Page 2 of 2 FirstFirst 12
Results 11 to 14 of 14

Thread: I am seeing some anomalies in Vixen 3.9

  1. #11
    Join Date
    Oct 2011
    Location
    Maryville, Illinois
    Posts
    2,082
    Post Thanks / Like

    Default Re: I am seeing some anomalies in Vixen 3.9

    Quote Originally Posted by RichieNorthcott View Post
    Example of timing issue.... I took this from my 3.8 'Sleigh Ride' sequence - the mark was aligned tightly on the 'whip' sound at the end of the track.
    Attachment 46128
    Does the audio align with the waveform or is it more aligned with the old timing mark? I need to figure out if it is an offset issue compared to how the old library decode the stream or if there is some latency in the playback. Since it appears the waveform is ahead, and if the audio tracks that, then latency may not be the issue.

  2. #12
    Join Date
    Jul 2019
    Location
    United Kingdom
    Posts
    141
    Post Thanks / Like

    Default Re: I am seeing some anomalies in Vixen 3.9

    Thanks Jeff. Summary of my current thoughts based on some further testing:

    - I think the audio spike is just a playback issue in 3.9. It relates to the audio in a sequence having already been played in part. For a show this may not be an issue since it doesn't happen on first run. I don't believe it's related to the timeline content shift issue, so it's more just a minor annoyance. I haven't tested what happens if there are several sequences in a show playlist. It may happen if the end of the previously played track is not 'silent'

    - If anything, I think the on-screen waveform vs 'what you hear' sync is actually much better in 3.9 vs earlier versions. I played back some sequences at super slow speed in both versions to confirm this. To answer your question - building a new sequence in 3.9 the waveform, marks & effects, audio sync seems spot on.

    - the sync drift may have crept it before version 3.9 since I see similar in my old sequences on previous verions. They were built on earlier versions - i.e. I see the same in 3.8, the sequence was built in 3.6u4 or possibly earlier. Perhaps we haven't picked up on this until now.

    Is there a way of shifting the whole sequence (all marks and effects) by a user-entered offset? This function would also need to shorten any items at beginning or end of sequence if there's overlap. An alternative of course would be to offset the audio file in Vixen (not currently possible) or edit in another application.

    Thanks again
    Last edited by RichieNorthcott; 09-24-2022 at 05:50 AM.

  3. #13
    Join Date
    Oct 2011
    Location
    Maryville, Illinois
    Posts
    2,082
    Post Thanks / Like

    Default Re: I am seeing some anomalies in Vixen 3.9

    I have noticed the pop here and there at the start and it is on my list of items to chase down. It was pretty minor on my system, but I can see where it may annoy others more.

    The waveform is more detailed in the new version and in my testing I believed it was a better representation so good to hear you agree. In case you had not noticed, you can choose a full or half waveform display. Under View -> Full Waveform toggle.

    I looked pretty closely at my pool of sequences here which span several years, however I tweak on the ones I use each year. In my testing with 3.9 I have found that most of them the audio leads the older version by about 15 - 20 ms. I did not really notice that when playing at full speed since the timing of the sampling by default is 50ms and can have a little variation in it anyway. But, I got concerned when you were seeing a much larger variation. I went back and reran my tests on my sequences and still see the same 15-20 ms lead. I would be interested in seeing your profile to give that a deeper look. My test case was to put marks in obvious spots with 3.8 and then compare where they landed in 3.9 compared to the waveform. This assumes the waveforms follow the audio in both versions.

    It would be interesting to further validate your thoughts on earlier versions, but I can't put my finger on any changes that might have caused that much drift. I feel like I would have noticed that since I am very conscious about tight sequencing. Worth further thoughts.

    There is already a way to bulk shift effects. Under Tools-> Bulk Effect Move. You can shift forward and backward all the effects in the sequence in one shot. I would have to look and see how it deals with those that may start right at the beginning or right at the end when they are shifted beyond the start or end.

    Marks are a bit more complicated, but can be done. If you enable the mark bar for each one, you can then select them all. Select the top first one at the beginning of the sequence and then the bottom right one at the end of the sequence holding shift. Verify they are all selected and then drag the whole lot of them left or right. If there are a lot of them it may be a little twitchy to work with, but it is at least an option to try. I have it on my list to add logic for a bulk shift like that based on time.

  4. #14
    Join Date
    Jul 2019
    Location
    United Kingdom
    Posts
    141
    Post Thanks / Like

    Default Re: I am seeing some anomalies in Vixen 3.9

    Thanks -

    The audio pop definitely relates to what was previously playing. If your track ends in silence, you don't get in on resuming playback.

    I just bumped into a function to move all marks in a group... right-click the group name in the marks window then you can offset by a time, positive or negative. You would need to move all mark groups by the same amount of course. So if the same offset was used in the bulk move utility as the the mark groups, then effectively everything gets shifted by the same amount. It may be worth considering (here comes a feature request.....) adding a checkbox to 'include all mark groups' in the bulk shift function.

    I've been testing just now with some new sequences in 3.9, connected to DMX fixtures. And - using a finger snap audio file to check with, which gives a very clear peak in the waveform. Timing seems perfect IMO. Might be worth shifting some of my old sequences when moving to 3.9, as I say I think in general it actually represents the audio much more accurately.

    Nice - like the full waveform toggle (hadn't found that)
    Last edited by RichieNorthcott; 09-24-2022 at 10:03 AM.

Page 2 of 2 FirstFirst 12

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •