Is it ever possible we could do better than binary for computers

[norobinhood]

TL;DR

could humans one day invent a system that is better than binary?
never no yes

hi there I wish to discuss possible ways we could one day come up with a programming code that is even better than binary?

i would like to discuss ideas openly?

anyone got ideas?

i have some if anyone wishes to discuss possible ways?

never / no / yes

 

[Vanadium 50]

You need to define better better.

 

[norobinhood]

better would mean the code would allow for getting unknown answers, faster than a binary code system

is this better? faster computers is outcome?

 

[Vanadium 50]

Why do you think non-binary computers (which means what, exactly) are faster? Ternary (base 3) computers have been built. If they were advantageous we'd all be using them.

 

[Halc]

I've programmed analog computers. They don't seem to do the kind of work most people want from a computer, but they were at the time faster than digital ones for tasks (analog signal processing mostly) for which they excelled.

Quantum computers are another example, which have more than just binary states. So you can for instance implement a gate the performs the square root of not. If the kinks can be worked out, these hold incredible promise for being 'better' than binary computers. They'll be able to break codes that are vastly beyond the ability of a binary computer to break in reasonable (billions of years) time.

 

[fresh_42]

come up with a programming code that is even better than binary?

better would mean the code would allow for getting unknown answers

Could it be that you confuse code and logic?

 

[norobinhood]

could be...

how would you do a code better than binary?

i am looking for other ideas from you brilliant people, if my idea is bad.

I have written wikipedia in less than 72 bits with my code, using triangles.
Unfortunately its quite hard to show you as my keyboard doesn't have the symbols lol

how would you do better than binary?
binary is 0 or 1
mine is more than binary...

how should we do it?

i used the number ' infinity never' because
science says you cannot prove a negative

i believe you cannot prove a negative...unless you try, and then you can get 99.9999 percent of an answer, which is a good outcome

never adds time and meaning to anything, especially when combined with no or yes

binary makes it hard for computers to understand meaning

my idea is more than binary.

i am trying to be the spark of creation for a community of geniuses...all who are better than me

please let's work together on better than binary..

thank you all so much sorry I am so stupid

 

[DaveC426913]

never / no / yes

You referred to this in your original post as a possible form of ternary.

I don't know what you envision a 'never' state to look like but to me, it seems like a static bit that never turns on. What purpose do you think that would serve, and why do you think it would be faster?

 

[phinds]

how would you do a code better than binary?

Coding and binary have nothing to do with each other. Binary is an implementation characteristic and coding is the logic of an algorithm.

binary makes it hard for computers to understand meaning

You need to define "meaning". It is possible that no computer will every understand "meaning" by some definitions and in any case has nothing to do with it.

 

[norobinhood]

computers have so far being unable to understand meaning...because they only have binary.

i will find a way to write the language and show you somehow

i would still prefer people suggest their own ideas, rather than just criticize my ideas...

if mine is wrong to you, what would you do different?

it has to be an idea that isn't binary otherwise its binary

if you think i don't understand code, what would you do better?

code is a language for computers...

i have a new one that allows meaning

infinity never is very useful.

do i like lady gaga? infinity never?

 

[DaveC426913]

An analogy to explain why binary is likely the fastest most efficient base to base calculation operations around involves the difference between multi-tasking and uni-tasking.

Try this out loud (or type it):
1.
Count from one to ten as fast as you can: 1,2,3,4,5,6,7,8,9,10.
Now say the alphabet from A to J: A,B,C,D,E,F,G,H,I,J.
How long did that take you?
2.
Count both simultaneously:
1,A,2,B,3,C,4,D,5,E,6,F,7,G,8,H,9,I,10,J.
How long did that take you?

The second one took a lot longer because you have to convert between number and letters between each.
That's multi-tasking.

The binary ON/OFF is a single processing action. It's uni-tasking.
Anything more complex, such as YES/NO/MAYBE is actually two binary operations in one, and they're different kind of operations. Switching between operations every time you flip a bit adds overhead that slows your processing down.

 

[DaveC426913]

computers have so far being unable to understand meaning...because they only have binary.

These two things are unrelated.
Computers not understanding meaning has nothing to do with binary processing.
You might enjoy reading up on quantum computers. That may be what you're looking for.

 

[norobinhood]

i love your explanation but i don't agree...
binary can answer 1 thing at once

mine can answer 2 at once sometimes
but you guys are brilliant i think we can do this.

any idea that is not binary is better at this moment

 

[DaveC426913]

mine can answer 2 at once sometimes

Except it's slower, so not ideal.

 

[norobinhood]

quantum computers would be brilliant do you know where I can get one?

if not we might have to invent them

pretty sure quantum computers will not use binary but I could be wrong!

lets all still try!

 

[norobinhood]

Dave C my friend its not slower if it answers 2 questions in one go ...its by definition faster?

sorry if i am wrong

 

[DaveC426913]

Dave C my friend its not slower if it answers 2 questions in one go ...its by definition faster?

Answering two questions requires two decisions. (Plus the overhead to switch between them.)

 

[phinds]

pretty sure quantum computers will not use binary but I could be wrong!

You are correct, they do not use binary. They exist in numerous place.

@norobinhood, I think what we have here is a situation where you are trying to think outside the box, which is admirable, but the problem is you have no clue what is already IN the box so your floundering is not likely to get you far. You would be better off spending your time studying computer basics for a while.

 

[norobinhood]

thank you for your kindness phinds...

i am trying to ask what you brilliant people would do to make something better than binary.

i don't know much but i do know what is in the box...its something binary

i am happy to discuss any ideas we have at all...that arent binary

i wish my idea to be a starting point...

i am all ears for an idea that works better than mine!

i am thankful that someone at least sees i am trying to spark some ideas, not claim that i know it all (obviously i dont).

i am autistic and very good with numbers...i have decided to work backwards and try a different way.

Thank you all for helping me so much

lets try and find a better way than mine?

 

[anorlunda]

The IBM 650 wasn't binary. It was decimal. So were many other computers in the 1950s. Was it better?

https://en.wikipedia.org/wiki/IBM_650
The 650 was a two-address, bi-quinary coded decimal computer (both data and addresses were decimal)

 

[norobinhood]

well decimal is not better, my idea is not 10, it is 3, but not ternary.

my idea is a new ternary if it helps people...but not t f t or other 1 0 1 1 1/2 0 etc

all current ternary is 3 discrete options...
mine is 2options plus another.

it has not been done before.

but wiki says ternary is still predicted by some scientists to one day be better than binary.

Dave you are a genius, but with a never option i can answer two binary questions with one answer...therefore i can answer two questions at once...(sometimes, not always) if i guess yes and it is right

i will find it its not no, and it is not never ever (2 answers, 1 question)

hard to explain I will send you the code I am trying to draw it.

Thank you all for kindness and patience.

Also all the answers and questions can still be binary of necessary.

Also computers can't currently understand meaning, maybe if we could teach them the concept of 'never' they might be able to?

probably not but i will keep trying anyway!

thank you all for your help this is one of the best days of my life, even if I am totally wrong!

 

[fresh_42]

There are several logical models which are not binary. E.g. Kleene's logic which has a third state "Indeterminate". I got the impression that you have something like this in mind. See
https://en.wikipedia.org/wiki/Many-valued_logic#Kleene_(strong)_K3_and_Priest_logic_P3
This is not a question of coding or how computers are organized, it is a question of a different logical model. However, there is more than one alternative to the binary predicate logic we normally use.

 

[DaveC426913]

This is not a question of coding or how computers are organized, it is a question of a different logical model.

An important distinction, worth repeating.

 

[norobinhood]

thank you I will read your links and more.
Sorry for long post

I agree I am describing a logic, and I believe it would be a useful method to answer questions quickly...for a human or computer.

I cannot yet prove this could be successful but whatever current ways of doing computing are...they all seem to be binary at heart...its all 1s and 0s. I believe this is true at least for home computers?

My concept is a triangle shape instead of 1s n 0s and may bring time, and maybe even some more meaning, and faster computers. It is hard to draw as there is no symbol for it on computers!

you could just add 00 to binary and call it never, and only use it when you know an answer is wrong forever?
will norobin eat octopus tomorrow?
no yes?
i would never eat octopus.

how could a computer answer this with binary?
will norobin eat octopus tomorrow?
no
will he ever eat octopus?
no

i guess maybe it depends on the questions you want to answer.
I might have just out done myself with my own stupidity I will read up on what you suggested.

please laugh at me but don't show me pictures of octopuses! not a fan

 

[DaveC426913]

My concept is a triangle shape

OK, so, to decide where a location is on a triangle, you need to determine an X coordinate and a Y coordinate.
That's two decisions.
Your one cycle takes as long to determine the answer as two binary cycles. So it's not faster.

But more than that, it needs to switch between two different decision types: X and Y.
That switch adds overhead. See the multi-tasking example in step 11 for why.

So your one cycle will decide two problems in twice-and-a-bit longer than binary will decide the same two problems. i.e.: slower.

 

[norobinhood]

fresh 42 thank you very much for the link...it was brilliant and helped me so much.
https://en.wikipedia.org/wiki/Many-valued_logic#Kleene_(strong)_K3_and_Priest_logic_P3

i believe my idea is similar to 3 valued logic, except instead of y n 'unknown' my idea is y n never

according to link
3 valued logic can be finitely valued 3 discrete options ( my idea is not this)
in modern ternary 3rd option is 'undecided'

my third option is 'never'

according to link
3 valued logic can be applied to fuzzy logic (fuzzy is by definition not perfect on its own)

according to link
to bring meaning to fuzzy logic, can be done by including a finite value logic (never)
when it is applied to propositions, only if it has 'computational semantics' (i believe this means 'meaning'

i thnk if you change 'undecided' to never and throw in some fuzzy logic all is not lost for us yet!

maybe my idea is only useful to help answer questions of meaning?

if it somehow combines fuzzy and 3 valued logic somehow, it could have one or two interesting applications

no where anywhere is 'never' mentioned as being one of the 3 valued logic options.

we should try it

thank you so much for the link i learned so much! Aristotle was brilliant! but no or yes isn't going to work for a none binary system

please anyone recommend me any links and i will try to do better

thank you so much I am so grateful. I am learning as i go

ps hopefully i haven't just started skynet by answering questions with meaning! (just joking I know I know nothing, I am just trying)

 

[Klystron]

A brief overlook of how we implement mathematical logic in computer circuits may help understanding.

This article describes digital signal processing. A takeaway relevant to your questions:

Simple digital signals represent information in discrete bands of analog levels. All levels within a band of values represent the same information state. In most digital circuits, the signal can have two possible valid values; this is called a binary signal or logic signal.[4] They are represented by two voltage bands: one near a reference value (typically termed as ground or zero volts), and the other a value near the supply voltage. These correspond to the two values "zero" and "one" (or "false" and "true") of the Boolean domain, so at any given time a binary signal represents one binary digit (bit). ...

with a brief expansion on multi-valued logics:

Digital signals having more than two states are occasionally used; circuitry using such signals is called multivalued logic. For example, signals that can assume three possible states are called three-valued logic.

We learn that binary computers do not actually manipulate "zeroes" and "ones" but operate on discrete voltage levels. Ones and zeroes symbolize logic states representing "True" and "False", "On" and "Off"; also including operators such as NOT. Your introduction of "never" might represent a signal out of the usable voltage bands 1) reference, ground or zero volts; and 2) voltage near the power supply output voltage.A common fallacy arose since the earliest days of electronic and mechanical computation. Since computers appear to perform tasks such as counting that humans perform and at much higher speeds, people, including educated professionals, expected computers to "think", to develop (artificial) intelligence. While progress has been made in technology and our understanding of computation and human intelligence, "computers understanding meaning" remains AFAIK an anthropomorphic fiction; i.e., endowing machines with human attributes.

 

[norobinhood]

thank you Dave I will think about what you said.
a few initial thoughts...
i never initially set out to write a language, but i will send it to you
On wiki it says binary code for the word 'wikipedia'is 9 lots of 8 numbers (all 1 and zero)
Each 1 and zero is made up of several smaller lines.

I am trying to draw a shape that covers more, with less. so either less numbers or lines to convey the same information.

I am not sure it is definitely relevant but I will try to make it shorter.

whether we can make a computer draw a triangle is one immediate but not insurmountable issue.

If I cannot do it, I will see what else could be done instead.
thank you i will think on what you said.

It might be that I am trying to achieve something different that computers don't do or don't need to do.

I will think and do better

 

[Drakkith]

It might be that I am trying to achieve something different that computers don't do or don't need to do.

I'm going to be brutally honest with you. You have no idea what you're trying to do. You have a vague idea in your head but you lack 99.9% of the knowledge that would let you figure out whether that idea is plausible or even meaningful.

Before trying to improve the wheel it's important that you understand how the wheel works first. Learn how computers work first, then try to improve them. You'll need to spend about about a decade in classes and in the industry to understand digital logic, semiconductor devices, programming, and to learn the math required to make any progress in this area. Only then will you be in any reasonable position to advance things.

 

[norobinhood]

Klyston thank you so much for the brilliant link I am learning so much

i think it implies that my idea might be tricky if using the 2 digital circuit states.

either we need to make it work or change the circuitry to 3 options.
"
Digital signals having more than two states are occasionally used; circuitry using such signals is called multivalued logic. For example, signals that can assume three possible states are called three-valued logic."
this is brilliant, it means that 3 valued logic
can be created using MULTI VALUED logic, (which is a type of circuitry?)

possibly, we might need some new computers, unless we can do it with existing two circuit system

will read more thank you I am learning so much.

I know I am still 99.999999 wrong but i still think we could try putting 'never' in a multi valued logic circuit ?

I will read more on 3 valued logic circuits!