First of all, thanks for the implementation! It looks very promising
and I like the macro/memory operator overload approach.

2014-08-22 20:20 GMT+02:00 Vesa <[email protected]>:
> Early results seem promising... I think I can notice some performance
> improvement, although haven't done any real benchmarking (not going to
> bother either, but if anyone wants to that'd be nice).

A good test case usually is to render a reference project at the
command line and measure the consumed and elapsed CPU time using the
"time" command. You should be able to measure notable improvements.
For small improvements oprofile could be an option even though it's a
little bit harder to set up.

Besides this it's probably harder to measure improved RT capabilities.
We could use the timer which is used for the CPU meter already to
profile the elapsed time for each mixer period and write the results
into a CSV file. Plotting the different CSV files could/should reveal
differences.

Toby

------------------------------------------------------------------------------
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
_______________________________________________
LMMS-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lmms-devel

Reply via email to