View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0010035 | ardour | bugs | public | 2025-11-13 21:23 | 2025-11-13 21:36 |
| Reporter | jumper149 | Assigned To | |||
| Priority | immediate | Severity | major | Reproducibility | always |
| Status | new | Resolution | open | ||
| Platform | x86_64 | OS | NixOS nixos-unstable | OS Version | (2025-11-01) |
| Product Version | 8.12 | ||||
| Summary | 0010035: Clicks in stem export (without processing) and also playback | ||||
| Description | The playback issue is not reproducible and random, so I'll just focus on the stem export. This issue also occured when Consolidating, but here I was also not able to reproduce it consistently. I have a medium sized session with a few plugins and about 100 tracks (I still expect this to work, because I use the export without processing). I use `Session -> Export -> Stem Export` on a single track. "Apply track/bus processing" is not checked. I have tried out exporting to different output files (namely BWAV 24-bit and BWAV 32-bit-float). Everyone of these exports reproducibly generates the same artifact in the waveform. The artifact looks like a snap to 0db for a single frame. What's really annoying about this bug, is that it doesn't happen very often. Only about once every minute. So you have to be quite thorough to check for it. | ||||
| Additional Information | I have already tried out a bunch of stuff that didn't have an effect: * Using ALSA backend * Using pipewire-jack backend * Install linux-rt kernel 6.6 * Install linux 6.12.57 * Switching interfaces (Motu M4 and Focusrite Clarett 8+ Pre) * Switching USB cables * Switching USB ports (and bus) * Switching between USB-A and USB-C ports * Setting rtprio 99 * Setting memlock unlimited Notably in the session where I have the problem I do sometimes get X-runs even with huge buffer sizes (4096). And in that session I also get this message in dmesg: ``` [ 6829.049366] retire_capture_urb: 56 callbacks suppressed ``` In the attached picture you can see the source track and the imported stem-export side-by-side. One track has the spike the other one not. When running playback on the track without the spike I also cannot hear anything there. The actual spike in the stem-export is very much audible. | ||||
| Tags | No tags attached. | ||||
