Difference between pages "Benchmarking" and "Talk:Benchmarking"

(Difference between pages)
m (format console output the right way)
 
m (point at my benchmarking stubby commits for web infra for causes)
 
Line 1: Line 1:
== Power usage ==
+
Maybe some details should be added to powertop. I never figured out how to get it work. I read it takes some times to make an average of your power consumption. Some times I run it all day long, but never get any relevant information about power usage. -- [[User:Pytony|Pytony]] ([[User talk:Pytony|talk]])
sys-power/powertop can guess the computer's current power usage in watts.
+
<console>
+
###i## powertop
+
Power usage (ACPI estimate): 15.7W (0.5 hours)
+
</console>
+
  
== CPU ==
+
== MOAR ==
=== Compiler performance ===
+
i have some web benchmarking stuff at [[web-server-stack]]  do as you will...  i suggest chopping either all or the majority from the article relocating here, and linking to here to say benchmark ur site! [[User:Threesixes|Threesixes]] ([[User talk:Threesixes|talk]]) 21:16, October 18, 2014 (UTC)
<console>
+
###i## time emerge -1 coreutils
+
real    2m17.574s
+
user    1m46.572s
+
sys    0m53.450s
+
</console>
+
 
+
== RAM ==
+
=== ramspeed ===
+
<console>
+
###i## emerge -av ramspeed
+
$##bl## ramspeed -b2 -m4 && ramspeed -b1 -m4
+
INTEGER & READING      2048 Kb block: 5027.82 MB/s
+
INTEGER & READING      4096 Kb block: 4999.53 MB/s
+
...
+
INTEGER & WRITING      2048 Kb block: 2003.86 MB/s
+
INTEGER & WRITING      4096 Kb block: 1951.05 MB/s
+
</console>
+
 
+
== Hard drive ==
+
=== hdparm ===
+
<console>
+
###i## hdparm -tT /dev/sda
+
/dev/sda:
+
  Timing cached reads:  2318 MB in  2.00 seconds = 1159.52 MB/sec
+
Timing buffered disk reads: 648 MB in  3.00 seconds = 215.92 MB/sec
+
</console>
+
 
+
=== dd ===
+
<console>
+
$##bl## dd bs=1M count=512 if=/dev/zero of=temp conv=fdatasync
+
###i## echo 3 > /proc/sys/vm/drop_caches
+
$##bl## dd bs=1M count=512 if=temp of=/dev/null
+
536870912 bytes (537 MB) copied, 3.13319 s, 171 MB/s
+
..
+
536870912 bytes (537 MB) copied, 2.17444 s, 247 MB/s
+
$##bl## rm temp
+
</console>
+
 
+
== GPU ==
+
=== Warsow ===
+
==== Install ====
+
<console>
+
###i## emerge -av warsow
+
$##bl## mkdir -p ~/.warsow-1.0/basewsw/demos
+
$##bl## cd ~/.warsow-1.0/basewsw/demos
+
$##bl## wget http://www.phoronix-test-suite.com/benchmark-files/pts-warsow-10-1.zip
+
$##bl## unzip pts-warsow-10-1.zip
+
</console>
+
==== Benchmark ====
+
<console>
+
$##bl## warsow +exec profiles/high+.cfg +timedemo 1 +cg_showFPS 1 +cl_maxfps 999 +demo pts-demo10 +next "quit" +r_mode -1
+
899 frames, 76.4 seconds: 11.8 fps
+
</console>
+
 
+
=== Nexuiz ===
+
<console>
+
###i## emerge -av nexuiz
+
$##bl## nexuiz-glx -benchmark demos/demo1 -nosound 2>&1 {{!}} egrep -e '[0-9]+ frames'
+
1910 frames 42.0330749 seconds 45.4404063 fps, one-second fps min/avg/max: 35 46 63 (90 seconds)
+
</console>
+
 
+
=== Xonotic ===
+
<console>
+
###i## emerge -av xonotic
+
$##bl## xonotic-glx -benchmark demos/the-big-keybench 2>&1 {{!}} egrep -e '[0-9]+ frames'
+
12568 frames 556.0637400 seconds 22.6017255 fps, one-second fps min/avg/max: 19 23 27 (207 seconds)
+
</console>
+
 
+
=== Glxgears ===
+
Note: glxgears is a very basic OpenGL support test, it is not a real benchmark tool!
+
<console>
+
###i## emerge -av mesa-progs
+
$##bl## vblank_mode=0 glxgears
+
1583 frames in 5.0 seconds = 316.485 FPS
+
</console>
+
 
+
[[Category:HOWTO]]
+

Latest revision as of 21:16, October 18, 2014

Maybe some details should be added to powertop. I never figured out how to get it work. I read it takes some times to make an average of your power consumption. Some times I run it all day long, but never get any relevant information about power usage. -- Pytony (talk)

MOAR

i have some web benchmarking stuff at web-server-stack do as you will... i suggest chopping either all or the majority from the article relocating here, and linking to here to say benchmark ur site! Threesixes (talk) 21:16, October 18, 2014 (UTC)