View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0002027 | ardour | bugs | public | 2008-01-15 17:36 | 2011-08-16 11:09 |
| Reporter | colinf | Assigned To | paul | ||
| Priority | normal | Severity | major | Reproducibility | always |
| Status | closed | Resolution | fixed | ||
| Product Version | SVN/2.0-ongoing | ||||
| Summary | 0002027: Empty & silent tracks produce "The disk system on your computer was not able to keep up with Ardour" | ||||
| Description | Empty tracks, or tracks containing only muted regions, have started causing Ardour to produce the dreaded: "The disk system on your computer was not able to keep up with Ardour. Specifically, it failed to read data from disk quickly enough to keep up with playback." error, as soon as you hit play. I think this was introduded by SVN commit 2834, that was supposed to not read inactive tracks from disk. It now seems to try to read from disk for the empty track, but of course gets no samples, which is then interpreted as an underrun. | ||||
| Tags | No tags attached. | ||||
|
|
1) are you sure this is not caused by *inactive* tracks? no changes were made that relate to silent/empty tracks. 2) assuming that its caused by an inactive track, then .... fixed in my code, it will be in svn today. still needs more testing, so feel free to provide some feedback. one problem is that tracks need a locate to resync after they are made active again. |
|
|
OK, I'll explain in a bit more detail. I have a session with three snapshots: * the first, saved by whatever the latest 2.0-ongoing SVN version around the 20th of August 2007 was, plays back perfectly in recent SVN revisions. * the second, with only a few small changes from the first, saved from revision 2903 or thereabouts, at first wouldn't play back without stuttering, and then very soon began producing the "disk not able to keep up" as soon as I hit play. Looking at the values of name(), total, and necessary_samples just before the call to DiskUnderrun() in audio_diskstream.cc showed that there were two active tracks in the project, one muted and one containing no regions, for which total was 0, and necessary_samples 128. Deleting the empty track and unmuting the muted one enabled me to carry on working with the session. * the third snapshot is of the fixed session. I can post the session files here if it's any help. |
|
|
I'm sorry: Paul, you're right. I went back and checked, and it is inactive tracks causing the "cannot keep up" error. I must have de-activated them when playback first started stuttering with the session in question, and I didn't realise that was the trigger for the error appearing. I think the stuttering is related, though: I'll keep digging... |
|
|
Revision 2935 fixes the inactive tracks 'disk cannot keep up' error: it also seems to cure the stuttering playback on the session I was having problems with, though that may just be coincidental. |
|
|
This issue can definitely be closed now. |
|
|
see notes. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2008-01-15 17:36 | colinf | New Issue | |
| 2008-01-16 20:12 | paul | Note Added: 0004658 | |
| 2008-01-17 09:36 | colinf | Note Added: 0004661 | |
| 2008-01-17 16:53 | colinf | Note Added: 0004664 | |
| 2008-01-18 11:56 | colinf | Note Added: 0004666 | |
| 2009-10-27 16:40 | colinf | Note Added: 0006905 | |
| 2009-10-27 17:44 | paul | cost | => 0.00 |
| 2009-10-27 17:44 | paul | Note Added: 0006906 | |
| 2009-10-27 17:44 | paul | Status | new => resolved |
| 2009-10-27 17:44 | paul | Resolution | open => fixed |
| 2009-10-27 17:44 | paul | Assigned To | => paul |
| 2011-08-16 11:09 | colinf | Status | resolved => closed |
