this is on tristate numbering system it may have some uses http://abhijit.info/tristate/tristate.html would this make a comp.tsp news group, or would it be more appropriate here? cheers
from comp.compression
Started by ●August 9, 2006
Reply by ●August 9, 20062006-08-09
jacko wrote:> this is on tristate numbering system > it may have some uses > > http://abhijit.info/tristate/tristate.html > > would this make a comp.tsp news group, or would it be more appropriate > here?I don't know. Tell me about comp.tsp. The idea of trinary numbers, even bipolar trinary, is older in computing than transistors. They are interesting to play with, particularly in puzzles, but implementing a system has never been practical.* Binary devices are either mostly on or mostly off; straight trinary adds more-or-less in between. Bipolar trinary holds more promise with mostly positive, mostly negative, and mostly off. To see why it never fulfilled that promise, try to design a four-trit adder. there are 81 combinations of 4 trits. Seven bits yield 120 combinations. Compare the complexity of a 3-trit adder to a 7-bit adder and the point will be clear. Now consider multiplication. Give up? M-ary number systems are used for communications to get the most information through a particular channel, and for making puzzles. See "Twelve Coins, Revisited" on my web site, http://users.rcn.com/jyavins/ Jerry ____________________________________ * At least one trinary computer was built. It demonstrated both the possibility and the impracticality, especially using core memory. -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●August 9, 20062006-08-09
Reply by ●August 9, 20062006-08-09
Reply by ●August 9, 20062006-08-09
Jerry Avins wrote:> jacko wrote:>> so...? > So what?Since Jacko added comp.compression with that post I guess he wanted people there to take up your comments. Jacko, Usenet is not a web board. Users of newsreader software may not have previous posts in a thread available from their news servers, so please always quote whatever you're referring to. Martin [restoring context from comp.dsp] Jerry Avins wrote:> jacko wrote:>> this is on tristate numbering system >> it may have some uses >> >> http://abhijit.info/tristate/tristate.html >> >> would this make a comp.tsp news group, or would it be more >> appropriate here? > > I don't know. Tell me about comp.tsp. > > The idea of trinary numbers, even bipolar trinary, is older in > computing than transistors. They are interesting to play with, > particularly in puzzles, but implementing a system has never > been practical.* Binary devices are either mostly on or mostly > off; straight trinary adds more-or-less in between. Bipolar > trinary holds more promise with mostly positive, mostly > negative, and mostly off. To see why it never fulfilled that > promise, try to design a four-trit adder. there are 81 > combinations of 4 trits. Seven bits yield 120 combinations. > Compare the complexity of a 3-trit adder to a 7-bit adder and > the point will be clear. Now consider multiplication. Give up? > > M-ary number systems are used for communications to get the most > information through a particular channel, and for making > puzzles. See "Twelve Coins, Revisited" on my web site, > http://users.rcn.com/jyavins/ > > Jerry > ____________________________________ > * At least one trinary computer was built. It demonstrated both > the possibility and the impracticality, especially using core > memory.-- Quidquid latine scriptum sit, altum viditur.
Reply by ●August 10, 20062006-08-10
"Jerry Avins" <jya@ieee.org> wrote in message news:eKqdnctnQMhzgUfZnZ2dnUVZ_sKdnZ2d@rcn.net...> jacko wrote: >> this is on tristate numbering system >> it may have some uses >> >> http://abhijit.info/tristate/tristate.html >> >> would this make a comp.tsp news group, or would it be more appropriate >> here? > > I don't know. Tell me about comp.tsp. ><snip>> > M-ary number systems are used for communications to get the most > information through a particular channel, and for making puzzles. See > "Twelve Coins, Revisited" on my web site, http://users.rcn.com/jyavins/ >one thing I wonder about here is for magnetic storage. albeit, one kind of problem I can imagine (intuitively) would be a tendency of values to try to "bleed over" on the media after a while (combined with the likely somewhat more complicated magnetic head...). -0+ the 0 could possibly polarize (at the cost of partly depolarizing a nearby value), or 1 size would polarize negative and the other positive. -?+ ... at least with a binary system based on opposites, I think, most of the time, the system would be fairly stable (the opposites would reinforce each other I think, at least assuming vertical polarization, horizontal is likely to be the opposite here, at least within smaller areas). I would then suspect (more intuition) the problem would be an overly large concentration of the same polarity in a single place, possibly causing bits to occasionally invert or similar. -+++++++++- might deform, eg: -++++-++++- or: -++-+++-++- or something... (just checked online, actual drives seem to use horizontal rather than vertical polarization, so quite possibly the effect would be different). I might be wrong on all this though... or such...> Jerry > ____________________________________ > * At least one trinary computer was built. It demonstrated both the > possibility and the impracticality, especially using core memory. > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������
Reply by ●August 10, 20062006-08-10
cr88192 wrote:> "Jerry Avins" <jya@ieee.org> wrote in message > news:eKqdnctnQMhzgUfZnZ2dnUVZ_sKdnZ2d@rcn.net... >> jacko wrote: >>> this is on tristate numbering system >>> it may have some uses >>> >>> http://abhijit.info/tristate/tristate.html >>> >>> would this make a comp.tsp news group, or would it be more appropriate >>> here? >> I don't know. Tell me about comp.tsp. >> > <snip> > >> M-ary number systems are used for communications to get the most >> information through a particular channel, and for making puzzles. See >> "Twelve Coins, Revisited" on my web site, http://users.rcn.com/jyavins/ >> > one thing I wonder about here is for magnetic storage. > > albeit, one kind of problem I can imagine (intuitively) would be a tendency > of values to try to "bleed over" on the media after a while (combined with > the likely somewhat more complicated magnetic head...). > > -0+ > > the 0 could possibly polarize (at the cost of partly depolarizing a nearby > value), or 1 size would polarize negative and the other positive. > > -?+ > ... > > at least with a binary system based on opposites, I think, most of the time, > the system would be fairly stable (the opposites would reinforce each other > I think, at least assuming vertical polarization, horizontal is likely to be > the opposite here, at least within smaller areas). > > I would then suspect (more intuition) the problem would be an overly large > concentration of the same polarity in a single place, possibly causing bits > to occasionally invert or similar. > > -+++++++++- > might deform, eg: > -++++-++++- > or: > -++-+++-++- > > or something... > > (just checked online, actual drives seem to use horizontal rather than > vertical polarization, so quite possibly the effect would be different). > > > I might be wrong on all this though... > > or such...It's easier to see that you're right if you equate signal degradation to noise. Remember the the capacity of a channel or medium depends not only on bandwidth, but on its noise floor. Infinite SNR implies infinite channel capacity. How would you like to record all the printed works in the Library of Congress on a piece of wire no more than an inch long? It's simple, really (but not necessarily practical). For versatility, record all text as unicode (16 bits), string all the characters together to make one big number, precede the whole with a binary point, and cut the wire to that length. To retrieve the archive, measure the wire and work backwards. Knuth, in /The Art of Programming/ describes an O(1) sort he calls "spaghetti sort". Encode each number as the length of a piece of raw spaghetti and put it into a box. When the time comes to sort the numbers, stand the strands on a flat table. Slam! they are in sorted order, with the largest number standing tallest. To get real, you just showed one of the many ways that binary makes best use of resources when robustness is important. It is easy to lay a book on a jostling table, front or back cover up. It is not so easy to stand it on edge. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●August 10, 20062006-08-10
"Jerry Avins" <jya@ieee.org> wrote in message news:bdWdnabIL-X_zkbZnZ2dnUVZ_t-dnZ2d@rcn.net...> cr88192 wrote: >> "Jerry Avins" <jya@ieee.org> wrote in message >> news:eKqdnctnQMhzgUfZnZ2dnUVZ_sKdnZ2d@rcn.net... >>> jacko wrote: >>>> this is on tristate numbering system >>>> it may have some uses >>>> >>>> http://abhijit.info/tristate/tristate.html >>>> >>>> would this make a comp.tsp news group, or would it be more appropriate >>>> here? >>> I don't know. Tell me about comp.tsp. >>> >> <snip> >> >>> M-ary number systems are used for communications to get the most >>> information through a particular channel, and for making puzzles. See >>> "Twelve Coins, Revisited" on my web site, http://users.rcn.com/jyavins/ >>> >> one thing I wonder about here is for magnetic storage. >> >> albeit, one kind of problem I can imagine (intuitively) would be a >> tendency of values to try to "bleed over" on the media after a while >> (combined with the likely somewhat more complicated magnetic head...). >> >> -0+ >> >> the 0 could possibly polarize (at the cost of partly depolarizing a >> nearby value), or 1 size would polarize negative and the other positive. >> >> -?+ >> ... >> >> at least with a binary system based on opposites, I think, most of the >> time, the system would be fairly stable (the opposites would reinforce >> each other I think, at least assuming vertical polarization, horizontal >> is likely to be the opposite here, at least within smaller areas). >> >> I would then suspect (more intuition) the problem would be an overly >> large concentration of the same polarity in a single place, possibly >> causing bits to occasionally invert or similar. >> >> -+++++++++- >> might deform, eg: >> -++++-++++- >> or: >> -++-+++-++- >> >> or something... >> >> (just checked online, actual drives seem to use horizontal rather than >> vertical polarization, so quite possibly the effect would be different). >> >> >> I might be wrong on all this though... >> >> or such... > > It's easier to see that you're right if you equate signal degradation to > noise. Remember the the capacity of a channel or medium depends not only > on bandwidth, but on its noise floor. Infinite SNR implies infinite > channel capacity. How would you like to record all the printed works in > the Library of Congress on a piece of wire no more than an inch long? It's > simple, really (but not necessarily practical). For versatility, record > all text as unicode (16 bits), string all the characters together to make > one big number, precede the whole with a binary point, and cut the wire to > that length. To retrieve the archive, measure the wire and work backwards. > Knuth, in /The Art of Programming/ describes an O(1) sort he calls > "spaghetti sort". Encode each number as the length of a piece of raw > spaghetti and put it into a box. When the time comes to sort the numbers, > stand the strands on a flat table. Slam! they are in sorted order, with > the largest number standing tallest. > > To get real, you just showed one of the many ways that binary makes best > use of resources when robustness is important. It is easy to lay a book on > a jostling table, front or back cover up. It is not so easy to stand it on > edge. >yes, ok. partly my reasoning was also based on trying to imagine the fields, but analogies work. yeah, neutral fields are unstable in the presence of positive and negative ones, and then you have to detect them, and setting a spot to neutral could be problematic as one has to adjust the field to a particular value, rather than just setting it... or such... not sure if I am making much sense here. it seems recently I have been starting to deal with some mental coherency issues. it used to be that I would need stress for stuff to break down, but now it seems the parts of my mind are breaking down their harmonious contracts and going off on their own absent any real sense of stress. this can't be good, then again, it could be a source of stress that I don't notice, but I don't know. at least everything is still working it seems, that is good. can't yet find the part that has failed, maybe the bus, I don't know. I guess for now I am ok though... it is odd though that right now things seems silent, and nothing is bothering me, and I seem to be located in physical reality. maybe it is a trick though, maybe they are hiding themselves, but normally I would have noticed. then again, this is the voice of the part which is skeptical, it is a different part which is known for ranting about spirits and magery, and this part seems to be being quiet, ok... each part is a part, and to each its own, but I am self as well, and the self needs correct functioning of the various parts. or maybe I have been tricked and I am the self and all is well. one can never be too sure what tricks they have pulled. maybe it is a trick of making me doubt them by making me once again question myself? I don't know... then again, I doubt that all is well here.> Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������
Reply by ●August 10, 20062006-08-10
cr88192 wrote: ...> then again, I doubt that all is well here.To keep it short, I sincerely hope that all *is* well. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●August 10, 20062006-08-10
"Jerry Avins" <jya@ieee.org> wrote in message news:XKWdnXCsTOXaeEbZnZ2dnUVZ_oidnZ2d@rcn.net...> cr88192 wrote: > > ... > >> then again, I doubt that all is well here. > > To keep it short, I sincerely hope that all *is* well. >yeah. I guess my state has stabilized partly. different possibilities exist in my case: I am not entirely mentally stable; I am a person who has generally turned away from using magics, but is still being bothered some by the presence of unpleasant spirits; ... oh well really, at least my coherency seems to have recovered for the time being. I am still left though absent a particularly strong sense of "reality", but this is nothing new... but, yeah, I experienced a lot of this, a lot worse, in highschool, but since then things had cleared up quite a bit for a number of years. note that I am autistic (in particular, aspergers), but I am not entirely sure if this is related to all this... (I guess it is a mystery of how common, or what it means exactly, for an autistic to develop some likely schizo traits...). then again, both are spectrum disorders, so it is possible there is some overlap (I have had some relatives that are both autie and schizo, but on different sides of the family, so yeah...). oh well, probably/hopefully nothing too major...> Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������
