Anonymous at Tue, 9 Jul 2024 02:18:53 UTC No. 16273607

>>16273597
computers haven't been turing complete in decades

Anonymous at Tue, 9 Jul 2024 03:35:57 UTC No. 16273640

Computer cabal must maintain their monopoly.

Anonymous at Tue, 9 Jul 2024 06:33:23 UTC No. 16273802

>>16273597
you arent suggesting anything better you fraud

Anonymous at Tue, 9 Jul 2024 06:50:54 UTC No. 16273812

>>16273597
No we have a super duper secret machine capable of predicting the future but you shouldn't tell anyone or else the secret police will find you and kill your grandpa with our time travel machine, I know you won't since I already know the future

Anonymous at Tue, 9 Jul 2024 06:52:06 UTC No. 16273813

>>16273607
Yes they have, infinite memory has nothing to do with being Turing complete, a finite memory machine can do any computation an infinite memory machine can do

Anonymous at Tue, 9 Jul 2024 07:05:06 UTC No. 16273821

>>16273813
>a finite memory machine can do any computation an infinite memory machine can do
No, it can't compute the nth digit of pi where n>memory.

Anonymous at Tue, 9 Jul 2024 07:54:51 UTC No. 16273872

>>16273813
Proof?

Anonymous at Tue, 9 Jul 2024 08:16:43 UTC No. 16273881

>>16273813
This is only true if an infinite number is a subset of a finite set of numbers. And that is not true in most math AFAIK especially if you combine two infinite numbers, such that each infinite number has no divisor of finite number sequences then for each infinite number, only an infinite memory machine may compute its state

Anonymous at Tue, 9 Jul 2024 08:22:17 UTC No. 16273882

>>16273881
Actually I am a retard and forget what I said

Anonymous at Tue, 9 Jul 2024 08:23:44 UTC No. 16273883

>>16273881
Although if this were true, does that mean reality is an infinite number machine?

Anonymous at Tue, 9 Jul 2024 08:39:20 UTC No. 16273890

>>16273607
>computers haven't been turing complete in decades
Then show me one Turing complete and physically manufactured computer

Anonymous at Tue, 9 Jul 2024 11:13:57 UTC No. 16274000

>>16273597
> Is this the best that computer "scientists" can come up with?
This attitude is wrong. I can tell that you don't know anything about Turing machines, because anyone who has learnt about Turing machines knows about the Church-Turing thesis, so why are you making snarky comments about something that you ignore?

Anyway, to answer your question: there are many other models for computation, such as Post-Turing machines, lambda calculus,the GOTO language, or register machines. However, all of them are equally powerful to Turing machines, so there is no best model for computation. There is a nice section in the Wikipedia page of Turing Machine if you are interested.

Anonymous at Tue, 9 Jul 2024 13:24:58 UTC No. 16274095

>>16273882
I refuse to

Anonymous at Tue, 9 Jul 2024 13:29:42 UTC No. 16274099

>>16273821
Yes it can, it just takes a long fucking time we don't have to store all the numbers, just the nth digit and whatever is needed for calculating the n+1 so stfu if you don't understand what a Turing machine is and how to build an algorithm

Anonymous at Tue, 9 Jul 2024 20:39:53 UTC No. 16274968

>>16273597
>Why is this the theoretical model for a computer?
It's A theoretical model for a computer. There are many others, the Turing machine is only one of many.

Anonymous at Tue, 9 Jul 2024 21:58:14 UTC No. 16275090

>>16273890
Minecraft Redstone.

Anonymous at Wed, 10 Jul 2024 22:00:01 UTC No. 16276617

>>16273597
I have yet to understand how a Turing machine would do computer graphics. Would it be possible for a Turing machine to compute its own GUI?

Anonymous at Wed, 10 Jul 2024 22:09:53 UTC No. 16276629

>>16273597
Turing machines are a thought experiment, not the basis for real life computers. Nerds build them for fun, but it's physically impossible to build a true one because a requirement is infinite tape length.

It probably seems obvious that a computer can solve any math equation, but people in the 1940s weren't so sure. The issue that a computer needs to break equations down into many very simple steps. The Turing machine thought experiment logically proves it's possible to compute anything which was kind of a big deal.

Anonymous at Thu, 11 Jul 2024 20:09:16 UTC No. 16277789

>>16274099
No it can't.
>just the nth digit and whatever is needed for calculating the n+1
Yeah, you "just" need to store the information for calculating the n+1, which, by definition, is larger than your memory.

Anonymous at Thu, 11 Jul 2024 20:15:54 UTC No. 16277799

>>16277789
Are you one of those assblasted AI fags that just realized computers will never be intelligent because they are equivalent to punch cards or a cassette tape?

Anonymous at Thu, 11 Jul 2024 20:31:57 UTC No. 16277816

>>16277789
>By definition
By your stupid definition, n is finite dumbshit, I just need enough memory and time to compute n+1.
I'm not saying we can represent the full number pi and that is not needed for it to be Turing complete

Anonymous at Thu, 11 Jul 2024 20:33:30 UTC No. 16277820

>>16276629
No it's not possible to compute anything/everything since undecidable problems exist.

Anonymous at Thu, 11 Jul 2024 20:44:10 UTC No. 16277836

>>16277799
I'm one of those who knows what a Turing machine is. If you study really hard, you might one day be able to join us.

Anonymous at Thu, 11 Jul 2024 20:45:59 UTC No. 16277841

>>16277816
>I just need enough memory and time to compute n+1.
Yeah, you "just" need enough memory to compute n+1... Here's a hint for you: assume that your memory consists of precisely one number. The last digit you've computed is 7. What's the next one?

Anonymous at Thu, 11 Jul 2024 21:01:09 UTC No. 16277856

>>16277841
I know it might be hard for you to comprehend since I have a feeling you don't have a background in computer science but having enough memory doesn't have to mean 1 number or infinite numbers stored. If I want to calculate the 1 decimal of pi I can use the 22/7 representation an so on. I never said the same finite memory machine can calculate all the digits of pi in one go as that is not needed for it to be Turing complete, but reduction proofs might just go over your head

Anonymous at Thu, 11 Jul 2024 21:08:14 UTC No. 16277865

>>16277856
>I have a feeling you don't have a background in computer science
Yup. I don't even have a high school diploma, and yet I know way more about Turing machines than you do. The cereal box you got your diploma from must have been from a pretty bad batch.

I honestly find it quite appalling how this statement was not enough to shut you up:
>Here's a hint for you: assume that your memory consists of precisely one number. The last digit you've computed is 7. What's the next one?
So? What's the next digit?

Anonymous at Thu, 11 Jul 2024 21:14:52 UTC No. 16277869

>>16277865
Why do I have to assume the memory has one digit? That just makes no sense at all even for a retarded argument you surely see how stupid that makes you sound. If you point at me where I claimed a 1 digit memory can compute any digit of pi I'll concede... I think you just are illiterate enough to not understand the meaning of the word *enough*
You claim you know more about turing machines yet fail to define one properly by using the definition of algorithm and finite states.

Anonymous at Thu, 11 Jul 2024 21:24:25 UTC No. 16277880

>>16277869
I made the mistake of thinking "one digit" would be simple enough for you to understand. So pick whatever number you want. Use 2 digits, 10 digits, or one googol digits if you so wish. Constant-bound it however you want to get your finite memory, and then show me how you get your n+1:th digit, where n is larger than your memory.
>yet fail to define one properly
I don't need to define what a Turing machine is because that's dictionary knowlegde. Look it up. I actually thought that a proper computer scientist like yourself should already know what a Turing machine is. Evidently not.

Anonymous at Thu, 11 Jul 2024 21:36:18 UTC No. 16277891

>>16277880
Surely you understand that if you define the machine by saying it has insufficient memory than it's trivial to show it won't be able to calculate. What I'm saying is there is a machine with finite memory that can compute. Sorry for not writing it clearly enough. If you define something by saying it fails when X then of course it fails when X, that doesn't exclude de possibility that there might be another definition by which X doesn't fail. That might be a little better but unfortunately you are not worth any more of my time.

Anonymous at Thu, 11 Jul 2024 21:43:56 UTC No. 16277899

>>16277891
Well, since you found this problem too hard and have given up, I can give the answer to this exercise. I really do hope, by the way, that you do not have any kind of degree in computer science, since this is just embarrassing. A high-school-dropout trounced you.

Anyway, here goes: A Turing machine with a finite tape is unable to solve this problem because there are only finitely many states the machine can be in, and your computation will at some point start going through a loop of those finite states. Intuitively, once you start computing digits of Pi, you will eventually reach the point, where merely saying "I've calculated the N:th digit, what's the next one?" is impossible because N itself won't fit in the memory.

Anonymous at Thu, 11 Jul 2024 22:42:54 UTC No. 16277942

>>16273813
t. dumbass

every computer ever built is a finite state machine exactly because every computer ever built has finite memory

an FSM with N states can make at most N-1 transitions without revisiting a previous state
the Nth transition will definitely revisit a previous state

Anonymous at Fri, 12 Jul 2024 09:25:51 UTC No. 16278498

>>16277789
At some point, even n and n+1 will be larger than your memory space, so you will still eventually be unable to parse n let alone calculate pi at n.

Anonymous at Fri, 12 Jul 2024 09:27:13 UTC No. 16278500

>>16277880
>or one googol digits
Nope that will definitely exceed the memory space, there is no such thing as a googol-bit register and there never will be.

Anonymous at Fri, 12 Jul 2024 10:54:51 UTC No. 16278561

>>16273597
>Why is this the theoretical model for a computer?
Because it came first and the guy who came up with it was thinking in terms of a very simple theoretical machine rather than something purely abstract, presumably because he wanted something more grounded in the technological reality of his time. You can come up with far better mathematical abstractions.

Anonymous at Sat, 13 Jul 2024 03:27:50 UTC No. 16279799

Von Neumann machine better

Anonymous at Sat, 13 Jul 2024 12:15:21 UTC No. 16280186

>>16277942
What does it matter when machine states >> actual states visited. The limitation is computing power, not memory.

Cult of Passion at Sat, 13 Jul 2024 12:37:00 UTC No. 16280202

>>16273597
>Is this the best that computer "scientists" can come up with?
Does it contain 200,000 feet of data-tape?

Anonymous at Sat, 13 Jul 2024 12:47:44 UTC No. 16280209

>>16273607
what are you talking about? all computers are turing complete by definition

Anonymous at Sun, 14 Jul 2024 14:43:48 UTC No. 16281385

>>16280209
This isn't true. There have been many computers that are not turing complete.