I compared DDB2 and Foobar2000 mobile playing the same album using equalizer and replaygain on Android 8.1. The file format was musepack and at the end of the album DDB2 had used 8% battery and Foobar2000 mobile had used 15%. The phone was in battery saving mode the whole time and the screen was turned off and the volume was the same. What might the reason be for the almost double battery usage when using Foobar2000 mobile? Why I did this test is because if I charge my phone fully it will be completely drained by the time my eight hour work day is over and it looks like for the time being I will be using DDB2 because of this but I wish to get back to Foobar2000 mobile as soon as it uses way less battery.
Maybe, difference is caused by equalizer. Can you test without it?
I noticed that the first test also had downmix channels to mono enabled so I deactivated it and also the equalizer but now the battery usage was more (24%).
On which device is a simple float to integer conversion operation so expensive?
I have a Nexus 5X.
Thanks for taking your time to document the battery usage. I will see what I can do about it for a future update.
One more thing: if you could please try toggling OpenSL output in advanced settings and check the impact on the battery life then. There's a chance that OpenSL will use less battery power, but it can't be enabled by default as it doesn't work correctly on specific old devices.
I have been using OpenSL enabled but I can try once without it.
Without eq, downmix to mono and using no OpenSL it used 23%.
I tried VLC with its equalizer enabled and it only used about 5-7%. The proböem with VLC though is that it doesnt't have a 15 band graphical eq and mono output.