I have access to a 3000+ Athlon64, and it's incredibly snappy.
It has no bottlenecks, however. It has a fast hard disk, lots of ram, etc.
Have you tried running something other than SuSE on it?
I've been using a 2400+ 32 bit Athlon system for about a year. I had to upgrade my wife's computer, so I decided to buy me a 3400+ 64 bit Athlon setup, and she can have my old system which is plenty fo her needs. So I got my 64 bit Athlon system set up with the 64 bit version of SuSE 9.1 and I'm kinde disappointed with the speed. I do see some speed improvement over the previous setup, but nothing dramatic. I expected close to twice the speed (32 ->64). Is this normal? Am I doing something wrong? What is your experince? My observation is based on how the system feels and responds not on any benchmark tests.
I have access to a 3000+ Athlon64, and it's incredibly snappy.
It has no bottlenecks, however. It has a fast hard disk, lots of ram, etc.
Have you tried running something other than SuSE on it?
I'm getting SuSE 9.2 in a few days, but I doubt the distro has anything to to with it. I just want to know what you guys' experience is with 64 bit CPUs.
Well I have no experience with it (yet!) but I'm consindering it.
As Far As I Know, 32->64 doesn't really mean you are going to have double speed improvements. The bits is just floating point calculations. Data now crunch twice more the size than before so it finished faster. That also is largely depends on your software's configuration. If you have 64 bits machine but your distro is not optimized for it, you won't notice the speed improvements. Despite its being advertised as 64 bits, SuSE's most programs still run in 32 bits mode (IIRC) and so is many others. The only OS part of it is 64 bits. I heard the same holds true for RH's.
So I'm guessing that until a few years down the road, we won't be seeing any good 64 bits softwares. Today's major applications such as MySQL, Apache, Oracle all run in 32 bits only although there are some works being done to move over to 64 bits.
This addresses your questions and confirms some of what Compunuts said.
http://arstechnica.com/cpu/03q1/x86-64/x86-64-2.html
Dynamic range: Work with bigger numbers in smaller times, address more memory without PAE (and hence no overhead of mapping the lower mem to the higher addresses)
From article: there's no magical performance boost inherent in the move from 32 bits to 64 bits, as people are often led to think by journalists who write things like, "64-bit computers can processes twice as much data per clock cycle as their 32-bit counterparts." Technically, this is true in a very restricted sense, but it would be better to say the following: "64-bit computers can process numbers that are 4.3 billion times as large as those processed by their 32-bit counterparts." It sounds a lot less sexy because it is, but at least no one is misled into thinking that 64-bitness makes a computer somehow twice as fast.
So the only advantage is that it can work with bigger numbers and address more memory. My subject says it all. I should have done my homework before jumping in. Ah well. At least anyone reading this thread won't set him/herself up for the same disappointment. :![]()
Me too. I got the MSI k8t neo with a athlon 64 3000+ for gaming. But its 32 bit XP, so I didn't expect wonders, just a nice running machine, which it is. I'm more reluctant now going to a 64 bit linux on that box based on your findings and Compunuts input. Maybe just for gee wiz.
Would a system running Gentoo be good for 64bit computing? Doesn't emerge d/l and complie everything from scratch... so everything would be optimized for 64bits, right?
compiling would got a bit faster but your programs wont be 64bit unless the code was written for 64bit cpus.
someone correct me if i'm wrong ;D I just woke up
Bookmarks