Matthew Simon Fletcher Posted September 2 Author Share Posted September 2 So another observation. I fixed the waveforms by regenerating the images and then saved the project. I opened it back up again and the waveform is back to being glitched! Link to comment Share on other sites More sharing options...
bvideo Posted September 2 Share Posted September 2 It's possible to zoom all the way in to the sample resolution. See anything then? Link to comment Share on other sites More sharing options...
sjoens Posted September 2 Share Posted September 2 I figured they were audio artifacts or remnants from an edited wave that's too low to hear or for the meters to pick up, but something is or was there as even one of your screen shots shows. I also thought they were sometimes caused by my audio interface. What I've done is splice the offending spot out and bounce the surrounding clips back together. When it persists, I chalk it up to OCD and move on. You could try exporting the clip into another DAW to see if it reports the same activity. Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 2 Author Share Posted September 2 (edited) 52 minutes ago, sjoens said: I figured they were audio artifacts or remnants from an edited wave that's too low to hear or for the meters to pick up, but something is or was there as even one of your screen shots shows. I also thought they were sometimes caused by my audio interface. What I've done is splice the offending spot out and bounce the surrounding clips back together. When it persists, I chalk it up to OCD and move on. You could try exporting the clip into another DAW to see if it reports the same activity. But if that was the case then why did re-generated the picture for the audio cause it disappear? And why would the glitch disappear when zooming in? If anything that would make it more significant in size. I'm adamant there's nothing there, e.g. a vocalist has recorded a verse one and then the artefact is visible in a chorus three for example with absolutely no trace of audio in the middle. Edited September 2 by Matthew Simon Fletcher Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 2 Author Share Posted September 2 1 hour ago, bvideo said: It's possible to zoom all the way in to the sample resolution. See anything then? I've zoomed and zoomed - I had the waveform as large as the screen. There is nothing visible, but if you zoom out past a certain point the glitch will be visible., Link to comment Share on other sites More sharing options...
Noel Borthwick Posted September 2 Share Posted September 2 What your see when zoomed in is the raw audio not the preview audio. Preview audio is computed from averaged out windows and stored in the Wov files. If you don't see it when zoomed in then it's not there and I wouldn't worry about it. Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 2 Author Share Posted September 2 (edited) Hi Noel, How zoomed in do you need to go before it goes from the preview audio to the raw audio? Because when I opened the project file at different times the level of zoom in which the spike was visible was different. Surely that should be consistent? Also why is it that regenerating the picture caused the spike to disappear but then after saving/reloading the project it became visible again? Lastly, even if the preview audio is averaged out, how would a spike suddenly materialise in a section in which there is no other audio? In the example above I could understand it because there is some content near it, but in other examples it can be minutes apart. So I can't see how the preview would could be calculating/drawing it as such? I get this may seem like a minor issue but when you're dealing with a lot of tracks at a high level, it's really key to be able to know where audio is/isn't, so you can quickly trim sections that aren't needed but without moving other content. Currently due to this it's taken me longer to tidy-up / check a project and conversely I've also accidentally deleted things, thinking they were simply a UI glitch due to this bug. Edited September 3 by Matthew Simon Fletcher Link to comment Share on other sites More sharing options...
Bristol_Jonesey Posted September 3 Share Posted September 3 Another thing to try - delete the contents of your picture cache. Let the program re-compute the waveforms next time the project is loaded. Obviously only delete the cache when Sonar is not running. 2 1 Link to comment Share on other sites More sharing options...
John T Posted September 3 Share Posted September 3 17 hours ago, Matthew Simon Fletcher said: I get this may seem like a minor issue but when you're dealing with a lot of tracks at a high level, it's really key to be able to know where audio is/isn't, so you can quickly trim sections that aren't needed but without moving other content. Currently due to this it's taken me longer to tidy-up / check a project and conversely I've also accidentally deleted things, thinking they were simply a UI glitch due to this bug. I agree with this. Also, one of my use cases is audiobook and podcast editing, and it's a tangible slow down having to check for phantom marks like this. 1 Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 3 Author Share Posted September 3 (edited) 5 hours ago, Bristol_Jonesey said: Another thing to try - delete the contents of your picture cache. Let the program re-compute the waveforms next time the project is loaded. Obviously only delete the cache when Sonar is not running. Thanks for your thoughts. I've tried various combinations of recreating the pictures in Cakewalk and deleting these to force re-creation. I am always able to see a spike. Exported the track discussed so far to audacity and you'll see there's nothing at 3:09 which is the timestamp. Nothing visible in Next either. Not sure how the generation process is different on Next, but given it displays fine can we back-port it? It seems like there's enough evidence from different people's projects that something odd is happening here. Edited September 3 by Matthew Simon Fletcher Link to comment Share on other sites More sharing options...
Noel Borthwick Posted September 4 Share Posted September 4 Can you send the source wave file that exhibits this? Perhaps it may shed some light on why it happens, assuming its reproducible. 1 Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 4 Author Share Posted September 4 (edited) Sure - will message you shortly! Here's another interesting observation. I reimported the part in question into the project and at first it didn't show the glitch, however after zooming in and out it now appears consistently. At one level of zoom a random spike also appears at the end of the track, but if you go in/out at all that one goes. Again no content anywhere near that so can't see how it could be approximating anything, and again why it's only at one level of zoom that's neither too far in / out is bizarre. Edited September 4 by Matthew Simon Fletcher Link to comment Share on other sites More sharing options...
SteveC Posted September 4 Share Posted September 4 4 hours ago, Matthew Simon Fletcher said: At one level of zoom a random spike also appears at the end of the track, but if you go in/out at all that one goes. Now that one I've seen a few times myself, particularly after slip-editing or sometimes right after recording. There's usually nothing there as you've mentioned, and zooming in removes the visual spike. 1 Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 4 Author Share Posted September 4 (edited) Hi Steve - thanks for confirming you've seen this behaviour Noel and I are discussing further. Edited September 5 by Matthew Simon Fletcher 1 Link to comment Share on other sites More sharing options...
sjoens Posted September 7 Share Posted September 7 Is this unique to Sonar or does it also occur with the same wave file in CbB or older SONAR? Link to comment Share on other sites More sharing options...
John T Posted September 7 Share Posted September 7 It's been in for years. My screen captures in this thread are from CbB. 1 1 Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 8 Author Share Posted September 8 Okay so it appears that swapping from ASIO to WDM/KS is a workaround for this. Can people confirm which driver they are using and whether also applies in their cases? Link to comment Share on other sites More sharing options...
lɐʍd Posted September 8 Share Posted September 8 59 minutes ago, Matthew Simon Fletcher said: Okay so it appears that swapping from ASIO to WDM/KS is a workaround for this. Can people confirm which driver they are using and whether also applies in their cases? i'm interested to know how a driver mode can affect an image being generated Link to comment Share on other sites More sharing options...
Matthew Simon Fletcher Posted September 8 Author Share Posted September 8 (edited) Yes - i'm very surprised by this finding, as I wouldn't have thought it was connected. Hence not trying it until now! Edited September 8 by Matthew Simon Fletcher 1 Link to comment Share on other sites More sharing options...
David Baay Posted October 30 Share Posted October 30 On 9/4/2024 at 2:07 PM, Matthew Simon Fletcher said: Hi Steve - thanks for confirming you've seen this behaviour Noel and I are discussing further. Did this ever go anywhere? I've also seen "phantom" transients come and go in waveform drawing when zooming for years. It was never a huge deal for me, but it would be nice to have it eliminated as visual glitches of any kind detract from the overall user experience and confidence that WYSIWYG and vice versa. 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now