Waveform poor performance and artifacts

Bug #998465 reported by Ferran Pujol
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Mixxx
Fix Released
High
Unassigned

Bug Description

I'm using:
-Intel(R) Core(TM)2 Duo CPU E7200 @ 2.53GHz (64 bits)
-2 GiB RAM
-Intel 82G33/G31 Express Integrated Graphics Controller at 1280x1024 60hz
-Maudio Xponent soundcard and controller
-UbuntuStudio 11.10 64 bits ~ (Ubuntu +Xfce)
-Mixxx 1.11.0-alpha2 x64
-Waveform: Filtered (GL) set 33fps

The waveform presents some artifacts like lines and "shadows" in both decks and works at slow fps.
It just happens with these skins, the other ones work well at 33 fps or more:
-Deere 1920x1080
-Deere 1920x1200
-LateNight 1280x1024
-LateNight 1280x800
-LateNight 1366x768
-LateNightBlues 1280x1024
-LateNightBlues 1280x800
-LateNightBlues 1366x768
-Outline 1024x600
-Outline 1024x768
-Outline 800x480 (no artifacts and a nearly good fps: it's very chaotic and has peaks down to ~25 fps)
-Phoney 1600x1200
-Phoney 1680x1050
-PhoneyDark 1600x1200
-PhoneyDark 1680x1050

Fps values are between 10~30. Some skins perform better than others at fps. Some skins have artifacts on both decks and others just in one. Artifacts are like lines or polygons that origin on the left side of the waveform display. Some skins have longer artifacts than others. I would give a more precise feedback of each skin if necessary. Not smaller skins always perform better than bigger ones.

Maybe an average fps display would help with a more precise feedback with future alpha or beta versions.

Revision history for this message
Thomas Vincent (vrince) wrote :

Faster 1.10-like waveform is comming.

Changed in mixxx:
milestone: none → 1.11.0
status: New → In Progress
importance: Undecided → High
Revision history for this message
Thomas Vincent (vrince) wrote :

Can you tell us more about the artifacts ?
Can you take a screen shot of it ?

Revision history for this message
Ferran Pujol (ferranpujol) wrote :
Download full text (103.4 KiB)

I ran mixxx from the command line. Here I copy the hole output. First I started Mixxx with a working skin set in preferences. The second output is from starting Mixxx with a non working skin.

Deere 1280x1024 SXGA (Working):

ferran@ferran-estudi:~$ mixxx
Debug [Main]: Mixxx 1.11.0-alpha2 "(built on: May 1 2012 @ 22:11:59; flags: hid hifieq mad optimize=9 qdebug shoutcast vamp verbose vinylcontrol)" is starting...
Debug [Main]: Qt version is: 4.8.1
Debug [Main]: Configuration file is at the current version 1.11.0-alpha2
Debug [Main]: Loading translations for locale "ca_ES" from translations folder "/usr/share/mixxx/translations/" : success
Warning [Main]: ControlObject::getControl returning NULL for ( "[Channel1]" , "vinylcontrol_mode" )
Warning [Main]: ControlObject::getControl returning NULL for ( "[Channel2]" , "vinylcontrol_mode" )
Debug [Main]: JACK client name set
ALSA lib pcm.c:2217:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.rear
ALSA lib pcm.c:2217:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.center_lfe
ALSA lib pcm.c:2217:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.side
ALSA lib pulse.c:243:(pulse_connect) PulseAudio: Unable to connect: S'ha refusat la connexió

ALSA lib pulse.c:243:(pulse_connect) PulseAudio: Unable to connect: S'ha refusat la connexió

ALSA lib pcm_dmix.c:957:(snd_pcm_dmix_open) The dmix plugin supports only playback stream
Cannot connect to server socket err = El fitxer o directori no existeix
Cannot connect to server socket
jack server is not running or cannot be started
Debug [Main]: Available QtSQL drivers: ("QSQLITE", "QMYSQL3", "QMYSQL")
Debug [Main]: src/library/trackcollection.cpp DB status: true
Debug [Main]: SchemaManager::upgradeToSchemaVersion already at version 17
Debug [Main]: TrackDAO::initialize QThread(0x24290b0, name = "Main") "qt_sql_default_connection"
Debug [Main]: CrateDAO::initialize()
Debug [Main]: CueDAO::initialize QThread(0x24290b0, name = "Main") "qt_sql_default_connection"
Debug [Main]: Promo dir: "/usr/share/mixxx//promo/1.8.0/index.html"
Debug [Main]: Traktor Library Location=[ "/home/ferran/collection.nml" ]
Debug [Main]: AnalyserWaveform::AnalyserWaveform()
Debug [Main]: Setting VAMP_PATH to: "/home/ferran/.mixxx/plugins/vamp/:/usr/bin/lin32_build/vamp-plugins:/usr/bin/lin64_build/vamp-plugins"
Warning [Main]: ControlObject::getControl returning NULL for ( "[Flanger]" , "lfoDepth" )
Warning [Main]: ControlObject::getControl returning NULL for ( "[Flanger]" , "lfoDelay" )
Warning [Main]: ControlObject::getControl returning NULL for ( "[Flanger]" , "lfoPeriod" )
Debug [Main]: Constructed LibraryScanner
Debug [Main]: iTunes Album Art path is: ""
Debug [Main]: Creating ControllerManager
Debug [Main]: Promo dir: "/usr/share/mixxx//promo/1.8.0/index.html"
Debug [Main]: loadSettings: 1 0 "SlowFade"
Debug [Main]: slotApply crossfader: 1 "SlowFade"
Debug [Main]: Setting VAMP_PATH to: "/home/ferran/.mixxx/plugins/vamp/:/usr/bin/lin32_build/vamp-plugins:/usr/bin/lin64_build/vamp-plugins:/home/ferran/.mixxx/plugins/vamp/:/usr/bin/lin32_build/vamp-plugins:/usr/bin/lin64_build/vamp-plugins"
Debug [Main]: DlgPrefBeats: qm-tempotracker Vamp plugin not found
Debu...

Revision history for this message
Ferran Pujol (ferranpujol) wrote :

Here I attach a compilation of screenshots made to 1280x1024. The artifacts change very quickly and are always visible.
Now I use Lubuntu 12.04 x64 fully updated, no improvements at all.

Revision history for this message
Thomas Vincent (vrince) wrote :

Thanks for the screen shot, I have had similar artifact long time ago, I'll re-re-check bound check.
When you say 'always visible' does that mean that i move along with waveform ?

I am suspecting too long rendering corrupting data.
Hopefully simpler and less demanding signal waveform are on their way.

Revision history for this message
Ferran Pujol (ferranpujol) wrote :

I mean that the artifact shape varies a lot (like blinking), but there is always visible artifacts. Artifacts always seem to origin from the middle point of the left border of the waveform area.

Revision history for this message
Thomas Vincent (vrince) wrote :

Ok so I know were to look. Thanks!

Revision history for this message
Daniel Schürmann (daschuer) wrote :

I have not seen those artifacts for long.
Is it probably already fixed?

Revision history for this message
Ferran Pujol (ferranpujol) wrote :

It is fixed on Mixxx 1.11.0-beta 2 running on the machine described earlier.
Thank you :)

Revision history for this message
Daniel Schürmann (daschuer) wrote :

Ok, then we can mark it as "fix committed".
Credits to the one who fixed it, not me.

Changed in mixxx:
status: In Progress → Fix Committed
RJ Skerry-Ryan (rryan)
Changed in mixxx:
status: Fix Committed → Fix Released
Revision history for this message
Swiftb0y (swiftb0y) wrote :

Mixxx now uses GitHub for bug tracking. This bug has been migrated to:
https://github.com/mixxxdj/mixxx/issues/6432

lock status: Metadata changes locked and limited to project staff
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.