implemented features (Re: Decision time....)

km b kmb810 at gmail.com
Mon Jun 4 17:42:43 PDT 2007


On 6/5/07, Michel Talon <talon at lpthe.jussieu.fr> wrote:

I can confirm what Rahul is saying, perhaps this is no suprise since we are
both theoretical physicists. I see 64 bits machines   being *twice* faster
than 32 bits machines (same machine, one in 32 bits Linux, the other in
64 bits Linux) on my computations, and particularly in symbolic maths
computations, e.g. running maple 32 bits and maple 64 bits, or floating
points computations. With this experience, i don't give any credence to the
computer people who pretend that there is no difference between the two
modes, or that you need >4 Gigs memory to see the difference. Our machines
have 2 Gigs and i see an enormous difference.
When you claim something generally, like 64-bit machines faster than
32-bit machines, it is but specific to your application and the area
you are working in. Applications that require heavy I/O would see no
or negligible performance increase. You don't need > 4G memory to see
the  difference, 64 bit processing mode solves many of the problems
related to x86 architecture to effectively utilize > 4G address space.
When you compare 64-bit machines 32-machines you'd definitely see
performance increase, due to enahnced register set and SSE instruction
set. However, try comparing the same applications performance on same
64-bit machine in two different environments, pure 64-bit environment
and pure 32-bit environment (which dragonfly is capable of) and let me
know what performance increase do you see?
And you can always run dragonfly on a 64-bit machine and it doesn't
mean the OS won't utilize the available resources of 64-bit
processors. The only difference is that you won't be able to utilize
4G address space and may be negligible performance decrease.
If you try to look at the DragonFly from the computer scientist's
perspective you'll realize that though it ain't ready, it is the way
an OS is meant to be.
I'd ask you guys from making such comments as the DragonFly ain't SMP
capable and won't be for a while just because you don't want to try it
(because it doesn't have fancy graphical installer of your famous
linux distro) . The negative comments like this have significant
impact on mindset of casual users who want to try and contribute to
the project.
Cheers
kmb
--
Something is wrong up on cloud # 9!




More information about the Kernel mailing list