Bug 505676
| Summary: | no way to handle pass-through audio (line-in -> line-out) | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Stig Hackvan <stig-redhat-bugzilla> |
| Component: | gnome-media | Assignee: | Bastien Nocera <bnocera> |
| Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | low | ||
| Version: | 11 | CC: | bnocera, davidz, lkundrak, lpoetter, paradox606, wtogami |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2009-09-18 15:41:12 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Stig Hackvan
2009-06-12 21:12:41 UTC
I've been trying to listen to my ipod through my line in jack, played out of my laptop's speakers, all morning. And its been extremely frustrating, there is no way to do it with fedora 11! Once this bug is fixed I'll reinstall fedora, going back to an earlier release until this pulse audio bug is resolved. Stig, insulting developers and being rude certainly won't help us care about your use case. I plugged in my iPod, selected my computer's internal sound card as the default input, checked that sound was coming through it on the selected connector. Then I launched: arecord | aplay And it outputted the sound to my speakers. The sound was of poor quality and saturated, but I'm sure there's a better way to do that with PulseAudio. Then I'm sure somebody could write a simple app where you'd choose the input and the output and connect the 2 without recording (through the monitor API of PulseAudio, or using pulsesrc/pulsesink in GStreamer). (PS: we don't use votes) We never took away any features. You can still do that with a line like: amixer -c0 sset Line 100% on In fact, PA nowadays allows digital playback monitoring too, by loading "module-loopback". Doing this will allow you to have a PCI card monitor input from an USB card which the analog path does not allow you to do. So, let me stress this: we have not taken away any monitoring-related features, no, in fact we have extended them. However -- we don't expose it in the UI. Playback monitoring is not really something we want to show, because it is an exotic feature. I know that some disagree on that, but uh, we have to agree to disagree on that. (In reply to comment #2) > Stig, insulting developers and being rude certainly won't help us care about > your use case. > > I plugged in my iPod, selected my computer's internal sound card as the default > input, checked that sound was coming through it on the selected connector. > > Then I launched: > arecord | aplay Due to latency/time deviation issues, using module-loopback inside of PA is probably a better idea. Indeed, works much better using: pacmd load-module module-loopback sink=470 source=732 The magic numbers are the index of the source and sinks. You can get them using "pacmd list-sources" and "pacmd list-sinks" respectively. Somebody could write a small app that does that using drop-downs for each source/sink and connecting them, and showing a level meter. (In reply to comment #5) > Indeed, works much better using: > pacmd load-module module-loopback sink=470 source=732 > > The magic numbers are the index of the source and sinks. You can get them using > "pacmd list-sources" and "pacmd list-sinks" respectively. Somebody could write > a small app that does that using drop-downs for each source/sink and connecting > them, and showing a level meter. Instead of tracking down the magic numbers you could just load the module without specifiying anything and the move the streams to the right sink/source via pavucontrol. (In reply to comment #6) > (In reply to comment #5) > > Indeed, works much better using: > > pacmd load-module module-loopback sink=470 source=732 > > > > The magic numbers are the index of the source and sinks. You can get them using > > "pacmd list-sources" and "pacmd list-sinks" respectively. Somebody could write > > a small app that does that using drop-downs for each source/sink and connecting > > them, and showing a level meter. > > Instead of tracking down the magic numbers you could just load the module > without specifiying anything and the move the streams to the right sink/source > via pavucontrol. Except that you'd get feedback. Nearly made my ears bleed when the 5.1 speakers were feeding back into the webcam's mike... Bastien, I wasn't insulting developers, I was quoting a flame (url included) between Lennart & some other fellow what was making a lot of noise about the removed functionality during the F11 pre-release cycle. Now F12 is approaching and Fedora is still calling this removed functionality is "not a bug" and "wontfix" (oh, excuse me...really, it's buried-so-deep-that-only-the-hardcore-can-find-it)... Still I differ. I filed this bug because I could see the trail of flames on the net (quoted) but not a bug report. The loss of functionality was such an irritation to me that I stopped tracking fedora because of the new inconvenience. I'm not the only one...it seems that paradox606.com shares my sense of "acute disappointment" over this abandoned support in the UI (heck, the Windows UI let's me patch line-in through to the output channel...no command-line incantations required there...) for basic audio functionality. THIS CHANGE IS A *REGRESSION*! It is NOT "progress"... And what's sorely disappointing about attempting to participate in the open source process is that so often such efforts are rebuffed with a wholesale disregard for the user experience. Bugzilla conversations occasionally bring me workarounds and coping strategies, but I very seldomly see fixes come out of them. It is truly depressing. ps: my .bashrc is huge and I have no qualms with the command line, but in order to patch line-in to line-out, there's no way you can convince me that digging around in the command line, loading kernel/pulseaudio modules with carefully-researched arguments is the "right way" to do it. This IS A BUG! (In reply to comment #8) > Bastien, I wasn't insulting developers, I was quoting a flame (url included) How is that any better? You're quoting insults, does that make it alright? > between Lennart & some other fellow what was making a lot of noise about the > removed functionality during the F11 pre-release cycle. Now F12 is approaching > and Fedora is still calling this removed functionality is "not a bug" and > "wontfix" (oh, excuse me...really, it's > buried-so-deep-that-only-the-hardcore-can-find-it)... It's not hidden any further than it used to be. You can still use gst-mixer to enable the feature. > Still I differ. I filed this bug because I could see the trail of flames on > the net (quoted) but not a bug report. > > The loss of functionality was such an irritation to me that I stopped tracking > fedora because of the new inconvenience. I'm not the only one...it seems that > paradox606.com shares my sense of "acute disappointment" over this > abandoned support in the UI (heck, the Windows UI let's me patch line-in > through to the output channel...no command-line incantations required there...) > for basic audio functionality. The support is still there in gst-mixer if you require it. > THIS CHANGE IS A *REGRESSION*! It is NOT "progress"... And what's sorely > disappointing about attempting to participate in the open source process is > that so often such efforts are rebuffed with a wholesale disregard for the user > experience. Bugzilla conversations occasionally bring me workarounds and > coping strategies, but I very seldomly see fixes come out of them. It's not a regression, the support was in an application that's still available. The feature is still available, and you can even make it sound better than it used to by using PulseAudio's feature. > It is truly depressing. It's truly depressing that nobody started working on a UI for it when I've explained what's needed, instead of wrongly complaining that it's a regression. > ps: my .bashrc is huge and I have no qualms with the command line, but in order > to patch line-in to line-out, there's no way you can convince me that digging > around in the command line, loading kernel/pulseaudio modules with > carefully-researched arguments is the "right way" to do it. I said "the right way", because it doesn't eat as much CPU, or sound worse than "arecord | aplay" would. And you don't need to load any kernel modules. > This IS A BUG! It's not. It's a feature request. Given the explanations above, I'll reiterate that you won't see this feature in gnome-volume-control, but that you could implement a UI for it reusing parts of gnome-volume-control's code. |