newsAdvanced Gtk+ Sequencer - News: Inaccurate attack calculation of bpm

 
 
Latest News
Refactored key c source files posted by jkraehemann, Thu 19 Apr 2018 06:49:50 PM UTC - 0 replies
Extended semantics and some removals posted by jkraehemann, Sun 08 Apr 2018 05:10:13 PM UTC - 0 replies
The new dispatcher: ags_channel_recursive_run_stage() posted by jkraehemann, Sun 18 Mar 2018 05:37:45 PM UTC - 0 replies
GSequencer development insights posted by jkraehemann, Sun 04 Mar 2018 02:15:03 PM UTC - 0 replies
GSequencer v2.0.0 - forecast posted by jkraehemann, Tue 06 Feb 2018 03:04:57 PM UTC - 0 replies
[Submit News]
[57 news in archive]

Inaccurate attack calculation of bpm

Item posted by Joël Krähemann <jkraehemann> on Sat 19 Aug 2017 11:56:06 PM UTC.

As having bigger buffers especially by using pulseaudio, the algorithm that calculates attacks of target buffer seems to fail.

The attack is calculated by one of these functions:

  • ags_devout_adjust_delay_and_attack()
  • ags_jack_devout_adjust_delay_and_attack()
  • ags_pulse_devout_adjust_delay_and_attack()

The goal of the algorithm would be having constant delay but alternating attacks. This formula was considering only buffer lengths of 512 or lower.

Rework the algorithm

The computation shall return an approximation attack with a certain tolerance of deviation. I would say it doesn't matter if there is a swing of 10 to 20 frames each delay.

It should figure out best fit of delay and approximated attack. The problem with bigger buffer lengths is the algorithm can't smaller delays accurately. This was done for simplicity but doesn't meet current requirements, yet.

By Joël

No messages in Inaccurate attack calculation of bpm

 

Back to the top


Powered by Savane 3.1-cleanup1