Editing Native HDV

mjroddy wrote on 8/28/2006, 10:03 AM
I don't get this.
We're told that Vegas 7 will be able to edit natively in HDV and I can see that as a work-flow benefit, but....
I thought it was a bad idea to edit native HDV. I thought it fell apart much more quickly (than Cineform) and if you hit an "I" (or was it "P") frame, you were hosed when it comes to CC, etc.
Going through the Vegas training, they were pretty specific that "You Don't Want To Edit HDV," and, at the time, this made total sense to me (which is surprising).
What has changed?

Comments

johnmeyer wrote on 8/28/2006, 10:07 AM
What has changed?

Because Vegas 6 couldn't do it, but now (apparently) Vegas 7 can.

I think 99% of the arguments about not being frame accurate were completely, totally bogus. That is not to say that it wasn't true for Vegas 6, but it IS to say that it did not necessarily have to be true with a properly engineered MPEG editing tool. I mean, think about it, when you play the video, there is a frame of video on the screen every instant in time. Do you think that if you replay the video that there will be a different frame of video there the next time? Of course not. You can always get the same frame up there. The problem for editing is that Vegas has to keep all the frames in each group of pictures (GOP) in memory, and remember all the complex relationships between them, in order to be able to display any given frame, and in order to go back and forth on the timeline. This is a lot more difficult than just decompressing each frame, independent of the others. However, once you figure it out, each frame can be displayed.

Now, having said that, the multi-generation loss in rendering from m2t to another m2t is still going to be an issue (and always will be). Thus, if your workflow requires multi-generation rendering, you still want to use intermediates. Also, I would expect -- although I do not know this for a fact -- that the intermediates will still provide faster response, although perhaps not enough to make them worthwhile for simple editing.

It will be very interesting to see how the Cineform folks position their technology once Vegas 7 is introduced. I have been VERY impressed by their technical people and by their no-nonsense approach to marketing. I really trust what I hear from them and what I read on their site. Therefore, that is where I recommend people go to get answers to these questions.

Spot|DSE wrote on 8/28/2006, 11:25 AM
You *still* don't want to edit native HDV if you're doing any level of compositing, re-rendering, etc. Even though you *can* edit native HDV, I'm still working with 4:2:2 instead, simply because it's better (and faster if you have an optimal system) to work with.
As far as frame accurate, the argument never was bogus in early application versions; decoders have come a long way since just 2 years ago. Hardware is more capable now from what it was two years ago as well. It's still not bogus that it takes additional horsepower to decode anything that isn't an I frame, even if it's a frame-complete DCT frame.