Re: Memory footprint differences between x86 and x64

I can't tell if this is creative spam, or a real question.

I've been getting scary spam latley with my home address in it (and various
other personal identifying things), so I know the spammer are getting more
creative.

On the other hand, the web site linked is in German, confirming that English
isn't Lukas's first language. Given his name, I was expecting italian,
but... who knows.

Overall, I vote +1 for this being a charming and heartwarming new type of
spam.

--
Chris Mullins, MCSD.NET, MCPD:Enterprise, MVP C#
http://www.coversant.net/blogs/cmullins

"Nip" <admin@xxxxxxxxxxxxx> wrote in message
news:CA17A1B6-F322-4478-98AD-1CF16F49657B@xxxxxxxxxxxxxxxx
Hi

Why it is like Chris Mullins wrote:
1. Every memory pointer needs 64 bits now instead of 32
2. The default integer size now 64 bits

--
Lukas Cavigelli
http://www.cavigelli.net

"Chris Mullins" <cmullins@xxxxxxxxx> schrieb im Newsbeitrag
news:eVzfDq2AHHA.3380@xxxxxxxxxxxxxxxxxxxxxxx
I've seen a number of people ask about memory footprint differences
between x86 and x64 over the last few months.

In the course of some recent testing, I jotted down the numbers for both
x86 and x64 with a very reproducable use case.

I ran my tests on a dual-core AMD Athlon with 4GB of memory running
Windows XP 64. To be certain of configuration, I explicity built my
projects as either x86 or x64, depending on which test I was doing.

In my test, I allocate 10,000,000 instances of a class that has 6 strings
inside it and add these instances into a List<MyClass>.

In short:
On x86 it took 1.55 GB of memory
On x64 it took 2.31 GB of memory.

The detailed results can be found here:
http://www.coversant.com/dotnetnuke/Default.aspx?tabid=88&EntryID=24

(Be sure to scroll down far enough to get to the "The Working Test")

This isn't a great test for memory footprint, so don't bet the company on
the results without doing your own verification, but I think the trends
my numbers show conintue to hold true in the general case.

--
Chris Mullins, MCSD.NET, MCPD:Enterprise
http://www.coversant.net/blogs/cmullins




.

Relevant Pages

  • Memory footprint differences between x86 and x64
    ... I've seen a number of people ask about memory footprint differences between ... In the course of some recent testing, I jotted down the numbers for both x86 ... and x64 with a very reproducable use case. ... I ran my tests on a dual-core AMD Athlon with 4GB of memory running Windows ...
    (microsoft.public.dotnet.framework.clr)
  • Re: Memory footprint differences between x86 and x64
    ... "Chris Mullins" schrieb im Newsbeitrag ... In the course of some recent testing, I jotted down the numbers for both x86 and x64 with a very reproducable use case. ... I ran my tests on a dual-core AMD Athlon with 4GB of memory running Windows XP 64. ...
    (microsoft.public.dotnet.framework.clr)
  • Re: x64 Terminal Server with x86 Applications
    ... One x64 advantage is you can have more memory on the TS, so you can support more sessions. ... With a x86 server there is a limit ... that this limit would be removed, but the x86 applications would ...
    (microsoft.public.windows.terminal_services)
  • p5n32-sli se deluxe and Vista x64 and 2gb (slow enough) then 4gb (horribly slow issues)
    ... Has anyone had trouble running Vista x64 on this motherboard.. ... I have a 1.86 Core2 CPU, the memory is the Corsair Twin2x2048-6400 ... opening windows was horrible.. ...
    (alt.comp.periphs.mainboard.asus)
  • Re: 16/32 processor operating mode
    ... Okay, x86 it is. ... hardware perspective, as I already mentioned, most memory accesses are ... called a "cache line") in a single operation. ... much of the data bus is active when accessing stuff on the bus. ...
    (alt.lang.asm)