Re: 197 page execution trace of DDD correctly simulated by HHH

On 7/1/2024 9:27 AM, Fred. Zwarts wrote:

Op 01.jul.2024 om 14:57 schreef olcott:

On 7/1/2024 3:27 AM, Fred. Zwarts wrote:

Op 30.jun.2024 om 19:25 schreef olcott:

On 6/30/2024 3:42 AM, joes wrote:
>

No, I mean: why does the inner simulator repeat instead of aborting,
the same as the outer one does?
>

Technically it is called detecting a repeating state.

Yeah, I know. My point is: all recursive calls both enter and detect
a repeating state.
>

>
The inner ones always see one less execution trace
than the next outer one, thus could only meet their
abort criteria after they have already been aborted.
>

>
Which indicates that they were aborted too soon, showing that the emulation was incorrect.

>
Unless the outer HHH aborts its simulation after some
fixed number of correct emulations or none of the HHH
ever aborts and HHH never stops running.

 But that does not make the result of the abort correct.

Not aborting will loop infinitely.

     If simulating halt decider H correctly simulates its input D
     until H correctly determines that its simulated D would never
     stop running unless aborted
THEN H IS NECESSARILY CORRECT TO ABORT
THEN H IS NECESSARILY CORRECT TO ABORT
THEN H IS NECESSARILY CORRECT TO ABORT

Aborting will abort too soon. Both cases are incorrect.
 

It doesn't seem like you care about the truth.

>
There is no passing the guy in front of you if you both
continue to run at the exact same speed, he will always
be ahead of you.
>

 But shooting his tyres is also incorrect.

--
Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer