| View previous topic :: View next topic |
| Author |
Message |
discomfitor l33t


Joined: 21 Feb 2003 Posts: 927 Location: None
|
Posted: Sat Nov 13, 2010 11:33 am Post subject: |
|
|
oh wow this thread is old _________________ There is no substitute for experience.
Imperfection indicates a lack of effort. |
|
| Back to top |
|
 |
skellr l33t


Joined: 18 Jun 2005 Posts: 698 Location: The Village
|
Posted: Sat Nov 13, 2010 1:25 pm Post subject: |
|
|
| You know your box is fast when it doesn't have time to break. |
|
| Back to top |
|
 |
Marctraider Guru


Joined: 24 Dec 2003 Posts: 387
|
Posted: Wed Nov 24, 2010 5:15 pm Post subject: |
|
|
lol indeed very old.
I must laugh about this ricer shit and all, however. I must add to the 'hdparm' section that I actually had some speed optimization from a certain parameter which was probably disabled in the BIOS on my old Toshiba (Pentium) laptop which I couldn't change there. It sped up my harddrive by almost half the speed. Since the IDE controller only worked in Programmed Input/Output (PIO) mode i needed all speed i could get. (were it to be 30-40% at least).
I had to enable the 32-bit transfer mode. This seemly wasnt enabled by default at that time by kernel.
My write/read speed went from 3~MB/s to like 6MB/s.
But it was not the processor which was holding me down. It was primarily the harddisk/harddisk space and memory usage which I had to 'optimize'. (-Os for the win!) lol
Now thats some speed increase dont you think? hehe.
But seriously, at a time i could consider myself a beginning ricer. But I stopped with it as soon as I got a faster computer. LOL. _________________ MOBO: Maximus II Gene
RAM: DDR2 OCZ 4GB
CPU: E6400 Conroe
GPU: HD2600XT
SATA: 3x 250GB. |
|
| Back to top |
|
 |
Yamakuzure l33t

Joined: 21 Jun 2006 Posts: 779 Location: Hamburg, Germany
|
Posted: Thu Nov 25, 2010 10:49 am Post subject: |
|
|
I bet the "ricing" discussion will have a very noisy "comeback" when GCC-4.5 is marked stable and people realize what Graphite can do.  _________________
| Kernel 3.3.1 sources wrote: | | drivers/net/ethernet/realtek/r8169.c:74 #define SafeMtu 0x1c20 /* ... actually life sucks beyond ~7k */ |
|
|
| Back to top |
|
 |
Ormaaj Guru

Joined: 28 Jan 2008 Posts: 312
|
Posted: Mon Nov 29, 2010 7:30 am Post subject: |
|
|
| Yamakuzure wrote: | I bet the "ricing" discussion will have a very noisy "comeback" when GCC-4.5 is marked stable and people realize what Graphite can do.  | Graphite has been around since 4.4 and you still have to explicitly enable the flags as of 4.5 AFAICT.
This box:
CFLAGS="-march=native -O3 -mmmx -floop-interchange -floop-strip-mine -floop-block -pipe"
Makes it look ricey. |
|
| Back to top |
|
 |
Yamakuzure l33t

Joined: 21 Jun 2006 Posts: 779 Location: Hamburg, Germany
|
Posted: Thu Dec 09, 2010 3:07 pm Post subject: |
|
|
Of course. But I was merely thinking of: | Code: | CFLAGS="-march=core2 -O2 -pipe -mmmx -msse -msse2 -msse3 -mssse3 -mcx16 -msahf -mtune=core2 -mfpmath=sse -ftree-vectorize -mno-push-args -floop-interchange -fgraphite-identity -ftree-loop-distribution -floop-parallelize-all -floop-strip-mine -floop-block -ftree-parallelize-loops=5"
CXXFLAGS="${CFLAGS}" |  _________________
| Kernel 3.3.1 sources wrote: | | drivers/net/ethernet/realtek/r8169.c:74 #define SafeMtu 0x1c20 /* ... actually life sucks beyond ~7k */ |
|
|
| Back to top |
|
 |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|