Re: Can D simulated by H terminate normally?

Liste des GroupesRevenir à theory 
Sujet : Re: Can D simulated by H terminate normally?
De : polcott333 (at) *nospam* gmail.com (olcott)
Groupes : comp.theory
Date : 02. May 2024, 20:46:45
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <v10n2l$ese$3@dont-email.me>
References : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
User-Agent : Mozilla Thunderbird
On 5/2/2024 6:03 AM, Richard Damon wrote:
On 5/2/24 12:21 AM, olcott wrote:
On 5/1/2024 7:28 PM, Richard Damon wrote:
On 5/1/24 11:51 AM, olcott wrote:
Every D simulated by H that cannot possibly stop running unless
aborted by H does specify non-terminating behavior to H. When
H aborts this simulation that does not count as D halting.
>
Which is just meaningless gobbledygook by your definitions.
>
It means that
>
int H(ptr m, ptr d) {
    return 0;
}
>
>
Your H is not simulating D at all thus not
"Every D simulated by H" quoted above
 Yes it is, it is just aborting the simulation before it started.
 
D is never simulated by H is not the same as D is simulated by H.
D is simulated by H entails that 1 to N steps are simulated and
then the simulation is aborted or the simulation is never aborted.
--
Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer

Date Sujet#  Auteur
10 Nov 24 o 

Haut de la page

Les messages affichés proviennent d'usenet.

NewsPortal