[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#176267: ITP: mplayer -- Mplayer is a full-featured audioand video player for UN*X like systems



On Wed, Jan 29, 2003 at 07:44:39AM +0100, Gabucino wrote:
> Andrew Suffield wrote:
> > wasted"? These figures will change almost randomly depending on what
> > else is running on the host at the same time.
> That's why I pasted the Idle time.

Eh? That doesn't stop the figures changing in an unconstrained manner.

> > They certainly don't reflect how an application will perform in the cases you
> > care about,
> Talk about exact things!

I don't know what *you* care about, but I'm fairly sure that the
numbers reported by top when running aaxine do not feature highly
amoung them.

> > especially if (like xine and mplayer) they adapt their performance profile
> > when system load is high - which any decent multimedia application should be
> > doing.
> "adapt their performance profile" ?? :)) It's called framedropping :)

Uhh? There are a whole range of things that multimedia applications
can do to adapt their performance profile. Dropping frames is one of
the simplest and most ineffective of them. Good applications will
dynamically vary the quality of their output as necessary.

> > This is closer to the domain of real benchmarks, but it's still at the
> > sort of level you get from a marketing department.
> ROTFL :)
> 1. __you are free to reproduce it__
> 2. you failed to point out _what_ and _why_ is wrong in benchmarking a gcc
>    compilation to get Idle CPU time, just say "marketing". :) I can't think
>    you ever did any benchmarks, sorry..
> 
> So unless you say _reasons_ why the benchmark was bad, please don't answer.

You quoted one of the more significant ones immediately below (ie,
what I pointed out before, regardless of you claiming I didn't).

> > (One golden rule is that if a benchmark was quoted without giving the
> > variance, it's neither objective nor useful).
> Variance?

Oh dear. Some basic knowledge of statistics is absolutely essential in
order to perform meaningful benchmarks.

> > > So thanks for your mail, but next time you doubt I can read 'top' output,
> > > think twice.
> > Uhh, the whole point was that top is not a useful benchmarking
> > tool. What mail were you reading?
> Ok, if you tell me that top is lying when it writes "0% CPU is idle when
> xine is running", then you better don't send that mail at all.

You really haven't actually read anything I have written, have you?

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ | Dept. of Computing,
 `. `'                          | Imperial College,
   `-             -><-          | London, UK

Attachment: pgprut8D7lkfM.pgp
Description: PGP signature


Reply to: