Jay Rose Posted June 30, 2018 Report Share Posted June 30, 2018 Fascinating NYTimes article on - the need for nanosecond sync precision between distant computers connected via internet; - why NASDAQ is spearheading the effort (and has it working); - how it's useful for things as widely afield as World Cup Soccer; - how this has been a goal since the 1960s! https://nyti.ms/2Izysff Quote Link to comment Share on other sites More sharing options...
Wandering Ear Posted June 30, 2018 Report Share Posted June 30, 2018 Cool article. 1/100 Billionth of a second is a huge leap in accuracy. For comparison, Dante uses PTP (also used by some stock exchanges currently) which is accurate to 1 Microsecond (1/1,000,000 of a second). Quote Link to comment Share on other sites More sharing options...
jpb2 Posted June 30, 2018 Report Share Posted June 30, 2018 I have always wondered why we don't use the Schumann resonance as the heartbeat of the world. We could easily adjust for the variables given that they are global. I'll bet some of you did not know the world produces a beat every 7.8 ms Not to mention it would be a part of US.... what we feel naturally (we cant hear that low but we sure can feel it!) https://www.sedonanomalies.com/schumann-resonance.html Quote Link to comment Share on other sites More sharing options...
Dalton Patterson Posted July 1, 2018 Report Share Posted July 1, 2018 I hope Rolex knows about this. I want a billionth/second Oyster GMT. Quote Link to comment Share on other sites More sharing options...
LarryF Posted July 1, 2018 Report Share Posted July 1, 2018 On 6/30/2018 at 11:07 AM, jpb2 said: I have always wondered why we don't use the Schumann resonance as the heartbeat of the world. We could easily adjust for the variables given that they are global. I'll bet some of you did not know the world produces a beat every 7.8 ms Not to mention it would be a part of US.... what we feel naturally (we cant hear that low but we sure can feel it!) https://www.sedonanomalies.com/schumann-resonance.html Not to be a PITA but 7.8 ms is 128 Hz, easily hearable. I think it's 7.8 Hz or a pulse every 128 ms. OK, now I've gotten my Sunday PITA cravings out of the way. Best Regards, Larry Fisher Quote Link to comment Share on other sites More sharing options...
Shastapete Posted July 1, 2018 Report Share Posted July 1, 2018 54 minutes ago, LarryF said: I've gotten my Sunday PITA cravings out of the way. Best Regards, Larry Fisher Now I want hummus Quote Link to comment Share on other sites More sharing options...
Constantin Posted July 1, 2018 Report Share Posted July 1, 2018 On 6/30/2018 at 7:07 PM, jpb2 said: I have always wondered why we don't use the Schumann resonance as the heartbeat of the world. We could easily adjust for the variables given that they are global. I'll bet some of you did not know the world produces a beat every 7.8 ms Not to mention it would be a part of US.... what we feel naturally (we cant hear that low but we sure can feel it!) https://www.sedonanomalies.com/schumann-resonance.html It‘s 7.8 Hz, as Larry stated. There is no indication that this beat occurs at precise intervals, or even frequencies. And it varies depending on the time of year and on certain weather phenomena, such as lightning Quote Link to comment Share on other sites More sharing options...
Jay Rose Posted July 2, 2018 Author Report Share Posted July 2, 2018 Wouldn't the resonant frequency of the planet change with things like cyclic gravitational changes (i.e., lunar), density shifts from polar ice melting and subsequent evaporation, and even the burning of carbon extracted from under ground? Besides, we're talking a requirement for nanosecond precision among computers. 7.8 Hz seems precise only to a hundred ms. Quote Link to comment Share on other sites More sharing options...
IronFilm Posted July 3, 2018 Report Share Posted July 3, 2018 Many of these high frequency traders are so obsessed with speed that they'll even position their offices close to the stock exchanges so as to minimize the travel time their internet connection has to go for a trade to be executed! Teeny tiny teeny tiny fractions of a second matter to these guys. Quote Link to comment Share on other sites More sharing options...
Jim Feeley Posted July 3, 2018 Report Share Posted July 3, 2018 Ya, Michael Lewis talks about all this in his very-readable book, Flash Boys: http://michaellewiswrites.com/#flash-boys By "all this" I mean high-frequency trading. But back in the days a few years ago before the latest round of "even faster" "needs." There was some pushback against Lewis's book and apparently he did make a couple basically small mistakes, but a couple friends who are experts at this stuff (as in: one writes automated trading code, another is a biz professor) think a lot of the pushback was basically interested parties trying to discredit the book because it exposed some kinda crappy --but fascinating-- behavior. Here's what Lewis wrote about some of the response to the book. He's a very entertaining writer, btw: https://www.vanityfair.com/news/2015/03/michael-lewis-flash-boys-one-year-later And to keep this kinda close to JWS, there's a film being made from the book: https://variety.com/2018/film/news/michael-lewis-flash-boys-netflix-1202814774/ Jay, thanks for the link to the article! Quote Link to comment Share on other sites More sharing options...
Jay Rose Posted July 3, 2018 Author Report Share Posted July 3, 2018 FWIW, the point of ns precision is so trades can be accurately time-stamped, and then executed in order no matter how much latency exists between the trader and the exchange. Will it work? Will sharks find a way to game the system? Yes and yes... Quote Link to comment Share on other sites More sharing options...
borjam Posted July 3, 2018 Report Share Posted July 3, 2018 1 hour ago, Jay Rose said: FWIW, the point of ns precision is so trades can be accurately time-stamped, and then executed in order no matter how much latency exists between the trader and the exchange. Will it work? Will sharks find a way to game the system? Yes and yes... Actually it's two different problems The first one is the synchronization accuracy. The second one is a trusted timestamping mechanism that can't be played by the users. I can imagine some cryptographic approaches, but it's a hard problem and huge resources will be devoted to subvert it. Quote Link to comment Share on other sites More sharing options...
mono Posted July 3, 2018 Report Share Posted July 3, 2018 https://sniperinmahwah.wordpress.com/ https://www.facebook.com/auroratomahwah/ Quote Link to comment Share on other sites More sharing options...
IronFilm Posted July 4, 2018 Report Share Posted July 4, 2018 10 hours ago, Jim Feeley said: apparently he did make a couple basically small mistakes Maybe the small mistakes were intentional so as to mislead any potential competitors. 10 hours ago, Jim Feeley said: And to keep this kinda close to JWS, there's a film being made from the book: https://variety.com/2018/film/news/michael-lewis-flash-boys-netflix-1202814774/ Awesome! I'll look forward to seeing this movie. I have a Math Degree plus a heavy background in Computer Science, so any of these technical trading kind of news do interest me. Quote Link to comment Share on other sites More sharing options...
Jim Feeley Posted July 4, 2018 Report Share Posted July 4, 2018 22 hours ago, mono said: Thanks Mono! That's pretty cool! 13 hours ago, IronFilm said: Awesome! I'll look forward to seeing this movie. I highly recommend the book. Lewis is a former bonds salesman at Solomon Brothers (which he wrote about in Liar's Poker); that made him famous so he has some knowledge of the biz and good access to people. Also, he's an AMAZING writer. Insightful, dramatic, really funny. Several of his books have become movies (Moneyball, The Blind Side, The Big Short), but really, his books are super readable. His Bio: http://michaellewiswrites.com/#top Quote Link to comment Share on other sites More sharing options...
mono Posted February 5, 2019 Report Share Posted February 5, 2019 Quote Link to comment Share on other sites More sharing options...
Jay Rose Posted February 5, 2019 Author Report Share Posted February 5, 2019 Reporter (and stock exchange paying for this PR) says this is inherently safer than a digital delay, because computers can be compromised. A straight digital delay can be built with no connection between data and control, and its control isolated from the web. So you'd need physical access to compromise it... the same way you could with a spool of fiber and a couple of splices. The fiber compromise would be easier to spot, that's all. When I got into this business, you used long spools of coax to time NTSC color TV camera signals so they'd all arrive at the switcher at the same time. Otherwise, dissolves would have weird color shifts. It was a breakthrough when electronic delay lines were added. Everything old is new again? Bigger question: why? What's the advantage to a company being listed on this exchange, rather than one of the big ones where program traders care about picoseconds? Is it just that the underlying company's valuation won't 'vibrate' quite as much in the second before it stabilizes? Does that outweigh the big traders' natural inclination to favor stocks where they can pick up a few pico's advantage? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.