Exported video has the same source timecode as the source clips, unless a clip is retimed or slowed down.
Hi daniel, does timecode need to be enabled. I’ve just done a test and am not getting timecode in my output video.
Hello, apologies I’m new to the forum, but I was posting this as a feature request! Currently VEAI doesn’t support these timecode features, and I’m writing them here to ask for it to be supported.
Timecode is not being preserved. This is true of many different sources we’ve tested, camera originals, transcodes, etc…
I’ve tested this with and without the “use timecodes” option turned on/off.
Timecode is being written out at 00:00:00:00 for frame 0.
This presents some significant issues with workflow.
I am aware there is an ability to re-stripe timecode using FFMpeg after the fact but that is an unfortunate step to have to take.
VEAI V3.0.0.2 beta now has the ability to export the original source timecode.
Exporting the original source timecode works with the 2X frame doubling AI models also.
Am I missing something? I’m not seeing this option anywhere in the beta (I’m on 3.0.0.4 now)
I can’t believe I am not seeing timecode in the app! Also it still does not preserve it on output. This needs to be solved to make it viable for pro users.
2 Likes
Is timecode throughput working yet? Would be really helpful in many workflows.
2 Likes
Just tested to see if original timecode is preserved in export. Seems to be not working. Clip starts at zero. Would be really useful.
2 Likes
Perhaps it is difficult to implement or expensive to license the feature of reading a video file’s timecode. As an alternative, maybe Topaz Video AI could allow the user to set the starting timecode in the viewer to make it easier to synchronize clip import and exports between programs.
2 Likes
As we are using mark in and mark out often to process multiple clips from a longer master clip (sometimes master clips can be an hour or more), source timecode support should be added. Also, ability to import EDL, XML or AAF would be fantastic for batch processing.
4 Likes
It would be nice to see some acknowledgement on this issue. Maybe Topaz doesn’t want to pay to license the feature or something. Seems like a basic feature to have, displaying the timecode embedded in the clip.
1 Like
I would realy love to have this feature, too. I really sucks to reconvert the clips in Compressor or Media Encoder and look up the timecode manually and make sure, you don’t have a typo…
I don’t think, this timecode feature is rocket sciences…
Absolutely! A professional workflow requires total transparency. I often render only the necessary parts of the video to save time and space. A source timecode is necessary to find the correct range. Please, make the clip source timecode visible. What’s worse, Video AI rewrites the timecode from 0 when rendering.
1 Like
We need to import footage with its original timecode intact, and the option to export it with the same timecode, or not. What do we have to do to get implementation of this capability escalated? Can we get a response from Topaz Labs as to whether this basic and indispensable feature is happening? Thank you.
1 Like
I second that! We do need the original timecode in the output render even if we just take parts of the clip. Otherwise we have to bring in all the footage by eye match, that’s stupid and time-consuming extra-work .
Timecode kept in preview, but not in „real“ export
I just discovered, that Video AI maintains the source timecode in the generated preview file but not in the „real“ export. But in the preview file, the sound is not exported. So we can either have time or sound.
I’m not a programmer, but as far as I can tell, this change shouldn’t be that difficult to implement.
Having timecode preserved is essential for a professional workflow. . It’s truly amazing to me that Topaz doesn’t seem to care about it and leave us to frame match by eye.
I think I found the „secret“: You have to stay in the same codec and the same frame rate and copy the audio.
I converted some NTSC videos to PAL in Compressor, which is much faster (and feels even better to me). I did not change the picture sizes (only 720x480 → 720x576 NTSC/PAL size switch), I already added the 8 audio channels I would need later. I always switch to progressive. The best codec for that is, in my view, ProRes LT as MNX not MOV. I keep the original time code in this conversion.
In Video AI, I tried different models but stayed with ProRes LT and Audio Mode Copy (hoping to keep the 8 channels).
After I finished, I was about to reconvert the Video AI file in Compressor to get the time code back. I was surprised. I was still there.
So I guess if I don’t change the codec and the frame rate, the time code will stay.
