I should perhaps point out that what I describe happens when opening and saving WAV files on Windows. I don't have access to an Intel Mac at the moment, and cannot test the beta myself.
1 person found this helpful
These files typically have markers identifying takes, points of interest etc. When saving them, Audition seems to convert Markers into Regions of zero length, which will then confuse the parent program (Sony Vegas in this case). Can anyone verify if this is still the case?
Thanks for the post and your patience on the reply. Digging through my RIFF specs is something I enjoy doing, but always requires a fresh pot of coffee. I think I see what you're saying (that is, I can notice a difference which looks like what you're describing). I tested with a WAVE file from Sony Sound Forge (I figured it was "close enough" to Sony Vegas). To "get the full picture" of Cues, Markers, or Regions in a WAVE file, you have to look at several Chunks within the WAVE file...
I created a file in Sony Sound Forge, it has 4 CuePoints found in the Cue chunk.
CuePoint ID: 1 ID: 2 ID: 3 ID: 4 Position 44544 135424 176384 42998 DataChunkID data data data data BlockStart 0 0 0 0 SampleOffset 0 0 0 0
It has 4 Labels found in the Labels Chunk.
Label CuePointID: 1 CuePointID: 2 CuePointID: 3 CuePointID: 4 Text Region 00:00:01.010 to 00:00:02.008 (Region 01 / Region 02) Marker 03 Region 04 3.999-4.580 CMD TEXT: This is a text command
It has 2 LabeledTextItems which correspond to the two Cues that are actually Ranges
LabeledTextItem CuePointID: 1 CuePointID 3 SampleLegnth 44032 25600 PurposeID 544106354 544106354 Country 0 0 Language 0 0 Dialect 0 0 CodePage 0 0 Text
These show up in the Public Beta as:
Now, if I resave this through the Public Beta Build, and analyze these chunks, the new file has:
CuePoint ID: 1 ID: 2 ID: 3 ID: 4 Position 42998 44544 135424 176384 DataChunkID data data data data BlockStart 0 0 0 0 SampleOffset 0 0 0 0 Label CuePointID: 1 CuePointID: 2 CuePointID: 3 CuePointID: 4 Text CMD TEXT: This is a text Command Region 00:00:01.010 to 00:00:02.008 (Region 01 / Region 02) Marker 03 Region 04 3.999-4.580
But now has 4 LabeledTextItems!
LabeledTextItem CuePointID: 1 CuePointID: 2 CuePointID: 3 CuePointID: 4 SampleLength 0 44032 0 25600 PurposeID 544106354 544106354 544106354 544106354 Country 0 0 0 0 Language 0 0 0 0 Dialect 0 0 0 0 CodePage 0 0 0 0 Text
And still shows up in the Public Beta as:
(yes they look deceptively the same)
So, the big difference is that we now have a LabeledTextItem for each CuePoint and not just the ranged CuePoints. I would imagine that the reason why we would do this would be an optimization such that if all you did was change the offsets or durations of markers, we could quickly save without having to resize or rewrite chunks in the WAVE file. On the other hand, if this is causing interoperability problems with other applications, I'll call it a bug.
Also, is unnamed Markers automatically given the name Marker 01, Marker 02 and so on? This is another issue I have, since it prevents my scripts from recognizing the markers once they go back into Vegas.
No, this is no longer happening. If the CuePoint doesn't have a label (see tables above), then we won't automatically populate them with Marker nn when we open the file.
Otherwise, let me just take this opportunity to thank you for an amazing tool. It has truly revolutionized the way I work.
Opportunity granted. Thanks for the feedback.
Thanks for the very extensive reply!
Now I just have to go find a cryogenic chamber so I can leap forward to the release day of Audition 4 for Windows =)