Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rutgers!aramis.rutgers.edu!athos.rutgers.edu!nanotech From: celit!daemon@ucsd.edu Newsgroups: sci.nanotech Subject: Re: Some problems of super-intelligence Message-ID: Date: 20 Dec 90 18:51:22 GMT Sender: nanotech@athos.rutgers.edu Organization: FPS Computing Inc., San Diego CA Lines: 55 Approved: nanotech@aramis.rutgers.edu In article autodesk!robertj@uunet.uu.net (Young Rob Jellinghaus) writes: > >In article landman@eng.sun.com (Howard A. Landman) writes: >>One possibility is that >>people would only use a small portion of their consciousness for dealing >>with the snail-paced physical world, treating it like we do mowing the lawn. >>The rest would be free to deal with other entities operating at their own >>speed (computers and other enhanced humans). > >Myron Krueger, of Artificial Reality Corp., just gave a tech forum here >at Autodesk. He's been working on interactive video/virtual reality-type >systems for a good many years, and in a discussion of tactile feedback >systems--telerobotics, for instance--you need a 1000 Hz feedback to be able >to do very delicate types of fine work. Light can travel about 100 meters >in 1/2000th of a second (it has to go to the other end & back). They've redefined the length of the meter? Light moves at about 300,000 km/s so in half a millisecond (300,000,000 m/s * 1/2000 s) it would travel about 150000 meters. >So there >is a fairly low limit to how quickly you can process--soon, everyone else >starts lagging way behind you! You know the delay when you're talking to >someone a long way away by phone? Kind of distracting, right? Well, imagine >that delay lengthened to several minutes, or a month, or a year.... You'd certainly notice the delay as you sped up. The delay you notice on a phone link is most likely satellite delay (up to geosynch and back is about 72,000 km, or about .24 seconds each way, for a round trip delay of .48 (that's from you, up to the satellite, down, to the party on the other end and back). So, to stop you from getting too frustrated it would probably be good to keep perceived delay under a 1/2 second. At a 1000 to 1 speed-up, to get your 1000Hz perceived feedback cycles you need to run at a 1000000 Hz, giving you a 150m round trip possible. -- David L. Smith FPS Computing, San Diego ucsd!celit!dave or dave@fps.com "You can"t build a national and international network using TCP/IP" --Laurie Bride, Boeing Computer Services [Internal nervous system latencies can get up to 0.1 sec, and yet we can do things requiring much finer timing than that. It is believed that nerve signals carry (possibly implicit) timestamps and the brain sorts them out, so that (for example) you get a stimulus on your foot at time 0, one on your nose at time 0.05, your brain receives the nose message at 0.06, and the foot message at 0.1; yet you consciously experience the foot stimulus as happening first. If this mechanism were integrated into the sped-up brain, you could expect to have control as good as you have over your own body, within a radius c*t/speedup=3e8*.1/1e3 =3e4 meters = 19 miles. At a speedup of a million this goes to 30m or 100 ft. --JoSH]