Native HDV editing revisited.

Laurence wrote on 5/23/2008, 11:49 AM
Until recently, I didn't take native HDV mpeg2 long GOP editing that seriously. My m2t clips would crash on the timeline, and if I did manage to get a smart-render successfully completed, often it would crash in WMP or give me other problems.

Now between using tapeless acquisition with my new HVR-Z7 and realizing that if I just turn off my virus protection while I capture with HDVSplit from tape, I can work wonderfully well with native m2t clips and that smartrenders into m2t masters are in fact reliable.

Two huge advantages of working this way are that I use far less disc space and my renders are lightening fast. The only real disadvantage is that if I render to something other than Bluray compatible m2t, I have an extra render on parts that don't smart-render (color corrections, deshaken parts, transitions, etc.) Other than that it is pretty cool.

Who else is working this way? Are you doing it on simple event stuff only or are you doing it on fully authored projects as well?

Comments

winrockpost wrote on 5/23/2008, 3:57 PM
I've never used anything but the m2t since vegas 8 . Done several long projects (one had 25 hours of tape, all captured with vegas ),no problem, but.. I dont use smart render, havent tried it in a while but it didn't work and i very seldom have clips that we dont do a little something to so it doesnt matter
johnmeyer wrote on 5/23/2008, 4:39 PM
I have used native editing exclusively since 7.0d (which I am still running). For simple projects, I can see no advantage to creating the Cineform intermediates.

Having said that, if I ever get a modern computer with some serious horsepower, I will probably also spring for the full Cineform package. At that point, I'm pretty sure I will capture directly to Cineform AVI and edit with those. The few times I have bothered to create the intermediates, the editing was a LOT faster (better timeline performance). However, I don't do any of the really high-end 32-bit stuff so the "break down" of editing with m2t that Spot and others sometimes mention is something I have never actually experienced, so that additional reason for using Cineform is not a factor for me.


winrockpost wrote on 5/23/2008, 5:36 PM
people actually use that 32 bit mess,,, baffles the hell out a me :)
blink3times wrote on 5/23/2008, 6:55 PM
I USED to do the M2T/smartrender thing, but I find myself doing track color corrections more and more which kills the smart rendering, and I get a lot of crashes using scene detection of anykind with M2T files.
Laurence wrote on 5/23/2008, 8:16 PM
The funny thing is that smart-rendering long GOP is supposed to be a problem because you get a screwed up GOP sequence. In actual practice however I find it an advantage rather than a problem in that with a regular GOP sequence you get some blockiness at the edit points whereas there are extra fully defined frames in the smart-rendered mpeg2 that make the transitions look a heck of a lot better.

I love looking at a finished master where most of the footage has never actually been rerendered.
Laurence wrote on 5/23/2008, 8:26 PM
In addition to the lossless video quality there is the savings in time since smart renders are for the most part just file copies. I find myself being more careful color and motion-wise while I shoot in order to have footage that will smart-render without any kind of correction.

Another little trick I've been using lately is to take the smart-rendered master and render it into a downrezzed DivX version for uploads on sites like Youtube using Dr DivX. This has the advantage two advantages over doing an mp4 render in Vegas: One is that it's faster. The second is that it doesn't bog your computer down during the render so you can still use it during the render for other stuff. This probably wouldn't mean much to anyone using a quad core system, but on my Core2Duo it is much faster and less intrusive to my workflow.