On Wed, May 19, 2010 at 7:18 PM, Aaron Griffin <firstname.lastname@example.org> wrote:
> On Wed, May 19, 2010 at 12:07 PM, Dan McGee <email@example.com> wrote:
>> On Wed, May 19, 2010 at 12:02 PM, Aaron Griffin <firstname.lastname@example.org> wrote:
>>> Came across my reader today
>>> http://www.phoronix.com/scan.php?page=article&item=ubuntu_arch_faster&num= 1
>>> Pretty neat.
>> Showing boot-up time and other more noticeable waits would be a much
>> better comparison; I think the benchmarks are not why people think
>> Arch feels faster.
> Right but the reason I think this is interesting is that the data
> shows it is "feels faster" not "is faster".
I basically agree with everything that was said and will try to
re-conciliate the different opinions :
1. This benchmark is not particularly useful, as we are often used
with Phoronix. They are often not well chosen and/or with bad
interpretation, so it often annoys people who know better about a
2. Phoronix's benchmarks have the merit to exist and to provide facts,
as opposed to many users who claims X is faster than Y, and NEVER
provide any facts/numbers/benchmarks, even when specifically asked. I
saw it happening many times in Arch community over the time, so now we
at least have some data to show them.
So I would still thank phoronix for providing benchmarking tools and
results, which are definitely useful and needed.
IMO the most interesting benchmarks are the ones comparing different
versions of the same software, but this would also better be made in
collaboration with upstream developers, as these are probably the only
people able to really make sense and explain results.
In any cases, thanks for the link Aaron