nandinga at gmail.com
Fri Oct 7 14:37:37 PDT 2011
Hmm yes... There are more comments about other strange results, and about
the conditions of the tests not being clear enough (32 or 64bit userland?).
Anyway I think there are clear benefits in using your whole processor and
architecture. My concern is mostly about compatibility and stability of the
software. Still have the feel that most tests have been done on 32bit
environments, and that's why I'm asking here.
But looks like the people using 64bit here is happy about it, so I'm givin'
it a try!
On Fri, Oct 7, 2011 at 4:09 PM, Brett Clark <brett.clark at zirous.com> wrote:
> It seems odd, but the test shows the 64-bit kernel as running disk
> transactions vastly faster than the 32-bit (page 5). Its using PostMark
> which is a filesystem benchmarking utility.
> 64-bit - 2297 TPS
> 32-bit - 235 TPS
> How in the world does a 64-bit cpu make your disk reads/writes happen 9x
> faster? How did that not raise a big red flag for the person performing the
> Since the tester didnt put in any real comments about the hugh difference
> in a highly unexpected area, im a big skeptical about all of their tests.
> On Thu, Oct 6, 2011 at 6:34 PM, Nando <nandinga at gmail.com> wrote:
>> Hi all!!
>> Given that we are in the mood these days, I have a question to ask. I've
>> seen this benchmarks
>> and started asking myself, why are we still stuck on 32bits? Is anybody
>> using 64bit linux distro for audio? Its only me stuck?? hehe...
>> Those benchmarks show HUGE differences in performance, and from the
>> comments I've learned that what you get is to use all registers in the CPU,
>> which are not used when running 32bit soft.
>> Ardour-Users mailing list
>> ardour-users at lists.ardour.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ardour-Users