Forums before death by AOL, social media and spammers... "We can't have nice things"
|    sci.physics    |    Physical laws, properties, etc.    |    178,769 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 176,949 of 178,769    |
|    Physfitfreak to Farley Flud    |
|    Re: Problem For Physfitfreak (monospace     |
|    03 Nov 24 01:30:59    |
      XPost: comp.os.linux.advocacy       From: physfitfreak@gmail.com              On 11/2/24 05:32, Farley Flud wrote:       > On Sat, 2 Nov 2024 00:15:11 -0500, Physfitfreak wrote:       >       >>       >> In an absolute way? Disregarding unrelated limitations?       >>       >> Then that base would be infinity :) Then any real and complex number       >> will be expressed in just one digit :)       >>       >       > Good guess.       >       > This stuff is quite new to me. I just stumbled upon it when       > doing some research on ternary (i.e. base 3) computers.       >       > It's all explained here:       >       > https://en.wikipedia.org/wiki/Optimal_radix_choice       >       > The "cost" of representing a number N in base b is (monospace       > font required):       >       > log(N)       > b * floor(------ + 1)       > log(b)       >       >       > Letting N --> inf we get the asymptotic cost:       >       > b       > ------       > log(b)       >       >       > Here is an image of the plot of this relation:       >       > https://i.postimg.cc/CLwHyyzP/asymptotic-cost.png       >       > The minimum of this curve will give the lowest "cost"       > of expression.       >       > Take the derivative, set it equal to zero, and solve.       >       > Using Maxima:       >       > diff(b/log(b),b);       >       > 1 1       > ------ - -------       > log(b) 2       > log (b)       >       >       > Set equal to zero and solve:       >       > solve(diff(b/log(b),b) = 0, b);       >       > [b = %e]       >       > There it is! The most efficient number base is e, Euler's       > constant.       >       > But since irrational bases are not practical, we use the closest       > integer base which is 3.       >       > Thus, ternary (base 3) computers are the most efficient in       > storing numbers.       >       >       >                     I'll go over that later (it is very late now). I read an article about       early Soviet computers built using a ternary number base when I was       beginning the third year of undergraduate physics. The department had       created an optional "computer programming" course that year for physics       students, or so I heard. This was early 1970s. I took that course and       attended the first two class sessions, then dropped the course right       away when I found out it was designed and created by the engineering       school, not the physics or math departments. I had a bad experience with       another optional course, "statics", a couple years earlier which had       caused us physics students grief and much anger in the way the material       were exposed to students and how strangely absurd the treatment we got       felt to us. That course was also an engineering school course. Both text       and the instructor were from the engineering school.              In that second session of the computer programming course, the       instructor gave us a handout about binary and ternary systems. Evidently       back then they were still thinking the choice Soviets had made (or had       tried at least) could be the one that become common.              Even that handout smelled of an engineer behind it creating it. For       instance, I remember I went back and forth in it to see where they       mention _which_ three numbers the Soviets had chosen. In binary of       course 0 and 1 had been chosen but in ternary that damn shit of a       handout would not say which three numbers were chosen! 0, 1, and 2? -1,       0, and 1? Even -2, -1, and 0? The creater of the handout was oblivious       to the significance of which set to choose and which one the Soviets       chose...              It disgusted me just like that "Statics" text had done, and I dropped       the course right after reading that handout.              Physics books are written very, very carefully! Because there's no other       way to say anything in there. They're not discussing things for       technicians. They're not discussing things for "engineers" or       "managers." They're discussing them for human, and for the sake of       _only_ finding stuff about nature; nothing else.              --- SoupGate-Win32 v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca