides of the piece of [TS]

00:44:09   metal causing to disintegrate but those [TS]

00:44:10   limits are way higher than the limits of [TS]

00:44:12   a sewing machine if you had a contest [TS]

00:44:13   between the best engineers in the world [TS]

00:44:14   how fast can you make a sewing machine [TS]

00:44:16   go versus how fast can you make an [TS]

00:44:17   electrical [TS]

00:44:18   the electric motor guys would destroy [TS]

00:44:20   the sewing machine and sew the risk [TS]

00:44:23   philosophy was that they figured this [TS]

00:44:25   out you know they said all right we're [TS]

00:44:27   making these awesome cpu's with these [TS]

00:44:28   really complicated instructions but geez [TS]

00:44:30   every time we try to make these things [TS]

00:44:31   go faster like how fast can we make that [TS]

00:44:33   instruction to do some complicated [TS]

00:44:36   operation I guess of all these steps [TS]

00:44:38   involved and they're all interlocked [TS]

00:44:40   with each other and this limit how fast [TS]

00:44:42   I can go with that what if we make the [TS]

00:44:45   CPU equivalent of that electric motor [TS]

00:44:47   where what it does is really stupid and [TS]

00:44:49   simple but we can do it really really [TS]

00:44:51   fast uh and so that's that's a little [TS]

00:44:55   definition of risk here from the the [TS]

00:44:56   Wikipedia article CPU design strategy [TS]

00:44:58   based on the inside that simplified as [TS]

00:45:00   opposed to complex instructions can [TS]

00:45:01   provide higher performance this [TS]

00:45:02   simplicity enables much faster execution [TS]

00:45:04   of each instruction so an example of a [TS]

00:45:10   RISC architecture is the ARM [TS]

00:45:11   architecture it's a emergency for acorn [TS]

00:45:14   risc machines i believe and i forget [TS]

00:45:15   what they we went through a [TS]

00:45:16   senate-passed show I'm not gonna look it [TS]

00:45:17   up again there's only the second we have [TS]

00:45:19   feedback about if any way is a risk [TS]

00:45:20   machine and they use called like load [TS]

00:45:22   store architectures and the reason load [TS]

00:45:24   store architecture comes up is because [TS]

00:45:25   given that complex instruction operation [TS]

00:45:29   from before take a piece of memory from [TS]

00:45:33   location 57 and move it to 128 on the [TS]

00:45:35   risc cpu that would be read the location [TS]

00:45:39   57 into register one right location [TS]

00:45:42   right register one out into location in [TS]

00:45:44   256 so it would be two operations [TS]

00:45:46   instead of one I may have gotten that [TS]

00:45:48   wrong with the details so basically you [TS]

00:45:50   can take any complicated operation and [TS]

00:45:53   decompose it into simpler operations so [TS]

00:45:56   the move from one place member to [TS]

00:45:58   another it's going to have to go through [TS]

00:45:58   the CPU anyway there is no instruction [TS]

00:46:00   that says move it from the settle [TS]

00:46:02   occasion the other one you have to issue [TS]

00:46:04   the two taken from here put it to their [TS]

00:46:05   take a rear porter there and that seems [TS]

00:46:08   like a loss like wasn't it much better [TS]

00:46:10   when you could just tell it to move from [TS]

00:46:11   one place to the other but in actuality [TS]

00:46:12   because those those risk instructions [TS]

00:46:15   are so simple you can make them go much [TS]

00:46:18   faster and you can interleave them with [TS]

00:46:20   each other and duels all sorts of other [TS]

00:46:21   things because the instructions [TS]

00:46:24   themselves are small and so everything [TS]

00:46:26   about risk machines was designed to be [TS]

00:46:28   more like the electric motor and less [TS]

00:46:29   like the sewing machine they [TS]

00:46:31   won't have any special hardware special [TS]

00:46:34   registers that only serve a single [TS]

00:46:36   purpose when you divide two numbers the [TS]

00:46:37   result always goes in this register they [TS]

00:46:39   wanted to have a big very uniform [TS]

00:46:41   register file with lots of what they [TS]

00:46:43   call general purpose registers we're [TS]

00:46:45   like when you divide two numbers the [TS]

00:46:46   result doesn't always have to go into [TS]

00:46:47   this register it can go in any of the [TS]

00:46:48   registers in fact all the registers the [TS]

00:46:50   same and we have a bunch of them because [TS]

00:46:51   there's going to be a bunch of these [TS]

00:46:53   little tiny instructions and flight at [TS]

00:46:54   once and maybe we can put more than [TS]

00:46:56   parallel and intro leave them with each [TS]

00:46:58   other we just want to make it regular [TS]

00:47:00   and uniform they did stuff like uh you [TS]

00:47:03   know either discouraging or outright [TS]

00:47:04   forbidding unaligned memory access you [TS]

00:47:06   know old machines would be like we can [TS]

00:47:08   read memory anywhere you don't have to [TS]

00:47:09   start from a particular location you [TS]

00:47:11   just tell me you can read you a bite at [TS]

00:47:12   a time from anywhere in memory and then [TS]

00:47:14   risk machines are more like now you can [TS]

00:47:15   only do a line memory access you have to [TS]

00:47:17   start on on a you know a multiple of 32 [TS]

00:47:19   bits or multiple 16 bits or something [TS]

00:47:21   some of the trunk like that if that's [TS]

00:47:23   inconvenient for you tough luck take up [TS]

00:47:25   the bigger piece bring it into the CPU [TS]

00:47:26   chop it up how you want uh and it's [TS]

00:47:29   basically you know making it much much [TS]

00:47:32   less pleasant to program an assembly [TS]

00:47:34   language because you're getting a bit [TS]

00:47:35   much dumber simpler machine that has [TS]

00:47:37   much more limitations and so many ones [TS]

00:47:39   into assembly programming be like I [TS]

00:47:40   don't want to program a risc cpu I got [TS]

00:47:42   to tell it how to do everything and have [TS]

00:47:43   instructions i wanted to aren't even [TS]

00:47:45   there I can't even pull a byte of memory [TS]

00:47:46   from here I gotta pull like 32 bits of [TS]

00:47:49   memory and mask it to get the bytes I [TS]

00:47:51   want and it's just this terrible right [TS]

00:47:52   uh that's not what I was designed for it [TS]

00:47:55   was designed for an age where compilers [TS]

00:47:56   right that for you and optimize it for [TS]

00:47:58   you purely because like the electric [TS]

00:48:00   motor they could spin a risc cpu really [TS]

00:48:02   really fast and in CPU world spinning [TS]

00:48:04   really fast is the clock speed it's [TS]

00:48:06   megahertz how many instructions per [TS]

00:48:08   second or how many cycles per second can [TS]

00:48:10   can happen oh the other thing that had [TS]

00:48:12   done these on risky p uses they would [TS]

00:48:14   try to make almost every single [TS]

00:48:15   instruction executes in a single clock [TS]

00:48:16   not everyone because this you know [TS]

00:48:18   complicated division instructions and [TS]

00:48:19   stuff like that but they would want they [TS]

00:48:22   would not want to have an instruction [TS]

00:48:23   like a one sis cpus where it's some big [TS]

00:48:25   complicated instruction like the song [TS]

00:48:26   machine it takes 17 cycles to execute [TS]

00:48:28   you know adding two numbers here they [TS]

00:48:30   wanted all their instructions to be so [TS]

00:48:32   simple that every single one of them [TS]

00:48:33   executed in a single cycle except for [TS]

00:48:35   maybe a couple of outliers so again the [TS]

00:48:37   entire instruction set the entire CPU [TS]

00:48:39   architecture was designed to make the [TS]

00:48:40   CPU go fast and not to make assembly [TS]

00:48:44   programmers like you [TS]

00:48:45   ye're and and or you know anything like [TS]

00:48:48   that they've expected a compiler to spit [TS]

00:48:49   out this stuff all right let me see what [TS]

00:48:54   up man these are there's some good rants [TS]

00:48:58   for me today yeah I'm really revisiting [TS]

00:49:01   old territory all right let's do the [TS]

00:49:02   second sponsor lynda.com online learning [TS]

00:49:04   company these guys have more than 77,000 [TS]

00:49:07   video tutorials i call them screencasts [TS]

00:49:09   they call video tutorials they're gonna [TS]

00:49:11   have to they're you know that's their [TS]

00:49:12   thing they know what the call they teach [TS]

00:49:15   you stuff they teach you everything from [TS]

00:49:16   design photography business programming [TS]

00:49:20   a you name it audio and video stuff even [TS]

00:49:22   have 3d animation stuff in there they're [TS]

00:49:24   adding new courses all the time all the [TS]

00:49:27   time and the way this thing works is you [TS]

00:49:28   get a membership and then you get access [TS]

00:49:30   to everything that's in there so that [TS]

00:49:32   you can teach yourself stop from experts [TS]

00:49:35   they have experts who are working in the [TS]

00:49:37   industry with real-world experience [TS]

00:49:39   creating these things teaching these [TS]

00:49:41   things and selling them there and it's [TS]

00:49:43   really really cool I love learning with [TS]

00:49:46   screencast and of course I mean I'm [TS]

00:49:48   probably biased because I've made [TS]

00:49:49   screencast before I used to make a [TS]

00:49:50   living doing screencast but I love this [TS]

00:49:52   stuff it's over on lynda.com that [TS]

00:49:55   spelled ly n da lynda.com membership [TS]

00:49:59   starts at 25 bucks a month but you get [TS]

00:50:01   unlimited 24 by 7 access to the best [TS]

00:50:04   videos out there they made a special URL [TS]

00:50:07   you can visit linda.com / 5 by 5 and if [TS]

00:50:12   you go there you'll get seven days for [TS]

00:50:14   free access to everything seven days [TS]

00:50:16   free and you'll support the show you'll [TS]

00:50:18   make John Happ you'll make me happy [TS]

00:50:19   Linda calm / 5 / 5 check them out so we [TS]

00:50:25   got some numbers on actual risk and CPUs [TS]

00:50:28   in terms of number of registers and [TS]

00:50:29   stuff just to reinforce the point about [TS]

00:50:31   the register whether they call the [TS]

00:50:32   register file how many differenter says [TS]

00:50:34   you have so the PowerPC cpu used to be [TS]

00:50:36   in max that's also a risc cpu and like i [TS]

00:50:39   said risks abused and have way more [TS]

00:50:40   registers and they're all kind of you [TS]

00:50:42   know no special purpose registers or few [TS]

00:50:45   special purpose or mostly generate lots [TS]

00:50:46   of general-purpose registers so power [TS]

00:50:49   busy had 32 general purpose registers [TS]

00:50:51   and 32 floating point registers and and [TS]

00:50:53   compared to x86 intel chips back when [TS]

00:50:56   they were 16 bit x86 chips [TS]

00:50:59   had six registers which were kind of [TS]

00:51:02   sort of dedicated special purpose not [TS]

00:51:05   all them are really general sort of like [TS]

00:51:06   I said you know when you did a divided [TS]

00:51:08   instruction the result always had to go [TS]

00:51:09   in a particular register and stuff like [TS]

00:51:10   that plus a couple of other things for [TS]

00:51:13   like stack pointers and stuff 32-bit x86 [TS]

00:51:16   had six general purpose registers plus [TS]

00:51:19   some extended versions of the you know [TS]

00:51:21   stack pointer registers and stuff 64-bit [TS]

00:51:24   they added a few more you got 14 general [TS]

00:51:26   purpose registers plus the other ones [TS]

00:51:28   and so you see like x86 has been trying [TS]

00:51:31   to add more registers but PowerPC from [TS]

00:51:33   the day was born had this you know huge [TS]

00:51:35   number registered more registers than [TS]

00:51:36   you would expect the human being to keep [TS]

00:51:37   track of because if you're writing [TS]

00:51:38   assembling you got to keep track of 32 [TS]

00:51:40   registers like I will tell that i put in [TS]

00:51:42   register 17 i want to add these two [TS]

00:51:44   numbers together was that in register 32 [TS]

00:51:46   or 57 it's just you can imagine writing [TS]

00:51:48   so maybe you can remember adding [TS]

00:51:50   assembly by hand and like it's hard to [TS]

00:51:51   keep track of these things that they're [TS]

00:51:52   all just called r 1 r 7 r 15 are three [TS]

00:51:54   and they're all the same and how can you [TS]

00:51:56   keep track of where anything is like [TS]

00:51:58   it's totally made for a machine because [TS]

00:51:59   the machine say okay I've got 32 [TS]

00:52:01   registers to work with and I can just [TS]

00:52:02   keep track of them and have sort of a [TS]

00:52:04   register allocation algorithm I figure [TS]

00:52:05   out how to use them when I compile this [TS]

00:52:07   code it's much better for a machine a [TS]

00:52:08   floating point is a good example to on [TS]

00:52:10   x86 based have a separate floating-point [TS]

00:52:13   unit called the x87 stack-based [TS]

00:52:15   floating-point unit which was a a [TS]

00:52:18   dedicated separate little thing itself [TS]

00:52:20   would eventually moved on die but it was [TS]

00:52:22   stack base where you put your number [TS]

00:52:23   push another number and then push on a [TS]

00:52:24   multiplication would multiply the top [TS]

00:52:27   two numbers on the stack but you could [TS]

00:52:28   also reach down lower into the stack and [TS]

00:52:30   do stuff and I it's not really conducive [TS]

00:52:32   to parallelization because when you have [TS]

00:52:35   a stack it's like a stack of plates [TS]

00:52:37   where you pushed push two plates on the [TS]

00:52:38   top and perform an operation that [TS]

00:52:39   combines those two plates into one plate [TS]

00:52:41   and puts the result back on top of the [TS]

00:52:42   stack and if you want have 17 people [TS]

00:52:44   operating at the same time on a stack [TS]

00:52:46   it's not really great because of the way [TS]

00:52:48   a stack works you can have some guy in [TS]

00:52:49   the middle yanking things in and out [TS]

00:52:51   because what if the guy on the top was [TS]

00:52:52   doing yet anyway stack-based floating [TS]

00:52:55   point was also a problem for them [TS]

00:52:56   because it was kind of designed in in a [TS]

00:52:59   different error when the most important [TS]

00:53:00   thing wasn't like parallel execution and [TS]

00:53:02   simplicity of things now what all is [TS]

00:53:04   article that is referred to at the very [TS]

00:53:06   front of the article by John Stokes were [TS]

00:53:07   in reference to like Cisco versus risk [TS]

00:53:09   at the post ristic post risk era that [TS]

00:53:12   was all [TS]

00:53:13   I'll sort of rehashing this cisco stuff [TS]

00:53:15   that I just talked about and explaining [TS]

00:53:17   how although that was the origin of the [TS]

00:53:21   risk movement it's much less relevant [TS]

00:53:23   today because these days when you have [TS]

00:53:25   this x86 quote-unquote CPU have these [TS]

00:53:28   complex complex instructions the way [TS]

00:53:31   they work internally is they take that [TS]

00:53:33   complicated instruction that says move [TS]

00:53:35   something from one memory address to [TS]

00:53:36   another and RAM and they break it down [TS]

00:53:38   into what they call micro operations so [TS]

00:53:40   they say okay even though you said move [TS]

00:53:43   this data word from address here at [TS]

00:53:45   address there I'm going to break that [TS]

00:53:47   single instruction into two smaller [TS]

00:53:49   instructions one that says get that [TS]

00:53:51   thing from memory and put it into this [TS]

00:53:52   register and other that says take that [TS]

00:53:54   register and stick and into memory all [TS]

00:53:55   right so they basically taken a sisk [TS]

00:53:57   instruction set and inside the CPU break [TS]

00:54:00   it down into something more [TS]

00:54:01   approximating a RISC instruction set and [TS]

00:54:03   the actual CPU is built to execute these [TS]

00:54:06   tiny little micro operations it's not [TS]

00:54:08   executing you know it doesn't the CPU [TS]

00:54:11   doesn't see this big complicated [TS]

00:54:12   instruction that says to do 50 things it [TS]

00:54:14   sees a stream of tiny little micro [TS]

00:54:16   operations that were created from those [TS]

00:54:18   single instructions so they build the [TS]

00:54:19   entire CPU internally to execute these [TS]

00:54:22   micro operations very efficiently and [TS]

00:54:23   very quickly and they just have a [TS]

00:54:25   decoder part on the front on the front [TS]

00:54:27   end that takes your complicated [TS]

00:54:28   instruction see and it decodes it into [TS]

00:54:30   these simpler ones and that's the idea [TS]

00:54:34   of getting at the post risk era where [TS]

00:54:35   it's like all right so the instruction [TS]

00:54:36   set itself how relevant is that if [TS]

00:54:38   everybody inside is building basically [TS]

00:54:40   risk CPUs and if you have a complex [TS]

00:54:43   instructions that you just need a [TS]

00:54:44   decoder now part of the CPU and an x86 [TS]

00:54:46   chip is dedicated to all the stuff that [TS]

00:54:49   has to say okay I've got this big [TS]

00:54:50   complicated instruction coming in how [TS]

00:54:52   many micro operations does that break [TS]

00:54:53   down to can i have some piece of [TS]

00:54:56   hardware that automatically breaks that [TS]

00:54:57   down into the micro operations or is it [TS]

00:54:59   a really complicated one but i have to [TS]

00:55:00   have like a little software microcode [TS]

00:55:02   program that tells me how to break this [TS]

00:55:03   down you need to dedicate part of the [TS]

00:55:05   chip just to dealing with this thing and [TS]

00:55:07   by the way have to keep track of these [TS]

00:55:08   little micro operations you have to sort [TS]

00:55:10   of reassemble the result in to like make [TS]

00:55:13   it look as if that complicated [TS]

00:55:14   instruction really executed like you [TS]

00:55:15   can't have things go out of order it has [TS]

00:55:17   to look to the assembly programmer just [TS]

00:55:20   like those old x86 chips from Intel that [TS]

00:55:23   actually did implement this complex [TS]

00:55:24   instruction set internally despite [TS]

00:55:26   fact that inside is basically a risk [TS]

00:55:28   like CPU executing all the stuff so you [TS]

00:55:30   have all this the front-end hardware and [TS]

00:55:32   the tracking hardware and the [TS]

00:55:33   reassembling of things and making sure [TS]

00:55:34   the exact timings and exact rules are [TS]

00:55:37   maintained and they behave exactly the [TS]

00:55:39   same as they used to uh and the argument [TS]

00:55:43   back in the 90s was that all right so [TS]

00:55:45   Intel may have figured out that this [TS]

00:55:46   wrist thing is a good idea because they [TS]

00:55:48   can make their CPUs run faster but they [TS]

00:55:50   have to dedicate all this time and [TS]

00:55:51   energy and parts of the chip and like [TS]

00:55:53   they have to you know they basically [TS]

00:55:55   have to have this whole whole bunch of [TS]

00:55:56   gunk inside their CPUs to translate [TS]

00:55:58   their crappy complicated instruction set [TS]

00:56:00   into the simpler one and keep track of [TS]

00:56:01   everything and that's bad because you [TS]

00:56:04   know the PowerPC you guys don't need to [TS]

00:56:05   do that they don't have the whole [TS]

00:56:06   section of the chip doesn't have to [TS]

00:56:08   exist to break down their instructions [TS]

00:56:09   into micro operations and stuff like [TS]

00:56:11   that is they already are micro and [TS]

00:56:12   everything's fine so therefore PowerPC [TS]

00:56:14   is going to win and risk is awesome and [TS]

00:56:16   Intel drools and so on so forth and from [TS]

00:56:19   a technical perspective I remember [TS]

00:56:21   feeling that way that like Intel looks [TS]

00:56:24   like the crusty crappy thing you're like [TS]

00:56:25   well they may have clever engineers and [TS]

00:56:27   they may have figured out to break up [TS]

00:56:28   their instructions to smaller ones and [TS]

00:56:29   stuff but they gotta deal with all those [TS]

00:56:31   that you know making that stuff to break [TS]

00:56:33   up the instructions and keeping track of [TS]

00:56:34   them and reassembling the results and [TS]

00:56:35   keeping the order or consistent that's [TS]

00:56:37   thanks x86 thinks it's ugly it's gross [TS]

00:56:39   right Intel also basically agreed that [TS]

00:56:42   it was ugly in gross and they trot have [TS]

00:56:44   been trying for years and years to get [TS]

00:56:45   rid of the ugliest and grossest part so [TS]

00:56:48   the stack based floating-point that's [TS]

00:56:50   not really great for doing things [TS]

00:56:52   quickly or in parallel they've been [TS]

00:56:55   deprecating that they have to keep [TS]

00:56:56   supporting it but they're like look [TS]

00:56:57   don't if if you're writing a compiler [TS]

00:56:59   for example I and someone's doing [TS]

00:57:02   floating-point math don't generate x87 [TS]

00:57:04   instructions for the stack place [TS]

00:57:05   floating-point thing because that's slow [TS]

00:57:07   and it's always been slow and it's not [TS]

00:57:09   going to probably get much faster we [TS]

00:57:11   still have to support it you just saw [TS]

00:57:13   all your old programs are running stuff [TS]

00:57:14   but instead what we have here is [TS]

00:57:17   something that they called SSE streaming [TS]

00:57:19   simdi extensions cingulate single [TS]

00:57:22   instruction multiple data it's a way to [TS]

00:57:26   process lots of data in parallel once [TS]

00:57:30   like if you have you know two sets of 50 [TS]

00:57:32   numbers and you want to add the first [TS]

00:57:33   set of 50 numbers to the second set of [TS]

00:57:35   50 numbers you're have instruction that [TS]

00:57:36   does it all at once and that may not [TS]

00:57:38   seem like it's the same thing as [TS]

00:57:39   floating point [TS]

00:57:40   but it also operates on floating point [TS]

00:57:41   numbers and they said look compiler [TS]

00:57:43   writers even though you may not think [TS]

00:57:45   you're doing like you know doing a [TS]

00:57:48   single instruction on 50 pieces of data [TS]

00:57:50   the floating part hardware we have in [TS]

00:57:52   this in these SSE parts of our chip is [TS]

00:57:55   way faster in that stack based one we [TS]

00:57:56   can do stuff in parallel don't generate [TS]

00:57:59   instructions for a stack-based floating [TS]

00:58:01   point if you see any floating point code [TS]

00:58:02   in your you know high level you know no [TS]

00:58:04   higher level bin assembly your C program [TS]

00:58:07   or whatever just target the RSS see [TS]

00:58:09   extensions and that will make your stuff [TS]

00:58:12   go faster so they have they had four [TS]

00:58:13   versions of SSC so far i think is for [TS]

00:58:15   you know I just keep version them and [TS]

00:58:17   making them more and more powerful and [TS]

00:58:18   these days if you compile a program for [TS]

00:58:21   an Intel chip you would hope that [TS]

00:58:23   there's no stack-based floating-point [TS]

00:58:24   stuff being sent to the chip despite the [TS]

00:58:25   fact that it the chip is still able do [TS]

00:58:27   that all the floating-point stuff is [TS]

00:58:28   targeting this more modern more [TS]

00:58:30   parallelizable non stack base flowing [TS]

00:58:32   point Intel also went for 32-bit 64-bit [TS]

00:58:35   and they took that while not Intel but [TS]

00:58:38   AMD actually AMD came up with a x86 64 [TS]

00:58:41   which is a 64-bit bunch of instructions [TS]

00:58:44   that only work on 64-bit chips chips ba [TS]

00:58:48   that are compatible with old x86 [TS]

00:58:51   programs and they took that as an [TS]

00:58:53   opportunity to make a new instruction [TS]

00:58:55   set that is more amenable to modern ship [TS]

00:58:58   designs than the old one because there [TS]

00:59:00   were no 64-bit instructions at all [TS]

00:59:02   before there was 64 bit chicks so we [TS]

00:59:04   don't have to just make like we don't [TS]

00:59:06   have to stick with the old ways this is [TS]

00:59:08   an opportunity this is a discontinuity [TS]

00:59:09   to come up with a better instruction set [TS]

00:59:12   they'd added registers they you know [TS]

00:59:15   made the rules about how the the 64-bit [TS]

00:59:17   instructions execute make it easier to [TS]

00:59:20   do encode and decode instruction [TS]

00:59:21   tracking and parallelization a lot stuff [TS]

00:59:24   Intel was off doing its own thing with [TS]

00:59:26   the Itanium and another right I remember [TS]

00:59:29   that day talk about a very long word [TS]

00:59:31   instruction computing all stuff that's [TS]

00:59:32   whole other topic but anyway so if I sit [TS]

00:59:35   decide that Intel's thing didn't quite [TS]

00:59:36   work out x86 64 caught on and now Intel [TS]

00:59:40   makes chips with x86 64 but this wasn't [TS]

00:59:43   another opportunity for them to get rid [TS]

00:59:45   of old crusty stuff say if you're [TS]

00:59:47   targeting if your compiler writer and [TS]

00:59:48   you're targeting 64-bit CPUs use these [TS]

00:59:50   cool new instructions and you have [TS]

00:59:51   access to more registers and see isn't [TS]

00:59:53   this thing getting [TS]

00:59:53   sir isn't our cpu architecture getting [TS]

00:59:55   to be more like those nice clean RISC [TS]

00:59:57   architectures where you had just had a [TS]

00:59:58   bunch of general purpose registers and [TS]

00:59:58   bunch of general purpose registers and [TS]

01:00:00   simple instruction set we're not there [TS]

01:00:01   yet but we're slowly trying to abandon [TS]

01:00:03   our craft right so that that is the path [TS]

01:00:05   that that Intel the the leading CPU [TS]

01:00:10   vendor with its complex instruction sets [TS]

01:00:12   abused has been pursuing to try to reap [TS]

01:00:17   the advantages of the risk revolution [TS]

01:00:18   without actually having a RISC [TS]

01:00:20   instruction set on it CPUs so hold this [TS]

01:00:24   thought about risk versus cysts for a [TS]

01:00:26   moment that hopefully you have some at [TS]

01:00:29   least sort of feel for and we do a [TS]

01:00:31   sidebar into one other topic that will [TS]

01:00:34   come back and turn out to be relevant [TS]

01:00:35   all right so all the stuff we've been [TS]

01:00:39   talking about is you know for processor [TS]

01:00:41   so what is a processor and how do we [TS]

01:00:43   make it that turns out to be very [TS]

01:00:44   important you all know the process a [TS]

01:00:46   little black chip that we see in the [TS]

01:00:48   diagram is the flat thing it's got a [TS]

01:00:49   little metal contact somewhere on it and [TS]

01:00:51   inside that little black package is a [TS]

01:00:53   semiconductor it's like a piece of [TS]

01:00:55   silicon it's very flat and the way they [TS]

01:00:58   make stuff on it is they use lithography [TS]

01:01:02   to etch things onto it and with ography [TS]

01:01:04   is just basically shining light or some [TS]

01:01:06   other light like thing through a mask [TS]

01:01:09   that block some of the light and let [TS]

01:01:10   some through and they use materials and [TS]

01:01:14   techniques so that when the light hits a [TS]

01:01:16   certain area it changes the material in [TS]

01:01:17   one way and like doesn't hit the area it [TS]

01:01:19   leaves the material the other way and [TS]

01:01:21   they do this to make very small things [TS]

01:01:24   basically by using different kinds of [TS]

01:01:27   light and different ways to focus it and [TS]

01:01:30   you know other techniques they can make [TS]

01:01:33   very small features that no machine [TS]

01:01:35   could ever edge like the little tiny [TS]

01:01:37   things that are on these chips but like [TS]

01:01:39   that they they can focus a light another [TS]

01:01:41   I don't know what you would call it but [TS]

01:01:43   electromagnetic radiation into very [TS]

01:01:46   small areas and so they can make the [TS]

01:01:48   mask actually relatively big and then [TS]

01:01:49   use lenses and stuff to focus it down [TS]

01:01:51   and make different features they do this [TS]

01:01:52   again and again until they wear away the [TS]

01:01:53   material on certain things and what [TS]

01:01:55   they're basically trying to make is a [TS]

01:01:56   transistor think thing they can switch [TS]

01:01:57   on and off and they do this by having [TS]

01:01:59   doing layers of material that allow [TS]

01:02:01   ektron electrons to flow through them or [TS]

01:02:03   not depending on if some current is [TS]

01:02:04   applied to some other layer of material [TS]

01:02:05   but the point is that these things are [TS]

01:02:07   very very small and the way they [TS]

01:02:11   describe the size of the [TS]

01:02:13   things that are etched on here with [TS]

01:02:15   lithography is called the feature size [TS]

01:02:18   or sometimes the process node I tried to [TS]

01:02:21   look up a good definition to this we all [TS]

01:02:23   hear these numbers bandied about we've [TS]

01:02:24   talked about on this show when we say [TS]

01:02:26   like oh the they're using ad I Shrunk [TS]

01:02:29   version of the a5 in the ipad 2 and the [TS]

01:02:32   new ipad 2 is it used to be about forty [TS]

01:02:34   five nanometers and now it's 32 [TS]

01:02:36   nanometers alright so nanometers is how [TS]

01:02:38   they measure that's that's the scale [TS]

01:02:40   things are measured these days it used [TS]

01:02:41   to be was it a microns or micrometers [TS]

01:02:44   and remember what it was before we went [TS]

01:02:45   down to nanometers but they it nanometer [TS]

01:02:47   is very small right and what they're [TS]

01:02:50   measuring is the call the half pitch is [TS]

01:02:53   the half that half the distance between [TS]

01:02:54   identical features on the edge piece of [TS]

01:02:58   silicon usually for a memory cell [TS]

01:03:00   because memory cells are very very [TS]

01:03:01   regular doesn't mean that every single [TS]

01:03:04   thing on a 32 nanometer chip is 32 [TS]

01:03:08   nanometers wide or 32 nanometers apart [TS]

01:03:10   but it's like it's a they need some way [TS]

01:03:13   to measure these things because larger [TS]

01:03:14   features are chunkier and smaller [TS]

01:03:16   simpler features can be skinnier but [TS]

01:03:18   they've just come up with a single [TS]

01:03:19   standard for how we talk about how small [TS]

01:03:21   you can make stuff and that number is in [TS]

01:03:24   marching down for all of our lives you [TS]

01:03:25   used to be you could not make things [TS]

01:03:27   that were you know very small at all [TS]

01:03:29   compared to what we have now and then we [TS]

01:03:30   just keep getting smaller and smaller [TS]

01:03:31   and smaller and smaller every year and [TS]

01:03:33   that means the entire thing shrinks [TS]

01:03:34   because you can make every piece smaller [TS]

01:03:35   than the whole thing gets smaller and [TS]

01:03:36   smaller and smaller right so Ivy Bridge [TS]

01:03:40   processors that's Intel's processes that [TS]

01:03:42   in the macbook air is macbook pros [TS]

01:03:44   Intel's current crop of desktop and [TS]

01:03:48   laptop CPUs they are done at the 22 [TS]

01:03:52   nanometer process note and they have all [TS]

01:03:55   bunch of other features you know about [TS]

01:03:57   their 3d transistors different ways to [TS]

01:03:59   etch things onto the the silicon thing [TS]

01:04:01   so too can jam more stuff in a closed [TS]

01:04:03   the thing but bottom line is that there [TS]

01:04:04   are 22 nanometers right uh and in [TS]

01:04:09   general as you shrink this feature size [TS]

01:04:10   as you make everything smaller and [TS]

01:04:12   smaller almost everything improves the [TS]

01:04:14   cost improves tremendously because your [TS]

01:04:16   unit of cost is like how many of these [TS]

01:04:18   little chips can I fit onto a silicon [TS]

01:04:19   wafer so looking way for some big [TS]

01:04:21   circular thing maybe the size of like a [TS]

01:04:23   basketball or something you know that [TS]

01:04:26   and if you can fit 10 chips on there and [TS]

01:04:28   it costs you you know a hundred dollars [TS]

01:04:30   to make one of those wafers then each [TS]

01:04:32   chip cost you ten bucks to make but if [TS]

01:04:33   you can get a thousand ships on that [TS]

01:04:34   same way for suddenly the cost per chip [TS]

01:04:36   goes way down so shrinking things is [TS]

01:04:39   good because your costs are like I cost [TS]

01:04:41   me a certain amount to etch a single [TS]

01:04:43   piece of silicon and that costs like is [TS]

01:04:47   dependent on a whole bunch other factors [TS]

01:04:49   but the point is if I can jam more and [TS]

01:04:51   more cpus onto that piece of silicon the [TS]

01:04:53   cost of making it doesn't go go you know [TS]

01:04:55   doesn't go up I would love to be able to [TS]

01:04:58   put more things on here so the more [TS]

01:05:01   things you can spit on to a single [TS]

01:05:03   silicon wafer you know the better it is [TS]

01:05:05   for you right the power required to run [TS]

01:05:07   these dinky little things goes down [TS]

01:05:09   because if you can make them smaller and [TS]

01:05:11   smaller if you can imagine if you have [TS]

01:05:12   to run it chip the size of a dinner [TS]

01:05:13   plate a single chip it would take a lot [TS]

01:05:15   of power to run that thing when you make [TS]

01:05:17   them small you don't need as much power [TS]

01:05:18   uh and the speed goes up for variety of [TS]

01:05:22   reasons but even just like in it the at [TS]

01:05:25   the very high end of the scale look if [TS]

01:05:27   the thing was the size of a basketball [TS]

01:05:28   it takes a certain amount of time for an [TS]

01:05:30   electrical signal or propagate from the [TS]

01:05:31   upper left corner to lower right corner [TS]

01:05:32   whereas if the thing is the size of your [TS]

01:05:34   thumbnail that distance shrinks and when [TS]

01:05:36   you're going really really fast they may [TS]

01:05:38   seem like oh I'll don't the electrons [TS]

01:05:40   travel close to the speed of light and [TS]

01:05:41   doesn't make that well it makes a [TS]

01:05:42   difference when you're trying to you [TS]

01:05:43   know go billions of cycles per second or [TS]

01:05:46   whatever how r this one thing that that [TS]

01:05:49   is a problem when you may think smaller [TS]

01:05:51   you make them really really small [TS]

01:05:52   suddenly it becomes really easy for [TS]

01:05:53   electrons that you didn't want to go [TS]

01:05:55   somewhere just sort of just slip through [TS]

01:05:57   like you make things so small that just [TS]

01:05:59   random electrons kind of rattling around [TS]

01:06:01   end up sort of leaking through or you [TS]

01:06:03   didn't want them to go just because [TS]

01:06:04   everything is so small like when things [TS]

01:06:06   are big and chunky you can put a big [TS]

01:06:07   wall between uh no electrons can go [TS]

01:06:09   through here I'm totally putting down [TS]

01:06:10   this big wall but when things are really [TS]

01:06:11   really really tiny that wall is so [TS]

01:06:13   skinny that occasionally maybe a couple [TS]

01:06:14   electrons start floating through there [TS]

01:06:16   and that's called leakage current and [TS]

01:06:18   there are various techniques to try to [TS]

01:06:19   control that but that's when you start [TS]

01:06:22   getting really small like down to the [TS]

01:06:23   size of like you were you measuring [TS]

01:06:24   feature sizes and you know how many [TS]

01:06:26   atoms are there it between this area in [TS]

01:06:29   that area then you have some problems [TS]

01:06:31   but in general doing shrinks is better [TS]

01:06:33   there's another article again by John [TS]

01:06:36   Stokes also an older our goal talking [TS]

01:06:38   about understanding Moore's [TS]

01:06:39   aw Moore's Law is the thing we hear [TS]

01:06:42   thrown around in the world of computing [TS]

01:06:44   a lot that most light people i think [TS]

01:06:47   understand as Moore's law means [TS]

01:06:48   computers get faster and faster all the [TS]

01:06:50   time which is not what Moore's law [TS]

01:06:51   actually says it's like a third level [TS]

01:06:53   consequence of it but Moore's law was [TS]

01:06:55   set as saying as he was just noting that [TS]

01:06:56   in the early days of semiconductors the [TS]

01:06:59   number of transitions we can shove on to [TS]

01:07:01   a particular area keeps doubling every [TS]

01:07:02   12 months basically he was noting that [TS]

01:07:04   they they keep being able to shrink [TS]

01:07:05   their feature sizes so before on this [TS]

01:07:08   one silicon wafer we could fit 10 CPUs [TS]

01:07:10   then 12 months from now we can fit 20 [TS]

01:07:12   and after that we can put 40 in it and [TS]

01:07:14   12 months after that we can fit ad and [TS]

01:07:15   it just keeps going up and up and he did [TS]

01:07:16   the graph means like wow soon we're [TS]

01:07:18   going to be able to shove huge amount of [TS]

01:07:19   transistors on to yeah right now we can [TS]

01:07:21   only do chips with a thousand [TS]

01:07:24   transistors of the size of your [TS]

01:07:25   thumbnail soon we'll be able to do chips [TS]

01:07:26   with you know ten times that and [TS]

01:07:29   especially in the early days of [TS]

01:07:31   computing if you gave a chip designer [TS]

01:07:33   more transistors to work with they could [TS]

01:07:34   make a more powerful CPU beach like all [TS]

01:07:37   before I could I can only add two [TS]

01:07:38   numbers together at once but now I have [TS]

01:07:39   double the number of transistors I can [TS]

01:07:41   just add another a door right nice that [TS]

01:07:43   and that they can both go at the same [TS]

01:07:44   time and i can add four numbers together [TS]

01:07:45   at the same time isn't that great and i [TS]

01:07:48   can add more memory and more cache [TS]

01:07:50   memory and also the things that get [TS]

01:07:52   better with shrinking things so being [TS]

01:07:54   able to have more transistors at your [TS]

01:07:56   disposal has led to you know for the [TS]

01:08:00   first couple decades of cpu development [TS]

01:08:02   every year computers would get faster [TS]

01:08:04   and faster because those guys had more [TS]

01:08:06   cpus that they could deal with and you [TS]

01:08:07   know they could either make your your [TS]

01:08:09   thing your existing cpu cost half as [TS]

01:08:11   much or they can make a new cpu that [TS]

01:08:13   cost the same amount that goes twice as [TS]

01:08:14   fast and that was a great run for how [TS]

01:08:16   long it lasted but not in general the [TS]

01:08:19   number of transistors doesn't exactly [TS]

01:08:21   equal speed but in the early days it did [TS]

01:08:22   now this was during the the risk versus [TS]

01:08:27   sisk wars we're like oh we've got this [TS]

01:08:29   increasing number of transistors but [TS]

01:08:31   your instruction set is stupid and we [TS]

01:08:33   have a better one and you've got a [TS]

01:08:35   parole this decode hardware on there [TS]

01:08:36   everything and risk was technically [TS]

01:08:39   superior because it didn't you know they [TS]

01:08:40   said we found a way to make you know [TS]

01:08:42   you're doing a sewing machine and we've [TS]

01:08:43   got an electric motor and we can make [TS]

01:08:44   this go so fast look at this we've got a [TS]

01:08:46   1 gigahertz alpha processor you can't [TS]

01:08:48   get the 1 gigahertz because you can't [TS]

01:08:50   run your sewing machine that fast and so [TS]

01:08:52   tell guys had to go back to the drawing [TS]

01:08:53   board how how how the hell can we make [TS]

01:08:55   this thing go faster we've got to break [TS]

01:08:58   things things down into small pieces and [TS]

01:08:59   then do our little internal core that [TS]

01:09:01   runs fast then have the external part [TS]

01:09:02   track everything and like now the risk [TS]

01:09:04   guys were laughing at him right you know [TS]

01:09:06   you guys suck you made a bad choice you [TS]

01:09:08   know you were there early we come in [TS]

01:09:09   later we have a better solution we're [TS]

01:09:10   going to crush you the dec alpha the [TS]

01:09:12   PowerPC all sorts of killer chips from [TS]

01:09:15   these risc cpu vendors in like the the [TS]

01:09:19   90s we're going to put Intel out of [TS]

01:09:21   business now Intel smart and they do a [TS]

01:09:25   lot stuff with decoding their [TS]

01:09:26   instructions and stuff like that but the [TS]

01:09:27   real thing that saved Intel's bacon is [TS]

01:09:30   the aforementioned Moore's law that [TS]

01:09:33   every year you could fit more and more [TS]

01:09:36   transistors on to the same size chip ah [TS]

01:09:39   and that is relevant because as the [TS]

01:09:44   number of transistors per unit area [TS]

01:09:46   increases and you know as the number of [TS]

01:09:48   transistors available to given cost [TS]

01:09:49   increases the relative percentage of the [TS]

01:09:52   total transistor count used by all that [TS]

01:09:54   ugly x86 decoding and tracking hardware [TS]

01:09:56   becomes a smaller and smaller percentage [TS]

01:09:59   of the total alright so you can do a [TS]

01:10:01   complete x86 implementation in like [TS]

01:10:03   29,000 transistors like the 8086 at [TS]

01:10:06   29,000 transistors right obviously [TS]

01:10:08   modern CPUs need you know they're doing [TS]

01:10:13   more complicated stuff than the 8086 but [TS]

01:10:15   it goes goes to show the like how many [TS]

01:10:17   transistors do you need to take x86 [TS]

01:10:19   instructions crack them into micro [TS]

01:10:21   operations track them as they go through [TS]

01:10:23   the machinery and reassemble them in the [TS]

01:10:24   end it's more than 29,000 ah you know [TS]

01:10:28   and of course it's not it's not as [TS]

01:10:31   simple as that was but I've you bid [TS]

01:10:33   transistors the current like I've your [TS]

01:10:35   bridge line of CPUs have 1.4 billion [TS]

01:10:37   transistors they have billion [TS]

01:10:40   transistors in them that's a lot right [TS]

01:10:42   so what percentage of that 1.4 billion [TS]

01:10:45   transistors is dedicated to dealing with [TS]

01:10:46   x86 decode tracking and reassembly stuff [TS]

01:10:50   the the press what eventually happened [TS]

01:10:53   was yeah x86 was uglier and the other [TS]

01:10:56   market forces involved or whatever but [TS]

01:10:58   the the penalty for having a crappier [TS]

01:11:01   instruction set just started to shrink [TS]

01:11:03   and shrink and Rigby's transistors were [TS]

01:11:05   just falling out of the sky [TS]

01:11:06   it's like oh we gotta dedicate half-hour [TS]

01:11:07   chip to do deal with x86 okay now we [TS]

01:11:09   have to get a fourth ok now we're [TS]

01:11:10   dedicating eighth now dedicating a [TS]

01:11:12   sixteenth eventually it's like it [TS]

01:11:14   doesn't matter you know you know how [TS]

01:11:16   many CPUs it takes for us to deal with [TS]

01:11:19   x86 it is an insignificant percentage of [TS]

01:11:22   the total number of CPUs of transistors [TS]

01:11:25   in the CPU in fact most of the CPU like [TS]

01:11:28   half of it is cash at this point you [TS]

01:11:29   know but it's it doesn't matter that we [TS]

01:11:31   have an uglier instruction set because [TS]

01:11:33   of Moore's law and many other factors [TS]

01:11:34   but in a large part because of Moore's [TS]

01:11:38   law the x86 penalty for having to do all [TS]

01:11:40   that ugly stuff went away and the risks [TS]

01:11:43   iska is theoretical cool advantage of [TS]

01:11:45   you know we have a simpler instruction [TS]

01:11:47   set we don't have to deal ugly decoding [TS]

01:11:48   and coding was no longer relevant and [TS]

01:11:51   along the way in that process since it [TS]

01:11:54   was so important for Intel to be able to [TS]

01:11:56   have as many transistors as is at its [TS]

01:11:58   disposal as a as it wanted and also [TS]

01:12:01   because you know doing die shrinks have [TS]

01:12:03   many other advantages Intel got really [TS]

01:12:06   good at doing those shrinks like say [TS]

01:12:10   okay this year we're going to be a 65 [TS]

01:12:12   nanometer and next year we're going to [TS]

01:12:13   be at 45 and then you know for that [TS]

01:12:15   we're gonna be a 32 and they're just [TS]

01:12:16   like a shrink shrink shrink because we [TS]

01:12:17   want more and more transistors on our [TS]

01:12:19   chips because it's you know it's it [TS]

01:12:21   gives us an advantage we want to have as [TS]

01:12:23   many as we want it helped them out when [TS]

01:12:25   they were trying to decrease the burden [TS]

01:12:27   of the x86 burden because like so what [TS]

01:12:29   we got to do this decode stuff whatever [TS]

01:12:30   we just need more see more princess just [TS]

01:12:32   throw them on there right put on you [TS]

01:12:35   know increase caches on our chips to [TS]

01:12:37   make them faster all stuff like that [TS]

01:12:38   yeah so so while the x86 burden is [TS]

01:12:43   technically gross and upsetting to [TS]

01:12:46   techno purists like me it at this point [TS]

01:12:49   it has very little practical effect or [TS]

01:12:51   rather its practical effects are able to [TS]

01:12:54   be managed by Intel because then tells [TS]

01:12:56   really smart and they have a lot of [TS]

01:12:57   transistors and as mentioned earlier [TS]

01:12:59   Intel's currently shipping 22 nanometer [TS]

01:13:01   chips us there they're a feature size [TS]

01:13:03   all right now Apple's a six they're cool [TS]

01:13:07   fancy CPU that's in the iPhone 5 and the [TS]

01:13:10   iPad for that is a 32 nanometer chip and [TS]

01:13:15   32 is bigger than 22 and this highlights [TS]

01:13:18   something [TS]

01:13:19   for many years now but the past several [TS]

01:13:21   years intel has consistently been one [TS]

01:13:23   generation at least ahead of everybody [TS]

01:13:26   else on feature size so intel has been [TS]

01:13:28   shipping 22 nanometers ships for a long [TS]

01:13:29   time and apples a six which just came [TS]

01:13:31   out in the iphone five fairly recently [TS]

01:13:33   is 32 nanometer Intel is ahead and then [TS]

01:13:37   by the way it helps fit shipping [TS]

01:13:39   full-fledged CPUs at 22 nanometers with [TS]

01:13:41   1.4 billion transistors it's not even [TS]

01:13:43   close to what the a6 is right most chip [TS]

01:13:47   makers when they start off on a new [TS]

01:13:48   process node the Merse they make memory [TS]

01:13:49   because memory is very regular very [TS]

01:13:51   simple not complicated at all they just [TS]

01:13:53   make memory chips and after they've [TS]

01:13:55   sorted that out okay we can make 32 [TS]

01:13:56   nanometer memory chips let's try making [TS]

01:13:58   some more complicated things and [TS]

01:13:59   eventually they work their way up to big [TS]

01:14:00   complicated cpu intel is shipping GPUs [TS]

01:14:03   with billions of transistors at 22 [TS]

01:14:04   nanometers Apple for all its amazing [TS]

01:14:07   power and money is shipping 32 nanometer [TS]

01:14:09   chips and this is in a context where [TS]

01:14:10   like boy you know incredibly power [TS]

01:14:13   sensitive you want it to be small you [TS]

01:14:15   want to use not a lot of power wouldn't [TS]

01:14:17   it be great if apple cup of 22 nanometer [TS]

01:14:19   a sex why don't they because intel has [TS]

01:14:21   been ahead of everyone else because they [TS]

01:14:22   invest tons of money in it they have the [TS]

01:14:24   most experience and it's been a pretty [TS]

01:14:26   sustained advantage it's not like Oh a [TS]

01:14:28   one-year Intel's ahead then another year [TS]

01:14:29   AMD is ahead no intel has been the big [TS]

01:14:31   dog here they are making ships a [TS]

01:14:34   generation of head where everyone else [TS]

01:14:36   has been doing in terms of feature size [TS]

01:14:38   all which finally brings us back to [TS]

01:14:42   Apple switching to arm and this may be a [TS]

01:14:45   good time to do the final sponsor before [TS]

01:14:46   I before I bring this home happy to [TS]

01:14:48   oblige at squarespace.com everything you [TS]

01:14:52   need to make an amazing website we use [TS]

01:14:54   them for a lot of the blogging stuff [TS]

01:14:56   that's kind of what they're known for [TS]

01:14:58   that's kind of they got their start but [TS]

01:14:59   you can build entire websites you can [TS]

01:15:01   have multiple blogs on the web sites you [TS]

01:15:03   can have galleries they're basically [TS]

01:15:05   completely managed environment leidy [TS]

01:15:07   create and maintain maintain is the key [TS]

01:15:10   a website a blog portfolio what's cool [TS]

01:15:14   about Squarespace is you don't have to [TS]

01:15:15   know anything maybe your maybe all you [TS]

01:15:18   do is worry about risk and sisk [TS]

01:15:19   processors and you don't want to [TS]

01:15:21   understand how to build a site you can [TS]

01:15:22   use this drag and drop stuff but if you [TS]

01:15:24   are an uber geek and you'd like things [TS]

01:15:27   like writing code in HTML and [TS]

01:15:28   controlling JavaScript and you want to [TS]

01:15:30   control every single aspect of the site [TS]

01:15:32   you can do that with their templates you [TS]

01:15:35   just switch to developer mode and you [TS]

01:15:37   can use SFTP or you can use get to [TS]

01:15:39   publish your site it's super super cool [TS]

01:15:42   and you can go to developers not [TS]

01:15:43   squarespace com to learn more about that [TS]

01:15:45   all of their templates at least until [TS]

01:15:47   you bust a mob start as completely [TS]

01:15:50   responsive things they were structure [TS]

01:15:52   automatically when you look at them on [TS]

01:15:53   an iOS device what this means is you're [TS]

01:15:55   not gonna have to deal with serving up a [TS]

01:15:57   crummy mobile version of your site to [TS]

01:15:59   someone you send so you build it once [TS]

01:16:01   and it looks good and everywhere or just [TS]

01:16:03   use their templates are already built to [TS]

01:16:05   do that so here's the way this works [TS]

01:16:07   this is not free they give you 24 7 [TS]

01:16:10   customer support you got to pay for that [TS]

01:16:11   it's not something that just happens [TS]

01:16:14   they don't have a staff of people there [TS]

01:16:15   that you know for nothing but it's 10 [TS]

01:16:18   bucks a month actually it's less than [TS]

01:16:19   that if you use the coupon code which [TS]

01:16:21   I'll tell you about in a second you want [TS]

01:16:24   the unlimited plan its twenty bucks a [TS]

01:16:26   month you send up for a year you get [TS]

01:16:27   twenty percent off you sign up for two [TS]

01:16:29   years to get twenty-five percent off [TS]

01:16:30   some people just might be month month [TS]

01:16:32   that's fine too you don't have to even [TS]

01:16:34   give them a credit card to try it out [TS]

01:16:35   and that's what I would like for you to [TS]

01:16:36   do trying it out will support the show [TS]

01:16:38   go to squarespace com / 5 by 5 you don't [TS]

01:16:42   have to give them anything you just come [TS]

01:16:44   up with a little name for your site and [TS]

01:16:46   get going the code that you want to use [TS]

01:16:48   is Dan sent me 11 well a couple people [TS]

01:16:51   were saying oh dance at me 11 code [TS]

01:16:53   wasn't working it to working now dan [TS]

01:16:56   sent me 11 this is going to get you ten [TS]

01:16:58   percent off whether you do the month to [TS]

01:17:00   month or the year it's all cumulative so [TS]

01:17:03   go check them out support this show [TS]

01:17:05   squarespace com / 5 by 5 code is dance [TS]

01:17:09   at me 11 more details in the show notes [TS]

01:17:11   all right Apple switching to arm meaning [TS]

01:17:19   Apple stops putting Intel CPUs in their [TS]

01:17:22   macs and replaces would this be with [TS]

01:17:25   this is the arm CP would this be a [TS]

01:17:27   bigger move John then switch to Intel [TS]

01:17:30   back in what was it Oh 306 yes it would [TS]

01:17:33   be no I explain why so what are the [TS]

01:17:35   challenges facing Apple if they decide [TS]

01:17:37   to do this first thing they have to do [TS]

01:17:39   is they have to match the performance of [TS]

01:17:42   Intel CPUs maybe you don't have to [TS]

01:17:44   exceed them but you at least have to [TS]

01:17:45   match it because if you have into [TS]

01:17:46   cpus now and you're saying we don't want [TS]

01:17:48   them anywhere we want arm CPUs well you [TS]

01:17:50   don't want to make everything slower [TS]

01:17:51   right and you don't have to match the [TS]

01:17:53   current Intel CPUs you have to match the [TS]

01:17:56   upcoming Intel CPUs like you know you [TS]

01:17:59   have to look at intel's roadmap and say [TS]

01:18:00   well they say in this year they're going [TS]

01:18:01   all with the cpu can we make something [TS]

01:18:03   that's as has that kind of performance [TS]

01:18:05   Intel is really good at making CPUs [TS]

01:18:08   right Apple's asics is by all accounts [TS]

01:18:12   very good and nicely designed everything [TS]

01:18:14   but Apple didn't have to design the [TS]

01:18:16   instruction set itself it sort of [TS]

01:18:18   licensed that from arm and the stuff [TS]

01:18:21   that is doing at CPU it's a smaller CPU [TS]

01:18:23   it looks like it was hand laid out and [TS]

01:18:25   stuff like that but it is not as [TS]

01:18:27   ambitious as the crazy stuff that [TS]

01:18:28   entailed does routinely to make its x86 [TS]

01:18:31   chips fast because x86 is still ugly and [TS]

01:18:33   disgusting and they manage to make that [TS]

01:18:34   go fast they've got serious skills there [TS]

01:18:36   look at AMD the only other real viable [TS]

01:18:39   competitor to Intel back in the old [TS]

01:18:42   desktop server CPU wars they am d [TS]

01:18:45   invented the x86 64 instructions that [TS]

01:18:47   ball Intel was off screwing things up [TS]

01:18:50   with the Itanium AMD has lots of you [TS]

01:18:54   know has some skilled there for a while [TS]

01:18:55   AMD was faster and better than intel [TS]

01:18:58   back when Intel was in the net burst [TS]

01:18:59   pentium 4 days where they just [TS]

01:19:01   concentrate entirely on clock speed to [TS]

01:19:02   the detriment of everything else that [TS]

01:19:03   turned out to not be a good idea so AMD [TS]

01:19:05   was ahead for a while uh but these days [TS]

01:19:08   intel has come roaring back and got [TS]

01:19:10   their house in order and they've just [TS]

01:19:11   been trouncing AMD in terms of [TS]

01:19:14   performance for many years now uh and if [TS]

01:19:16   you look at like the Geekbench scores of [TS]

01:19:18   like you know people say what about [TS]

01:19:20   we'll just use the a6 in a macbook air [TS]

01:19:22   like I'm picking geekbench just because [TS]

01:19:23   it's a site that I could find you know [TS]

01:19:25   there's very difficult to benchmark CPUs [TS]

01:19:26   and say what makes one fast or whatever [TS]

01:19:28   but no matter how bad and synthetic [TS]

01:19:30   these benchmarks are at a certain point [TS]

01:19:32   it makes a difference so like the the [TS]

01:19:34   iphone 5 from the geekbench score is it [TS]

01:19:36   was like 1600 or something like that [TS]

01:19:39   under there their scores confuse the [TS]

01:19:41   ipad 3 someone said an ipad 3 with an a6 [TS]

01:19:43   but the ipad 3 doesn't have an asic so [TS]

01:19:45   that's also computing but anyway the [TS]

01:19:46   ipad the fastest ipad score i could find [TS]

01:19:48   was like 5,000 the macbook air scores [TS]

01:19:52   that I could find we're like 8,000 right [TS]

01:19:54   and then the Mac Pro scores for like [TS]

01:19:57   thirty six thousand so [TS]

01:19:59   the mac pro at the top of 36,000 and the [TS]

01:20:02   iphone 5 with an a6 at 1000 that's like [TS]

01:20:04   an order of magnitude difference they're [TS]

01:20:05   in range between what what these little [TS]

01:20:07   arm CPUs can do no one's saying you're [TS]

01:20:09   going to put an a6 into a mac but it [TS]

01:20:11   just goes to show that Apple has yet to [TS]

01:20:13   build an arm CPU that is performance [TS]

01:20:16   competitive with it so it's not like [TS]

01:20:17   they have something ready to go even [TS]

01:20:19   like the a-15 which is a the reference [TS]

01:20:23   design from RM CPUs or whatever there's [TS]

01:20:25   a long gap between anything any actual [TS]

01:20:27   shipping arm CPU and the fastest [TS]

01:20:29   shipping Intel CPUs so somehow Apple [TS]

01:20:32   would be it's their responsibility to [TS]

01:20:33   now bridge that gap because again you do [TS]

01:20:35   not want to ship the next generation of [TS]

01:20:37   MacBook Airs that are slower than the [TS]

01:20:38   current generation so you have to match [TS]

01:20:39   them that's hard to do right second and [TS]

01:20:42   this is the real killer you want to [TS]

01:20:44   match their power consumption because [TS]

01:20:45   the whole reason you're you know [TS]

01:20:47   supposedly going to arm is like oh we [TS]

01:20:49   use arm CPUs and our phones and they sit [TS]

01:20:50   battery and then we keep using these big [TS]

01:20:52   hot power-hungry Intel CPUs and our [TS]

01:20:55   MacBook Airs uh well you can't just take [TS]

01:20:58   the current crop of arm CPUs and stick [TS]

01:21:01   them to a Mac because they're way too [TS]

01:21:02   slow and if you can make them as fast [TS]

01:21:04   now you have to make them as fast and [TS]

01:21:05   make them you know and not make them [TS]

01:21:08   consume as much power as Intel's [TS]

01:21:10   upcoming line of stuff uh and one of the [TS]

01:21:13   biggest factors not the biggest factor [TS]

01:21:15   and how much power your CPU takes is the [TS]

01:21:17   process size and as previously mentioned [TS]

01:21:20   Intel is an entire generation ahead on [TS]

01:21:22   process while you are making 32 [TS]

01:21:25   nanometer chips Intel's shipping 22 [TS]

01:21:27   nanometer chips and that's a huge [TS]

01:21:28   advantage when it comes to to power [TS]

01:21:31   right so for desktops and laptops is it [TS]

01:21:35   possible to match or exceed Intel's [TS]

01:21:38   performance and its performance per watt [TS]

01:21:40   at a bigger process node like can you [TS]

01:21:42   make can you make a cpu at 30 nanometers [TS]

01:21:45   that has better performance and better [TS]

01:21:47   performance per watt than intel when [TS]

01:21:49   they get to use 22 nanometers probably [TS]

01:21:51   not that's probably pretty much a [TS]

01:21:52   physical impossibility no matter how [TS]

01:21:54   awesome you are and how in company intel [TS]

01:21:57   is an intel is not incompetent right and [TS]

01:21:59   and even if you could what would your [TS]

01:22:02   advantage be say okay i think that at 32 [TS]

01:22:05   nanometers i can make a chip that it's [TS]

01:22:06   better price performance than intel even [TS]

01:22:08   though they get to make all the chips of [TS]

01:22:09   22 nanometers right ah say you could do [TS]

01:22:12   that can you do it [TS]

01:22:13   lower costs you know because they can [TS]

01:22:16   fit way more chips onto a wafer and [TS]

01:22:18   they've been doing it for longer and by [TS]

01:22:20   the way you have to find someone to [TS]

01:22:22   manufacture this chip for you and with [TS]

01:22:24   the articles mentions I could use a TSMC [TS]

01:22:26   taiwan semiconductor manufacturing [TS]

01:22:27   company like you know dedicated [TS]

01:22:29   semiconductor foundry you give them a [TS]

01:22:31   design they manufacture it they're not [TS]

01:22:33   in the process they're not in the [TS]

01:22:34   business of making ships you just say [TS]

01:22:35   okay you've got a design fab it for me [TS]

01:22:37   make these chips again they're they're [TS]

01:22:39   consistently a generation behind what [TS]

01:22:41   Intel can produce an Intel does not [TS]

01:22:43   really rent out its boundaries to its [TS]

01:22:45   competitors oh yeah until it's not going [TS]

01:22:46   to say hey you want to make a chip [TS]

01:22:48   Sierra will make your arm chips for you [TS]

01:22:50   20 nanometers just give us a design and [TS]

01:22:52   we'll turn them out no that's not really [TS]

01:22:53   what they're interested in for obvious [TS]

01:22:54   reasons right when Apple switched from [TS]

01:22:57   PowerPC to Intel i wrote an article [TS]

01:23:00   about this i linked in the show notes in [TS]

01:23:02   general it was a relief because finally [TS]

01:23:05   apple fans and Apple itself could stop [TS]

01:23:09   worrying about oh god Intel's got some [TS]

01:23:10   new chips coming out and they're really [TS]

01:23:12   awesome and their generation ahead and [TS]

01:23:13   process node and they're there really [TS]

01:23:15   fast and we need power pc chips that can [TS]

01:23:18   compete with those and we have to go beg [TS]

01:23:19   IBM to make them for us or big motor [TS]

01:23:22   role to do something that's just finally [TS]

01:23:24   it was really fly right we're off that [TS]

01:23:26   bandwagon whatever Intel comes out with [TS]

01:23:27   we get to reap those benefits now [TS]

01:23:29   suddenly Apple users for the past I [TS]

01:23:31   don't know how many years it's been for [TS]

01:23:32   a long time now we've not had to worry [TS]

01:23:34   about sea views we've not had to worry [TS]

01:23:35   about Max or slow because they use [TS]

01:23:36   crappy CPUs they use the same ones that [TS]

01:23:38   everyone else which also happened to be [TS]

01:23:39   the best ones in terms of price and [TS]

01:23:41   performance which are from Intel going [TS]

01:23:43   back to arm would mean that you are back [TS]

01:23:45   on the hook of like oh god oh god [TS]

01:23:47   Intel's coming out with new chips we [TS]

01:23:48   need to we need to make something that's [TS]

01:23:52   as good as these new Intel chips and [TS]

01:23:53   Apple hurry up and designed that arm [TS]

01:23:55   chip that's gonna you know compete with [TS]

01:23:57   Intel chips and that horse race will be [TS]

01:23:59   back which would be exciting but also [TS]

01:24:00   depressing if Intel starts kicking our [TS]

01:24:02   butts uh one obvious question here is [TS]

01:24:04   like okay so if Intel's always ahead and [TS]

01:24:06   process note and they're making 22 [TS]

01:24:07   nanometer chips why don't all the [TS]

01:24:08   iphones and ipads use intel chips ah [TS]

01:24:11   that's a good question for the senior [TS]

01:24:13   leadership at intel because part of the [TS]

01:24:14   problem has been that thus far intel has [TS]

01:24:16   not well first of all when you get down [TS]

01:24:19   to small sizes suddenly that x86 burden [TS]

01:24:22   becomes more relevant again because when [TS]

01:24:24   you're making a 1.4 billion transistor [TS]

01:24:26   chip yeah the [TS]

01:24:27   number of transistors you need to deal [TS]

01:24:28   with x86 crap is a small percentage but [TS]

01:24:31   if you make a much much much much [TS]

01:24:33   smaller chip with far fewer transistors [TS]

01:24:35   like the a6 suddenly the excess stuff [TS]

01:24:38   there starts to count you're like oh [TS]

01:24:39   geez I would really like if I didn't [TS]

01:24:40   have to deal with the x86 stuff and the [TS]

01:24:42   arm chips don't have to deal with their [TS]

01:24:43   very simple very straightforward risk [TS]

01:24:45   machines that's why a 32 nanometer a6 [TS]

01:24:49   chip can even compete in the mobile [TS]

01:24:50   space because the x86 burden is relevant [TS]

01:24:53   they're like the best of both worlds [TS]

01:24:54   would be take apples a6 design and have [TS]

01:24:57   Intel phablet at 22 nanometers but that [TS]

01:24:59   doesn't exist because until it's not [TS]

01:25:00   going to do that for you and Intel is [TS]

01:25:02   trying to make wimpier and wimpy or CPUs [TS]

01:25:04   it's got its Adam line of processors and [TS]

01:25:05   they just keep going you know Intel's [TS]

01:25:07   going down down down scale and as it [TS]

01:25:09   goes down scale smaller chips fewer [TS]

01:25:11   transistors the x86 stuff starts to [TS]

01:25:13   become relevant again which must stick [TS]

01:25:14   and Intel's crawl but what else can you [TS]

01:25:16   do because that's what they've got until [TS]

01:25:18   used to have holdings in arm and that [TS]

01:25:20   divested itself from arm and dedicated [TS]

01:25:22   itself to x86 and as you know that's all [TS]

01:25:24   they show to about whether they're doing [TS]

01:25:26   the right move over there but the bottom [TS]

01:25:26   line is that you've got arm way down at [TS]

01:25:29   the low end with a very simple low power [TS]

01:25:31   chip design but a generation behind in [TS]

01:25:33   manufacturing not that our manufacturers [TS]

01:25:35   stuff but like whoever they're fabbing [TS]

01:25:36   them it's not Intel I mean you and Intel [TS]

01:25:38   trying to bring it's it's incredibly [TS]

01:25:40   successful desktop products down market [TS]

01:25:42   and mostly failing to do so and that's [TS]

01:25:45   that's the pensioner that's the gap we [TS]

01:25:46   have right now so the obvious [TS]

01:25:50   explanation that I see it for this [TS]

01:25:52   floated rumor about Apple going to Intel [TS]

01:25:55   and this is you know what everyone [TS]

01:25:56   thinks as soon as they see it it's like [TS]

01:25:57   apples not ditching Intel there's [TS]

01:26:00   nothing viable they can move to this is [TS]

01:26:01   just a negotiating ploy so that Apple [TS]

01:26:03   can get good prices out of Intel for its [TS]

01:26:05   next generation of chips or so that [TS]

01:26:07   Apple can influence Intel and say until [TS]

01:26:09   we really want you to do XY and Z and [TS]

01:26:11   make him tell us into them because you [TS]

01:26:13   know they all ride we heard you might go [TS]

01:26:14   to arm don't do that let's kiss and make [TS]

01:26:17   up right ah that's kind of like when [TS]

01:26:21   Apple was talking to Intel for years and [TS]

01:26:23   years when he was shipping powerpc macs [TS]

01:26:25   part of us like oh they're just talking [TS]

01:26:26   to Intel because they're trying to force [TS]

01:26:28   IBM or motorola or whoever was at that [TS]

01:26:30   time to make them the chip that they [TS]

01:26:31   wanted they're not actually going to go [TS]

01:26:32   to Intel but they have to keep talking [TS]

01:26:34   with Intel because otherwise you know [TS]

01:26:37   IBM won't make the g5 for Apple right [TS]

01:26:39   it's all a negotiating ploy [TS]

01:26:41   but the thing is eventually that [TS]

01:26:43   negotiating ploy apparently stopped [TS]

01:26:45   working uh and I've actually did go to [TS]

01:26:48   Intel and so that that type of [TS]

01:26:50   phenomenon over the course of many many [TS]

01:26:52   years could happen here where these arm [TS]

01:26:54   you know bringing the next arm rumor [TS]

01:26:56   keeps getting floated and floated and [TS]

01:26:57   they keep using it is negotiating [TS]

01:26:59   leverage at a certain point until calls [TS]

01:27:01   they're bluffing says you know what [TS]

01:27:02   we've got to take care of our own [TS]

01:27:04   business oh you know we've got to do [TS]

01:27:06   something and yeah the chips were going [TS]

01:27:07   to make are perfectly suited to you but [TS]

01:27:09   you're just gonna have to deal with it [TS]

01:27:10   maybe until it does leave them and go to [TS]

01:27:11   arm and starts up that race again uh I [TS]

01:27:14   don't know final couple closing thoughts [TS]

01:27:17   here yes Intel's market cap is a as of [TS]

01:27:22   the time of a these show notes 1.4 [TS]

01:27:24   billion Apple has more cash than that [TS]

01:27:28   like in its bank account at some of its [TS]

01:27:30   long-term you know not not immediately [TS]

01:27:32   liquid or whatever but could Apple by [TS]

01:27:35   Intel probably ever really wanted to if [TS]

01:27:39   you wanted to spend all it's like [TS]

01:27:40   spending you know you get you you save [TS]

01:27:41   money for years and years and years and [TS]

01:27:43   you spend it all in one spot yeah Apple [TS]

01:27:45   could probably try to buy Intel [TS]

01:27:46   antitrust would be a problem there or [TS]

01:27:50   whatever but I don't know of a parolee [TS]

01:27:52   wants to be in that business apple [TS]

01:27:53   doesn't buy like the people who make its [TS]

01:27:55   products it you know contracts it out [TS]

01:27:57   and maybe buys hardware for them and [TS]

01:27:59   funds the development of their thing but [TS]

01:28:00   they don't want to be in the business [TS]

01:28:01   they want someone else to deal with it [TS]

01:28:02   the ugly low-margin part of the business [TS]

01:28:04   they just want the high Martin so I [TS]

01:28:05   don't think apples going to buy Intel ah [TS]

01:28:07   apple's market cap is currently five [TS]

01:28:09   hundred billion and rapidly dropping uh [TS]

01:28:12   because of their stock being slam [TS]

01:28:13   recently but there's this I mean I'm [TS]

01:28:16   yeah I know you're not a stock analysis [TS]

01:28:18   analyst then you won't do an analysis on [TS]

01:28:20   the show but uh is this a good time to [TS]

01:28:23   buy Apple stops I don't know I'm not [TS]

01:28:26   gonna you won't even go into it you [TS]

01:28:27   won't do it I I don't know if it's a [TS]

01:28:30   good time the stock market I don't [TS]

01:28:31   underst no one understands the stock [TS]

01:28:32   market I don't know sorry certainly it's [TS]

01:28:34   better to buy now than it was the by [TS]

01:28:35   Mary was 700 dollars but I don't know I [TS]

01:28:38   don't know I can I don't know why the [TS]

01:28:40   stock is going down I don't know why it [TS]

01:28:42   keeps going down so if you don't know [TS]

01:28:43   that's probably a bad time to buy [TS]

01:28:45   anything if you don't understand what's [TS]

01:28:46   happened I don't understand what's [TS]

01:28:47   happening so I give you know advice [TS]

01:28:49   about whether you should buy it or not [TS]

01:28:50   but so I'll take that as a strong [TS]

01:28:52   recommendation to buy [TS]

01:28:54   yeah but anyway that's okay that's the [TS]

01:28:56   situation Apple is in I don't think it [TS]

01:29:00   can feasibly go to arm CPUs in its max [TS]

01:29:02   unless it dedicates a huge amount of [TS]

01:29:05   money and time over the course of the [TS]

01:29:07   next several years to coming up with [TS]

01:29:09   something that can compete with intel [TS]

01:29:10   intel has its own problems because many [TS]

01:29:13   many more mobile cpus are being bought [TS]

01:29:15   then desktop CPUs and that ratio keeps [TS]

01:29:17   getting worse and worse for them so [TS]

01:29:18   intel has to get into the mobile space [TS]

01:29:20   somehow it seems to me though that these [TS]

01:29:22   two forces should lead to an arrangement [TS]

01:29:24   whereby Apple and Intel's interests [TS]

01:29:28   become aligned at some point until once [TS]

01:29:30   in the mobile space Apple wants [TS]

01:29:32   processors a certain way Apple talks to [TS]

01:29:34   Intel eventually they come to some sort [TS]

01:29:35   of agreement and this arm stuff is just [TS]

01:29:37   noise right but on the other hand like [TS]

01:29:39   you know Bob Mansfield coming back and [TS]

01:29:41   have big plans for the semiconductor [TS]

01:29:42   division or whatever it could just be [TS]

01:29:45   good business just like it was good [TS]

01:29:47   business to constantly be talking with [TS]

01:29:48   Intel and constantly be having that [TS]

01:29:50   version of you know Mac OS 10 for Intel [TS]

01:29:52   in the labs or whatever it's probably [TS]

01:29:55   good business for Apple to at least [TS]

01:29:56   investigate what would it take to get an [TS]

01:29:58   arm CPU design they'll be competitive [TS]

01:29:59   with Intel and how many years would it [TS]

01:30:01   take how much money would it take and [TS]

01:30:02   let's let's have that project and that's [TS]

01:30:05   also consider how much would it take if [TS]

01:30:06   we just paid Intel to make our chips for [TS]

01:30:08   us and can we convince in tell toofab [TS]

01:30:10   our stuff at 22 nanometers can Apple [TS]

01:30:12   itself invest enough money to suddenly [TS]

01:30:14   match Intel's ability to fab chips you [TS]

01:30:17   know and their expertise probably not [TS]

01:30:19   like many options are on the table but i [TS]

01:30:22   think this this entire story at this [TS]

01:30:24   point is like this is something to watch [TS]

01:30:26   in the same way that you should have [TS]

01:30:27   been watching those silly rumors about [TS]

01:30:29   Apple talking to Intel or talking to AMD [TS]

01:30:31   for years and years and years before [TS]

01:30:32   they actually did the switch it's like [TS]

01:30:34   it's easy to dismiss that stuff but i [TS]

01:30:36   think it's it's good idea just keep an [TS]

01:30:38   eye on that because Apple has many [TS]

01:30:39   options here and none of them are really [TS]

01:30:41   that good and same with Intel they have [TS]

01:30:42   many options and not really that good [TS]

01:30:44   either but I think what I think is that [TS]

01:30:45   eventually if Apple and Intel keep going [TS]

01:30:50   along the current path their interest [TS]

01:30:52   will be sufficiently aligned such that [TS]

01:30:53   they will become be able to come to some [TS]

01:30:55   mutually beneficial agreement that will [TS]

01:30:57   not involve Apple 100% jilting them and [TS]

01:31:00   making its own arm processors ah because [TS]

01:31:02   I think that is a really really tall [TS]

01:31:04   order and the biggest thing against it [TS]

01:31:07   is because they [TS]

01:31:07   tells still ahead of everyone by a [TS]

01:31:09   generation and processed note if that [TS]

01:31:11   changes if suddenly everyone else [TS]

01:31:13   catches up the Intel and Apple can get [TS]

01:31:14   its chips fabbed at the same size as [TS]

01:31:16   Intel can then that's a whole new [TS]

01:31:18   ballgame but like I said for years and [TS]

01:31:20   years intel has been ahead and it seems [TS]

01:31:21   to be a sustainable advantage on their [TS]

01:31:23   on their part and they invest tremendous [TS]

01:31:25   amount of money billions billions of [TS]

01:31:26   dollars into maintaining that advantage [TS]

01:31:28   very few other companies have the kind [TS]

01:31:31   of money to invest in that or the [TS]

01:31:32   expertise or the people Apple probably [TS]

01:31:35   has the money but probably doesn't have [TS]

01:31:36   the expertise and I don't think they can [TS]

01:31:37   hire away all of them tells people so [TS]

01:31:39   I'll let us say keep watching this space [TS]

01:31:43   but for now don't worry next year's crop [TS]

01:31:46   of max will not have ARM processors in [TS]

01:31:48   them or at least all the moon anyway [TS]

01:31:50   maybe one of them will who knows but [TS]

01:31:54   yeah don't hold your breath all right [TS]

01:31:58   even if you say so I say so pretty good [TS]

01:32:03   argument it's not much of an argument [TS]

01:32:06   there little background and then an area [TS]

01:32:08   to watch do you expect on arm max next [TS]

01:32:13   year not next year ner max a single [TS]

01:32:16   pilot program arm Mac and experimental [TS]

01:32:18   one just one nothing behind the scenes [TS]

01:32:21   behind closed doors I'm sure but you [TS]

01:32:24   know here if you think about what [TS]

01:32:26   everything that Apple has done as far as [TS]

01:32:29   wanting to control their own destiny the [TS]

01:32:34   move to Intel is in large part so that [TS]

01:32:38   they could have more control that was [TS]

01:32:41   the whole issue that they had with power [TS]

01:32:43   pc chips is they could never get [TS]

01:32:45   anything fast they switched until they [TS]

01:32:47   get the day get the fast stuff right and [TS]

01:32:49   not so much because they controlled [TS]

01:32:51   Intel but just because Intel's interest [TS]

01:32:53   was to make the best jede falsche stuff [TS]

01:32:55   yeah right and so now they were never [TS]

01:32:57   second leg Intel Intel will you please [TS]

01:33:00   make us good desktop making good desktop [TS]

01:33:02   processors right we can't let the big [TS]

01:33:03   them right but now it's getting to the [TS]

01:33:04   point where Apple's like Intel please [TS]

01:33:06   like I know you want to make these [TS]

01:33:07   processors but we need very very [TS]

01:33:09   specific things we want to retina [TS]

01:33:11   display can you build in the GPU can [TS]

01:33:13   drive a Retina display all this noise [TS]

01:33:15   i'm on the intel HD graphics 3000 just [TS]

01:33:17   barely or 4000 we're just barely can [TS]

01:33:19   drive a Retina display but we totally [TS]

01:33:21   wanted to [TS]

01:33:21   reten everywhere can you please make [TS]

01:33:23   your in ship GPU better and was like I [TS]

01:33:25   know that's not really all not a [TS]

01:33:27   priority for us right now yeah so [TS]

01:33:29   suddenly Apple's having to beg Intel a [TS]

01:33:32   little bit but yeah going away from [TS]

01:33:33   power pc is like finally we don't the [TS]

01:33:34   beggining one for a CBO they they still [TS]

01:33:36   by the way i will use the opportunity to [TS]

01:33:38   get as much leverage as I possibly could [TS]

01:33:39   an Intel like you know until is gonna [TS]

01:33:41   make chips just for us and we have a [TS]

01:33:43   site member they made that crazy shrink [TS]

01:33:46   version for the original macbook air [TS]

01:33:47   like they got Intel to make like a [TS]

01:33:49   custom chip just for them it's like a [TS]

01:33:51   jam a cpu I think it was the original [TS]

01:33:53   macbook air like Apple has been using [TS]

01:33:55   its leverage like you know at first it [TS]

01:33:57   was like don't you want us we're like a [TS]

01:33:59   trophy brand you can say your stuff is [TS]

01:34:00   in Apple CPUs and then it was like oh we [TS]

01:34:03   sell a lot of intel cpus and your other [TS]

01:34:05   you know and now it's like apples like [TS]

01:34:07   you know how you have any idea how many [TS]

01:34:09   mobile cpus we sell if you want any part [TS]

01:34:11   of that business ever you better listen [TS]

01:34:12   to us I know you're not getting any of [TS]

01:34:14   it now but someday you know talk to us [TS]

01:34:16   maybe you can convince us to change all [TS]

01:34:17   of our iphones and ipads and iOS devices [TS]

01:34:19   to x86 if you've ever maked these an x86 [TS]

01:34:21   chip you know like the negotiations are [TS]

01:34:24   getting more strained between those two [TS]

01:34:26   companies but i have to think that [TS]

01:34:29   eventually they will their heads will [TS]

01:34:32   come together and we'll get something [TS]

01:34:33   good out of this well let's say you know [TS]

01:34:35   what I mean I think Apple like you said [TS]

01:34:36   Apple would be thrilled if Intel would [TS]

01:34:38   just do that kind of thing but then each [TS]

01:34:40   if you think about it each of the things [TS]

01:34:42   that Apple relies on for from other [TS]

01:34:45   vendors that it doesn't control that it [TS]

01:34:47   feels that it's like there's this rumor [TS]

01:34:51   that just came out that the new imacs [TS]

01:34:54   are going to be delayed because of a [TS]

01:34:56   welding issue this is something that [TS]

01:34:58   Apple pretty much can can control so [TS]

01:35:00   that's their fault but if they can't get [TS]

01:35:02   enough screens you know that that that's [TS]

01:35:05   got a bug now we can't get enough [TS]

01:35:06   screens if they can't get a cpu that's [TS]

01:35:08   gonna do what they want if they can't [TS]

01:35:10   offload graphics to the cpu the way that [TS]

01:35:11   they want these are things that you [TS]

01:35:13   would think apple's gonna say well we've [TS]

01:35:14   got some money let's let's just do this [TS]

01:35:18   ourselves I don't like that they want to [TS]

01:35:21   mean Samsung as we haven't you mentioned [TS]

01:35:23   like they don't want to pay Samsung now [TS]

01:35:25   to make all the ASIC cpus and they're [TS]

01:35:27   like that's why they've been pulling [TS]

01:35:29   this stuff in-house like we don't know [TS]

01:35:30   I'm Jack may be five we want to make a [TS]

01:35:32   custom design ship and we want to be [TS]

01:35:34   able to pay anyone to fab [TS]

01:35:35   think the ax is not just being fat by [TS]

01:35:37   sams I may be wrong about this but yeah [TS]

01:35:39   but paying Samsung to make your cpu's [TS]

01:35:41   that's bad Samsung is the only other [TS]

01:35:43   person making money in the smartphone [TS]

01:35:44   market you don't want to be paying them [TS]

01:35:46   any money so yeah I can imagine Apple [TS]

01:35:48   wanting to go elsewhere for that [TS]

01:35:49   expertise but like they can't just they [TS]

01:35:52   can't just leave Samsung immediately [TS]

01:35:53   there's a certain you know can you get [TS]

01:35:55   what Samsung gives you from someone else [TS]

01:35:57   at the same price and if the answer is [TS]

01:35:58   no you got to grit your teeth and keep [TS]

01:35:59   going to Samsung all the while Sam we [TS]

01:36:01   got to have a plan to get away from [TS]

01:36:02   Samsung like wooden tap beloved oocyte [TS]

01:36:04   pleat like if Apple can get in tell [TS]

01:36:06   toofab the a7 in its best process node [TS]

01:36:09   at that time that would be a humongous [TS]

01:36:12   victory but it is not an Intel strategic [TS]

01:36:14   interest to become like a fab four other [TS]

01:36:16   people ship designs right so the you [TS]

01:36:19   know most in need to come together on [TS]

01:36:21   something we want we want your [TS]

01:36:22   technology we don't want your chips then [TS]

01:36:24   tells like we want to sell our chips [TS]

01:36:25   because becoming a fab is a low-margin [TS]

01:36:27   business and we get much better margins [TS]

01:36:28   by making our own you know so I don't [TS]

01:36:32   know this I would love to beat see these [TS]

01:36:35   negotiations because they're happening [TS]

01:36:36   now like what is the a7 gonna be who's [TS]

01:36:38   gonna make it and where is it going to [TS]

01:36:41   be shipped like maybe that maybe it's [TS]

01:36:43   too late for those and those discussions [TS]

01:36:44   have already taken place right but for [TS]

01:36:45   the one after that for the a8 and for [TS]

01:36:47   any possible like arm and Intel arm and [TS]

01:36:51   Max type negotiations those have to be [TS]

01:36:52   happening now like we need a plan we [TS]

01:36:54   need to talk to people we know what we [TS]

01:36:56   don't want to do we don't want to give [TS]

01:36:57   Samsung any more money and we also don't [TS]

01:37:00   want to use any of the chips that [TS]

01:37:02   Intel's currently offering us because [TS]

01:37:03   they're way too power hungry we want ra8 [TS]

01:37:06   design manufactured in the best [TS]

01:37:09   manufacturing process in the entire [TS]

01:37:10   world uh at a really cheap price and [TS]

01:37:13   Intel doesn't want that they want [TS]

01:37:15   something different so yeah those they [TS]

01:37:17   need to come together on something but I [TS]

01:37:19   you know the only the only way i can [TS]

01:37:20   imagine apple going full split off is if [TS]

01:37:22   they have like a concrete plan where [TS]

01:37:25   they really believe that they can make a [TS]

01:37:27   line of max that is not embarrassingly [TS]

01:37:30   slower or crappier than what they would [TS]

01:37:31   have been if they had intel cpus and [TS]

01:37:33   intel is going to do everything in its [TS]

01:37:35   power to make sure that apple doesn't [TS]

01:37:37   never even thinks that's possible like [TS]

01:37:39   they just need to make apple like so [TS]

01:37:41   despair for their ability to match intel [TS]

01:37:43   cpu so they won't even attempt it [TS]

01:37:45   because I until will show them their [TS]

01:37:46   roadmap and say you're never going to [TS]

01:37:47   have [TS]

01:37:48   it's gonna compete with this look at [TS]

01:37:49   this do you this is art this is our road [TS]

01:37:50   map that's what they did to get apples [TS]

01:37:52   business they said yeah current line of [TS]

01:37:54   CPUs suck but look at this upcoming lot [TS]

01:37:56   of core processors they are awesome they [TS]

01:37:58   are really fast they're low power and [TS]

01:38:00   that was all true and Intel delivered on [TS]

01:38:03   that commitment like they you know back [TS]

01:38:05   then was like why doesn't happen with [TS]

01:38:06   AMD don't they make better CPUs and [TS]

01:38:07   stuff but Intel shows in their roadmap [TS]

01:38:09   Apple said fine you're going to make [TS]

01:38:10   that then go make that and Intel did go [TS]

01:38:13   make that and until makes amazing CPUs [TS]

01:38:14   that powered you know you know decades [TS]

01:38:16   worth of great intel macs or ever many [TS]

01:38:19   years has been but now we're coming to [TS]

01:38:20   that point again and so intel does have [TS]

01:38:22   credibility here they come to apple and [TS]

01:38:23   say look yeah don't even think about [TS]

01:38:26   trying to make your arm cpus can you [TS]

01:38:27   compete with this you can't match us [TS]

01:38:29   just you know we need to come to an [TS]

01:38:30   agreement here don't even try you're [TS]

01:38:31   gonna you're gonna spend billions and [TS]

01:38:33   billions of dollars and it's not gonna [TS]

01:38:35   work out for you just look out what [TS]

01:38:36   happened to AMD we crush them don't even [TS]

01:38:38   attempt it but we're friends let's be [TS]

01:38:39   friends business it's like war but with [TS]

01:38:46   money keep your friends close enemies [TS]

01:38:48   closer yes and your frenemies i don't [TS]

01:38:52   know where you keep them but like I [TS]

01:38:53   guess somewhere in the middle all right [TS]

01:38:57   even wrap this up think we're done all [TS]

01:38:59   right so you can go to five by five TV [TS]

01:39:01   such hypercritical sash 94 and you will [TS]

01:39:03   get they will get all of the notes and [TS]

01:39:07   links that jon is put there you can [TS]

01:39:09   learn about sisk and risk on other [TS]

01:39:12   things you can follow John on Twitter he [TS]

01:39:15   is siracusa s I RAC USA USA and I'm damn [TS]

01:39:23   Benjamin on Twitter he is also on Alpha [TS]

01:39:25   app.net at Siracusa I'm Dan over there [TS]

01:39:28   and I guess I said you don't wanted me [TS]

01:39:33   to talk about the other one cuz you're [TS]

01:39:35   not using I used tent that is 10 thought [TS]

01:39:39   I oh whatever I do all right so he's [TS]

01:39:42   Syracuse over there too we appreciate [TS]

01:39:44   you listening if you would like you can [TS]

01:39:46   rate the show in iTunes you can [TS]

01:39:48   visit the sponsors it's a wonderful way [TS]

01:39:50   to help support the show in the hood [TS]

01:39:51   network we thank you very much for [TS]

01:39:53   listening and John will see next week [TS]

01:39:55   about [TS]

01:40:01   [Music] [TS]