| # | User | Message | Date |
| 8 | Cyphre | My quick guess the current audio code is too synchornous. I think it would need to be rewritten to use at lest doublebuffering so it can 'stream' sound data blocks without any hickups. I'll try to investigate more... | 5-Sep-08 15:47 |
| 7 | Cyphre | also the WAIT has no effect on the result. If you jus do N times INSERT on the soundport without WAITing you can hear the same delays between playing newly inserted soud data. | 5-Sep-08 15:42 |
| 6 | Cyphre | The code above is very simple test where you can easily hear the 'broken beat' in the sound loop everytime soundport is playing newly inserted sound data. The WAV test sound is precisely cut so if you play it in a loop using any ohter 'player' the beat is exact without any delays. | 5-Sep-08 15:39 |
| 5 | Cyphre | print "loading looped sample(around 600Kb)" wav: load-thru/binary http://www.rebol.cz/~cyphre/hardfloor-loop.wav sound-port: open sound:// while [wait sound-port][ print "inserting looped sound" insert sound-port wav ] close sound-port | 5-Sep-08 15:35 |
| 4 | Cyphre | Thanks for sharing the source Carl, I'll try to look into it over the weekend. It can take some time as I'm not much fammiliar with the waveout api. I was playing with directsound only so far. | 5-Sep-08 15:35 |
| 3 | Carl | BTW, is there some test code to show the sound problem, so we know for sure when it is fixed? | 4-Sep-08 21:30 |
| 2 | Carl | Regarding sound code... note that the code: waveOutSetVolume(hwaveout,(vol<<16)|vol); is no longer called every time. | 4-Sep-08 21:29 |
| 1 | Carl | For fix notes. | 4-Sep-08 21:28 |