[MPlayer-dev-eng] Re: Compile options

[Andrew Savchenko]

> > > Could you tell me which decoder you are testing?
> >
> > Yes. It is mpeg-4 stream, decoder ffodivx, vfm ffmpeg
>
> That's why you couldn't confirm my results: Only (?) ffh264 is
> affected by -O2.

Why do you think so (that -O2 affects only ffh264)?

Moreover, I've make h264 short film (2 minutes) as I promised earlier.
Your results are uncormfimed again. On my system playback of h264 movie 
(using [ffh264] vfm: ffmpeg (FFmpeg H.264)) is clearly faster with -O4 
than with -O2.

Here is the main results:
********************************************************
Results for normal distribution, reliability 0.95, 10 measurements,
t(0.95,10)=2.26:
======================================
for -O4:

BENCHMARKs: VC:   (17.168 \pm 0.015)s VO: (61.302 \pm 0.031)s  A:  
0.000s  Sys:  (0.243 \pm 0.001)s   = (78.71 \pm 0.04)s

real    (1m18.86s \pm 0m0.04s)
user    (0m52.24s \pm 0m0.11s)
sys     (0m0.414s \pm 0m0.027s)

user+sys     (0m52.65s \pm 0m0.11s)

======================================
for -O2:

BENCHMARKs: VC:   (17.254 \pm 0.014)s VO: (61.487 \pm 0.015)s  A:  
0.000s  Sys:  (0.244 \pm 0.001)s   = (78.976 \pm 0.017)s

real    (1m19.17s \pm 0m0.11s)
user    (0m52.50s \pm 0m0.09s)
sys     (0m0.411s \pm 0m0.032s)

user+sys     (0m52.91s \pm 0m0.10s)

======================================
diffrences in performance between O2 and O4:
\Delta(O2-O4):

BENCHMARKs: VC:   (0.086 \pm 0.021)s VO: (0.185 \pm 0.034)s      A:  
0.000s  Sys:  (-0.001 \pm 0.001)s   = (0.266 \pm 0.043)s

real    (0m0.31s \pm 0m0.12s)
user    (0m0.26s \pm 0m0.14s)
sys    (-0m0.003s \pm 0m0.042s)

user+sys     (0m0.26s \pm 0m0.15s)
********************************************************

Full test logs are in attachment (all conditions are the same as 
earlier).
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test.tar.bz2
Type: application/x-tbz
Size: 3178 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20060918/2ffd502c/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20060918/2ffd502c/attachment.pgp>