[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [fluid-dev] Three thoughts

From: josh
Subject: Re: [fluid-dev] Three thoughts
Date: Wed, 28 Oct 2009 13:51:55 -0700
User-agent: Internet Messaging Program (IMP) H3 (4.1.6)

Quoting David Henningsson <address@hidden>:

* I had a look at the dev docs and corrected some minor things, looks
much better now that Josh has got rid of all the internal stuff, good
work. "Data structure index" looks funny though with only F and _ as
letters. :-)

Lets blame Doxygen for that.  F'ing F and _!  ;)

* I thought Jack had a rtprio of 70 by default, so we should probably
put ourselves at 60 (audio) and 50 (midi)? Should we remove the
booleans and let a value of 0 mean rtprio turned off?

Both suggestions sound good to me. When using Jack though, those settings aren't actually used, since Jack is responsible for calling the callback using its own priority. But right now the ALSA sequencer driver would indeed be at a higher priority if used with Jack, which is of course bad.

I think the rtprio was somewhat arbitrarily chosen to be 90 and I was just continuing to use it, without much thought. 60 and 50 sound fine to me, with 0 meaning off (get rid of booleans). I'll do that.

* Did we break synth.dump and thus the -d switch? Or has it never worked?

Hmm. I just noticed that there is indeed code for the dump option at least when enabled by the -d option. It just uses the fluid_midi_dump_postrouter() and fluid_midi_dump_prerouter() functions for the MIDI router, which displays events. It seems like the synth.dump option is completely useless though, except to get status of whether -d was supplied or not, which also seems less than useful. Perhaps we should get rid of the synth.dump setting.

Its a little strange having verbose and dump too. Right now the verbose option will display output inside of the synthesis thread too, which is probably not good. Not knowing exactly what the option was for, I just assumed it would only be enabled for debugging purposes.

// David


reply via email to

[Prev in Thread] Current Thread [Next in Thread]