Re: H(D,D) cannot even be asked about the behavior of D(D) V2

On 6/16/2024 1:06 PM, Richard Damon wrote:

On 6/16/24 1:44 PM, olcott wrote:

On 6/16/2024 10:02 AM, Alan Mackenzie wrote:

olcott <polcott333@gmail.com> wrote:

On 6/16/2024 9:16 AM, joes wrote:

Am Sun, 16 Jun 2024 07:44:41 -0500 schrieb olcott:

On 6/16/2024 2:50 AM, Mikko wrote:

>

Whenever a decider is run it answers the question it is made to answer.

Not necessarily. Just because everyone falsely assumes that D correctly
simulated by H must have the same behavior as the directly executed D(D)
does not make this false assumption true.

>

You still need to explain how you can call a simulation that differs from
the behaviour of its input "correct".

>
Indeed, you do.
>

I have proven it many times and this proof is simply over
everyone's heads.

>
Nonsense!  How about, instead of "proving", actually explaining?  If a
simulation differs from its original, it's not a simulation; it's just a
random program.
>

When I ask what your C programming skill level is, this *is not* a
rhetorical question.

>
The question has nothing to do with C programming.
>

>
typedef void (*ptr)(); // pointer to void function
int H(ptr P, ptr I);
>
int D(int (*x)())
{
   int Halt_Status = H(x, x);
   if (Halt_Status)
     HERE: goto HERE;
   return Halt_Status;
}
>
Unless I make every single detail 100% explicit false
assumptions always slip though the cracks. The ONLY way
to make EVERY SINGLE DETAIL 100% EXPLICIT is the x86
programming language.
>
There cannot possibly be any H that correctly emulates
the x86 machine code of D according to the semantics
of the x86 programming language such that the emulated
D ever reaches its own emulated final state at machine
address [00001f58].
>

 Which is just a strawman, as the requirement on H is NOT to answer about "D correctly simulated by H" but about "the program represented by the input directly executed", or equivalently, simulated by an actual UTM, which is a simulator that NEVER stops until it reaches a final state.
 

This is simply over-your-head.
I am very glad of that because the alternative would
possibly condemn your soul to Hell.

For this input, D(D), since H(D,D) returns 0, D(D) will Halt, so H is just wrong by definition, and you by the attempt to use a strawman.
 

_D()
[00001f33] 55         push ebp
[00001f34] 8bec       mov ebp,esp
[00001f36] 51         push ecx
[00001f37] 8b4508     mov eax,[ebp+08]
[00001f3a] 50         push eax        ; push D
[00001f3b] 8b4d08     mov ecx,[ebp+08]
[00001f3e] 51         push ecx        ; push D
[00001f3f] e87ff7ffff call 000016c3   ; call H(D,D)
[00001f44] 83c408     add esp,+08
[00001f47] 8945fc     mov [ebp-04],eax
[00001f4a] 837dfc00   cmp dword [ebp-04],+00
[00001f4e] 7402       jz 00001f52
[00001f50] ebfe       jmp 00001f50
[00001f52] 8b45fc     mov eax,[ebp-04]
[00001f55] 8be5       mov esp,ebp
[00001f57] 5d         pop ebp
[00001f58] c3         ret
Size in bytes:(0038) [00001f58]
>
Once the above is understood (people quit denying verified facts).
thenn (then and only then) I can show how this applies to Turing
machines.
>

 No, YOU are the one denying DEFINED FACTS.

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