AXP vs A64

[ME BIGGD01]

I found this article pretty good and will help some with the descision if they go to upgrade.

here ya go Link to article (http://www.viperlair.com/reviews/cpu_mobo/amd/3200_64/)

[Rivers]

Nice article bigg. I am considering an upgrade in the near future and reading that made me 2nd guess myself about buying a A64 754 and think more about the socket 939 but I guess I'll just wait a few months and see. I think waiting a bit more is a wise idea and it'll give me time to save more money. :oooo:

[ME BIGGD01]

it all comes down to budget. you can get an 754 chipset board cheap with alot of features. they are not that much slower than 939. one thing most don't realize about the a64 chips is that they have hardware inside the chip that prevents viruses. today that is a great selling point because no intel chips have that.

[Rivers]

I did not know that as well......in any case I'm still waiting a few more months before building my new rig so I'll have some time to decide. I am updating in the near future though for sure. :thumbs:

[JIMINATOR]

Originally posted by ME BIGGD01@Jul 6 2004, 12:08 AM
one thing most don't realize about the a64 chips is that they have hardware inside the chip that prevents viruses. today that is a great selling point because no intel chips have that.
uh, to be precise, what it actually prevents is buffer overflow exploits.
these are more common for unix. i don't think i have seen any
for windows, since outlook + express is such an efficient means
means of spreading viruses.

a buffer overflow exploit is, say when you submit a form for a webpage,
and the program only allows for a 4k response. if it gets stuffed with 10k of
data, that will overwrite the stack. then when that piece of the program
finishes instead of returning, it will jump to the embedded program
in the stack. it is easy to prevent with proper coding, but with multitudes
of programs, you never know.

the amd will be able to mark stack and data areas as non-executable,
so instead of running the bad code, it will fault and the program will
crash, which is probably a better outcome.

[ME BIGGD01]

are you saying that there is no such thing as a buffer overflow in windows?

i forgot to mention "it does in the article" that xp will support this feature in sp2 which (fingers crossed) should be this month.

[JIMINATOR]

no, any program can be subject to buffer overflow.
but a virus writer would need to be able to input data to the
program somehow, and unless it is some type network program
(mail, ftp, http) the access is not going to be available.
for normal programs, if they can input data, then the system
is already compromised, although the exploit could possibly
be used for something like getting admin permissions,
but again, this is more common for unix systems.