hello,

a very dirty solution to restard the DSP when audio stuck is to make a kind of watchdog :

the attached patch is very basic but restart audio a few second after if stuck.

hope that help.

cyrille

Lex Ein wrote:

Can Pd detect its own console output errors? If so, how? Can Pd detect its own DSP state (on/off)? If so, how?

After about 2 hours of operation, I'm getting an "audio stuck" message; a fix is to stop and start DSP.

I'd like to DETECT the "audio stuck" condition, and only then issue DSP stop/start.

Dual 1GHz Xeon, Win2K fully updated, Pd 0.37.1

Of course, ultimately, I'd like the "audio stuck" messages to go away. I'm stuck with mmio drivers - asio doesn't start. No big, I guess.

---

PD-list mailing list PD-list@iem.at to manage your subscription (including un-subscription) see http://iem.at/cgi-bin/mailman/listinfo/pd-list

#N canvas 644 320 662 530 12; #X obj 23 108 snapshot~; #X obj 42 72 metro 1000; #X obj 23 136 change; #X obj 23 162 t b; #X obj 23 224 delay 3000; #X msg 23 316 ; pd dsp 0 ; pd dsp 1; #X obj 38 189 bng 15 250 50 0 empty empty empty 0 -6 0 8 -262144 -1 -1; #X obj 23 19 osc~ 0.1; #X obj 42 45 loadbang; #X text 117 102 the output of the snapshot should change every 1000ms unless audio is off.; #X text 118 185 bang every 1000ms when audio is on; #X text 116 335 restart if audio stuck; #X text 122 15 osc~ output value only if audio is on; #X text 117 211 test if audio is still on (a bang restart the delay to 3000ms , so the delay output a bang only if it does not receive input for 3000 ms , i.e only if audio stop); #X obj 57 284 loadbang; #X msg 96 408 ; pd dsp 0 ;; #X text 186 417 try this : audio sould restat in 3000ms; #X connect 0 0 2 0; #X connect 1 0 0 0; #X connect 2 0 3 0; #X connect 3 0 4 0; #X connect 3 0 6 0; #X connect 4 0 5 0; #X connect 7 0 0 0; #X connect 8 0 1 0; #X connect 14 0 5 0;