Some more linear HW VC thoughts
output_alsa.c set_mixer function on alsa.mixer_linear definitely fails its purpose. No. It fails. Because the inputs ldB rdB are basically of no use. Then I think we're faceing "over-logarithmic" data. 1. The 16.16 fixed point left/right input from slimproto is not linear to start with - I guess the associated VC curve is already logarithmic ( if anybody can point me where this is done on LMS - don't hesitate to do so) 2. The values of this logarithmic curve are run over "ldB = 20 * log10( left / 65536.0F );" which gives us another logarithmic curve. 3. And then these double logarithmic values are applied against a HW VC, where the driver offers another round of HW based logarithmic VC values. This I wouldn't call linear. Would anybody agree to this? Then I'm wondering if the 16.16 fixed point values from slimproto would cover a 100dB+ VC range as offered by modern DACs. The 72dB range from ancient slim times would work of course. Of course I understand that replaygain and crossfade are causing another major challenge. I do think that on a linear HW VC control replaygain and crossfade won't work properly anyhow and could therefore be bypassed. Live would be much easier, if the linear LMS VC data (1-100) could be fetched through slimproto. Of course we could all go back and enjoy the our SL software VC control. Though I do wonder now if that one has a double logarithmic logic applied!?!? Or perhaps I misunderstood the whole subject. If so. Please apologize for bugging you. :::' my audioblog - latest series: RaspBerry PI - \"The Audio Engine\" ' (http://soundcheck-audio.blogspot.com)::: ------------------------------------------------------------------------ soundcheck's Profile: http://forums.slimdevices.com/member.php?userid=34383 View this thread: http://forums.slimdevices.com/showthread.php?t=97046 _______________________________________________ unix mailing list [email protected] http://lists.slimdevices.com/mailman/listinfo/unix
