Posted by Ken Maltby on 11/16/99 11:39

"Bob" <spam@uce.gov> wrote in message
news:43eba1b1.10875265@news-server.houston.rr.com...
> On 9 Feb 2006 10:23:34 -0800, "Goro" <evilninjax@yahoo.com> wrote:
>
>>> For example, Shrink could be provinding a pointer in one numerical
>>> representation like unsigned int (because the clip is small) whereas
>>> AGK could be treating it as unsigned long, which would put garbage in
>>> the most significant digits and force the pointer way past the place
>>> it needs to point to.
>
>>this isn't even remotely what's happening.
>
> Prove it.
>
>

First you can't do it in C or Assembler, by accident. Any
competent programmer would pay some attention to the errors
that would be thrown up if you tried to do what you describe.
I know that to be true for MS C 5.0 and later. Then there is
what the OS will allow in terms of data exchange. But the
most laughable thing is your idea that a larger precision format
has any trouble accurately representing a lower precision number.

Luck;
Ken

[Back to original message]