# The limits of computing.

Discussion in 'Intelligence & Machines' started by TheFrogger, Apr 20, 2017.

1. ### Michael 345New year. PRESENT is 69 years oldValued Senior Member

Messages:
7,099
I wear glasses which can see small details like that

3. ### Michael 345New year. PRESENT is 69 years oldValued Senior Member

Messages:
7,099
The lower 3 zeros do not exist

I don't mean in the sense they are zero

You cannot have a naked +0 or -0

They must be mated with another number for the operator

(the + or the -)

to come into effect

Hence the first = sign is incorrect

negating the second =

causing the 3rd 0 to be non existent also

5. ### TheFroggerValued Senior Member

Messages:
1,325
n+0=n
n-0=n
n*0=n

therefore n/0=n

Zero is ignored because there is NOTHING THERE...it does not exist.

-0|0|+0

In 1+0 and 1-0 the signs are both operators AND signs...

7. ### arfa branecall me arfValued Senior Member

Messages:
6,144
But I can have a naked +1 or -1?

+1 and -1 are integers. 0 is an integer.

Are you saying I can write: 1 - 1 = 0, but I can't write 1 + (-1) = 0?

8. ### Michael 345New year. PRESENT is 69 years oldValued Senior Member

Messages:
7,099
Husband comes home and calls out to wife

Hunny I bought a plus one

Wife calls out

A plus one what?
--
Husband comes home and calls out to wife

Hunny I bought a plus zero

She takes him to the doctor
---
Later that evening he says

Hunny I returned it to the shop we are now minus one

Minus one what?
---
Hunny I returned zero to the shop we are now minus zero

She took him back to the doctor
---
Hunny I bought a plus one but took the plus one back

That's nice dear
---
Hunny I bought a plus zero but took the plus zero back

She took him to the hospital and left him there overnight
---
Moral

Don't treat nothing as if it has existence

9. ### TheFroggerValued Senior Member

Messages:
1,325
Yes zero does not exist. And it is blasphemy to attempt to name or quantify such a thing!

Messages:
136
Um, what? Division by zero generally causes an exception fault: for example, fixed-point [0x09] or floating point divide [0x0f] exceptions. This is generally a hardware fault, in my experience. What am I missing here?

11. ### sideshowbobSorry, wrong number.Valued Senior Member

Messages:
5,010
That certainly shouldn't happen. Software should never let hardware just blow up.

Messages:
136
The hardware doesn't 'blow up,' the hardware detects the error. With a few exceptions, hardware faults are those detected by hardware during the execution of software instructions. In the example, above, the ALU [Arithmetic Logic Unit] detects the fault in the execution of the instruction, puts the error code in a register (along with other pointers, depending on the actual hardware), raises the error by causing an interrupt & exits.

13. ### James RJust this guy, you know?Staff Member

Messages:
32,160
Um... what?!

n*0 = 0.

No. Because imagine that we multiply both sides of that equation by zero. We then get:
$\frac{n}{0}\times 0 = n \times 0$
$n = 0$
which is clearly not true for all n.

Of course, I made an assumption there, that 0/0 = 1, and that's not really valid either. If, instead, we had 0/0 = 17, then we'd conclude that all numbers are 17 from the above argument, which is, again, clearly bunk.

Zero is a number. It is only ignored if you ignore it.

This is meaningless rubbish.

Context is important when you're doing mathematics. So is knowing some mathematics.

14. ### sideshowbobSorry, wrong number.Valued Senior Member

Messages:
5,010
As I said, "Computers can divide by zero. They can return either infinity or just "not a number". If you tell the hardware to divide by zero, it will do it. It's just electronic circuitry. When you put voltages in here, something has to come out there. The actual electronic result of division by zero could vary from processor to processor. Any error that is raised has to be arbitrarily added, doesn't it? It could just as well raise an error if you tried to multiply by six.

Messages:
136
You have used hardware I've never encountered.

16. ### sideshowbobSorry, wrong number.Valued Senior Member

Messages:
5,010
"Zero" is just a concept that WE have superimposed on some electronic condition. Put zero in and SOMETHING - some electronic condition - has to come out. We can superimpose the concept of "error" on that condition if we choose, or not.

What hardware have you used that doesn't work that way?

Messages:
136
Why are you desperately arguing some esoteric, philosophical position? Zero IS NOT some concept or any other crap we have imposed on some electronic condition. It is a specific gate in a gate-array that causes the result to fall through when 'true' to a hardware-detected error condition. Stop trying to make it some mystical human crap.

18. ### sideshowbobSorry, wrong number.Valued Senior Member

Messages:
5,010
Huh? I'm talking about strictly physical phenomena.
And that "hardware-detected error" is what some engineer decided was an "error".

Again: You put zero in at one end of the black box and something comes out the other end. That something can be an error condition or not, depending on what the box's engineer decides. The "error-condition" is arbitrarily imposed on the electronic condition.