1

I am not a computer scientist and this is my first question.

This question is a question in layman terms and I also want the answer in layman terms.

When I searched hyper-computation. There was a list of models of hyper-computers.

Now, my question is that what are the problems with those models and if it is possible to create models of computers which can compute non-Turing computable things then doesn't, that already proves the Church-Turing thesis wrong and if it does prove that Church-Turing thesis is wrong then why people still consider it to be a thing?

Brian Tompsett - 汤莱恩
  • 1,845
  • 16
  • 24
Rounak Sarkar
  • 113
  • 6

2 Answers2

5

The Church–Turing thesis is about physically realizable machines. To the best of our knowledge, hypercomputation models cannot be realized in the physical world. They are a figment of our imagination.

If someone would find a new law of physics which enables solving the halting problem, that would make a big fuss in the world of science. But I am not holding my breath.

Yuval Filmus
  • 280,205
  • 27
  • 317
  • 514
1

I'll just address one model. WP says: The Zeno machine performs its first computation step in (say) 1 minute, the second step in ½ minute, the third step in ¼ minute, etc. By summing 1+½+¼+... (a geometric series) we see that the machine performs infinitely many steps in a total of 2 minutes. We can imagine all kinds of devices. But, looking around, we see that researchers who want to study an idealization of a physically-realizable machine, in order to get insights into what is mechanically computable, do not find that studying these is fruitful.

Jim Hefferon
  • 138
  • 6