I get <br><br><span>Using 768743 loops per ms, running every load for 30 seconds Benchmarking kernel 2.6.15-suspend2-r6 at datestamp 200603030849<br> --- Benchmarking simulated cpu of Audio in the presence of simulated ---
<br>Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.013 +/- 0.0135 0.019 100 100<br>Video 0.019 +/- 0.0279 0.476 100 100 X
0.035 +/- 0.126 2 100 100<br>Burn 0.012 +/- 0.0123 0.018 100 100 Write 0.085 +/- 1.25 30.5 100 100 Read 0.03 +/- 0.0319
0.097 100 100<br>Compile 0.14 +/- 1.11 17.3 100 100 Memload 0.051 +/- 0.262 4.2 100 100<br> --- Benchmarking simulated cpu of Video in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None
0.032 +/- 0.0331 0.045 100 100 X 0.035 +/- 0.15 2 100 100 Burn 0.012 +/- 0.0124 0.018 100 100<br>Write 0.036 +/- 0.398
16.7 100 99.9 Read 0.025 +/- 0.0274 0.109 100 100<br>Compile 0.112 +/- 1.54 50 99.8 99.7 Memload 0.055 +/- 0.77 30.7 100
99.9<br> Using 768743 loops per ms, running every load for 30 seconds Benchmarking kernel 2.6.15-suspend2-r6 at datestamp 200603030957 --- Benchmarking simulated cpu of Audio in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met
<br>None 0.014 +/- 0.0142 0.018 100 100<br>Video 0.031 +/- 0.0317 0.042 100 100 X 0.041 +/- 0.136 2 100 100<br>Burn
0.012 +/- 0.0125 0.038 100 100 Write 0.062 +/- 0.713 17.5 100 100<br>Read 0.031 +/- 0.0327 0.091 100 100 Compile 0.048 +/- 0.413
8 100 100<br>Memload 0.041 +/- 0.141 3.07 100 100 <br>--- Benchmarking simulated cpu of Video in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met
<br>None 0.013 +/- 0.014 0.018 100 100 X 0.033 +/- 0.155 2 100 100<br>Burn 0.012 +/- 0.0126 0.039 100 100 Write
0.065 +/- 1.57 66.4 99.9 99.9<br>Read 0.023 +/- 0.0258 0.124 100 100 Compile 0.064 +/- 0.64 16.7 100 99.9<br>Memload 0.056 +/- 0.421
10.2 100 100 <br>--- Benchmarking simulated cpu of X in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0 +/- 0.00132
0.015 100 100<br>Video 0 +/- 0.00331 0.035 100 100<br>Burn 0 +/- 0.00136 0.016 100 100<br>Write 0 +/- 0.00315 0.037 100 100
<br>Read 0 +/- 0.00284 0.035 100 100<br>Compile 0 +/- 0.00212 0.029 100 100<br>Memload 0 +/- 0.00519 0.065 100 100<br><br>
--- Benchmarking simulated cpu of Gaming in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU<br>None 0 +/- 0 0 100<br>Video 0 +/- 0 0 100
<br>X 0 +/- 0 0 100<br>Burn 85.3 +/- 198 568 54 Write 0.054 +/- 1.3 31.1 99.9<br>Read 0 +/- 0 0 100 Compile
93.7 +/- 185 676 51.6<br>Memload 0 +/- 0 0 100</span><br><br>... note that the only time I get less than 99% in the right hand column is in the Gaming section. Your 83% Video in the presence of X could well be related to your MythTV problem.
<br><br>Which I/O scheduler are you using (I'm using CFQ).<br><br><br><div><span class="gmail_quote">On 3/15/06, <b class="gmail_sendername">Nicklas Bergfeldt</b> <<a href="mailto:nicklas.bergfeldt@his.se">nicklas.bergfeldt@his.se
</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Nick Crabtree wrote:<br><br>> To see whether buffering is the issue, you could try activating laptop
<br>> mode, which groups disk write activity into "chunks" that are spaced<br>> at larger intervals than they normally would be. Although this is<br>> intended as a power saving idea to let a laptop's hard drive spin
<br>> down, it would in this case check whether the 30second interval jumps<br>> are due to disk cache flushes. emerge laptop-mode-tools to install it,<br>> and run /usr/sbin/laptop_mode start to start it.<br><br>
<br>I tried that and it did not have any effect :-(<br><br><br>> Of course, if you are running other stuff and don't have the free RAM<br>> for more than 30 seconds of TV then you will still get disk activity<br>> and laptop mode won't help. But on a quiet system (no web browser or
<br>> email client running, for example) then you should have plenty of<br>> spare RAM for this.<br><br><br>That should not be any problem since this computer is solely used for<br>mythTV...<br><br><br>> You might also want to try benchmarking using interbench
<br>> <a href="http://members.optusnet.com.au/ckolivas/">members.optusnet.com.au/ckolivas/</a><br>> <<a href="http://members.optusnet.com.au/ckolivas/">http://members.optusnet.com.au/ckolivas/</a>>*interbench*/
<br><br><br>Would this indicate that I need faster hardware, or is it enough for mythTV?<br><br>Sky /usr/src/interbench/interbench-0.30 > make<br>gcc -W -Wall -g -O2 -s -pipe -c -o interbench.o interbench.c<br>interbench.c
:1404:18: warning: unknown escape sequence '\]'<br>gcc -W -Wall -g -O2 -s -pipe -c -o hackbench.o hackbench.c<br>gcc -lrt -lm interbench.o hackbench.o -o interbench<br>Sky /usr/src/interbench/interbench-0.30 > ./interbench
<br>loops_per_ms unknown; benchmarking...<br>514666 loops_per_ms saved to file interbench.loops_per_ms<br>Creating file for read load...<br><br>Using 514666 loops per ms, running every load for 30 seconds<br>Benchmarking kernel
2.6.15-gentoo-r1 at datestamp 200603151111<br><br>--- Benchmarking simulated cpu of Audio in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met<br>None 0.007 +/-
0.0153 0.218 100 100<br>Video 0.153 +/- 0.264 0.517 100 100<br>X 1.42 +/- 3.15 12 100 100<br>Burn 0.005 +/- 0.00633
0.016 100 100<br>Write 0.033 +/- 0.178 4.02 100 100<br>Read 0.024 +/- 0.025 0.071 100 100<br>Compile 0.053 +/- 0.364 8 100 100
<br>Memload 0.128 +/- 0.705 8.01 100 100<br><br>--- Benchmarking simulated cpu of Video in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met
<br>None 0.005 +/- 0.00574 0.023 100 100<br>X 4.15 +/- 9.3 33.3 99.1 83.4<br>Burn 0.006 +/- 0.0123 0.127 100 100<br>Write
0.028 +/- 0.186 5.75 100 100<br>Read 0.023 +/- 0.0281 0.22 100 100<br>Compile 0.036 +/- 0.491 20.7 100 99.9<br>Memload 0.091 +/-
0.87 24.7 100 99.8<br><br>--- Benchmarking simulated cpu of X in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met<br>None 0 +/-
0.00283 0.032 100 100<br>Video 0.003 +/- 0.0386 0.476 100 100<br>Burn 0 +/- 0.000872 0.012 100 100<br>Write 0.214 +/- 1.21 10
93.8 92<br>Read 0.036 +/- 0.42 6 99.3 98.7<br>Compile 0.202 +/- 1.04 8 96.2 93.3<br>Memload 0.149 +/- 0.983 10 92.1
90.7<br><br>--- Benchmarking simulated cpu of Gaming in the presence of simulated ---<br>Load Latency +/- SD (ms) Max Latency % Desired CPU<br>None 0.001 +/- 0.0216 0.354 100<br>Video 0.008 +/-
0.157 2.94 100<br>X 12.9 +/- 31.3 208 88.6<br>Burn 195 +/- 269 367 33.9<br>Write 33.7 +/- 99.2 818 74.8<br>Read 6.27 +/-
6.71 10.3 94.1<br>Compile 326 +/- 395 570 23.5<br>Memload 15.7 +/- 33.7 191 86.4<br><br><br>Sky /usr/src/interbench/interbench-0.30 ><br><br><br>/Nicklas<br>
_______________________________________________<br>mythtv-users mailing list<br><a href="mailto:mythtv-users@mythtv.org">mythtv-users@mythtv.org</a><br><a href="http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users">http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-users
</a><br></blockquote></div><br><br clear="all"><br>-- <br>Nick Crabtree<br><a href="mailto:nickcrabtree@gmail.com">nickcrabtree@gmail.com</a>