Re: 64 bit or 32 bit
From: Christopher Browne (cbbrowne_at_acm.org)
Date: 04/08/04
- Next message: phil-news-nospam_at_ipal.net: "Re: Creating a file that self-destroys at each reboot"
- Previous message: Vasili I Galchin: "Re: TCB for pthread in linux kernel 2.4"
- In reply to: Måns Rullgård: "Re: 64 bit or 32 bit"
- Next in thread: Måns Rullgård: "Re: 64 bit or 32 bit"
- Reply: Måns Rullgård: "Re: 64 bit or 32 bit"
- Reply: Kasper Dupont: "Re: 64 bit or 32 bit"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: 8 Apr 2004 02:52:29 GMT
In the last exciting episode, mru@kth.se (Måns Rullgård) wrote:
> Grant Edwards <grante@visi.com> writes:
>
>> On 2004-04-07, Villy Kruse <vek@station02.ohout.pharmapartners.nl> wrote:
>>
>>>> There is no relationship required between the size of integral
>>>> types and the size of pointers. That's how Microsoft can get
>>>> away with sizeof(void*)==8 but sizeof(int)==4 in Win64. A
>>>> mistake, in my opinion, but once again they didn't check with
>>>> me first.
>>>
>>> A mistake alsof implemented by among others ny Sun
>>> Microsystems. Anyway, for several reasons we sometimes do
>>> need a 4 byte interger type so if short is 2 bytes and int 8
>>> bytes then the 4 byte integer type will be missing.
>>
>> unsigned long short aFourByteVariable;
>
> So when processors become 256-bit, what will we get?
>
> char = 8 bits
> short short = 16 bits
> short = 32 bits
> long short = 64 bits
> int = 128 bits
> long = 256 bits
I don't think we'll head there _any_ time soon.
The leap from 16 to 32 bits took quite a while, but was exceedingly
worthwhile because 32 bits of addressing space and 32 bits of
"resolution" were a whole lot more useful than 16 bits.
Since then, it has taken _years_ for people to get much interested in
the transition to 64 bits. It was only in the last year that it
became easy to get computer systems with more memory than 32 bits can
address, and while 32 bits isn't enough to represent Bill Gates' net
worth, 64 bits is plenty sufficient.
The reason to jump to 128 or 256 bits would be because 64 bits could
not address enough memory. When you consider that 2^64 is an
astoundingly enormous quantity, further leaps would seem unnecessary
in any sort of short term.
--
output = reverse("moc.enworbbc" "@" "enworbbc")
http://cbbrowne.com/info/x.html
"In other words -- and this is the rock solid principle on which the
whole of the Corporation's Galaxy-wide success is founded -- their
fundamental design flaws are completely hidden by their superficial
design flaws." -- HHGTG
- Next message: phil-news-nospam_at_ipal.net: "Re: Creating a file that self-destroys at each reboot"
- Previous message: Vasili I Galchin: "Re: TCB for pthread in linux kernel 2.4"
- In reply to: Måns Rullgård: "Re: 64 bit or 32 bit"
- Next in thread: Måns Rullgård: "Re: 64 bit or 32 bit"
- Reply: Måns Rullgård: "Re: 64 bit or 32 bit"
- Reply: Kasper Dupont: "Re: 64 bit or 32 bit"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
