Splitting events with envlopes -> don't match up?

philfort wrote on 5/21/2003, 1:07 AM
I've got an event that has a velocity envelope on it, with a number of points set. I split that event, and now the ends of the event that were split no longer match up - not even close! Shouldn't Vegas realize that there was a velocity envelope, and adjust the start/end of the clips accordingly?
I'm using 4.0b. This worked in VV3 (with track envelopes - is there a way to use track velocity envelopes in 4?)

Comments

kameronj wrote on 5/21/2003, 6:09 AM
What do you meant that you have points set and now the ends don't match?

I want to try and recreate the issue. Sounds like an interesting one.
jetdv wrote on 5/21/2003, 8:38 AM
I was just able to duplicate this by adding a velocity envelope, adding several nodes just randomly set faster and slower, and then randomly picked a spot about about 2/3 of the way through and split. The frames to the left are right of the split point are no where close to the same location in the clip.

I don't know what you mean about VV3 with track level velocity envelopes. Velocity envelopes have always been event level.
philfort wrote on 5/21/2003, 9:38 PM
You're right, I don't know what I'm talking about... for some reason I thought velocity envelopes were track level in VV3.

And actually, this problem repros in VV3 too - which is weird, because I'm almost certain I've split events with velocity envelopes before, and had them line up perfectly. Guess I'm going crazy.
jetdv wrote on 5/21/2003, 9:54 PM
I suspect with a single speed change it probably would. With several speed changes, it seems to get confused.
T*Bob Tubb wrote on 2/18/2005, 9:53 AM
just searched for this today and couldn't find a whole lot on it, so thought I'd bring it back up.

yeah, I've had a problem with this for a while, as I do a lot of velocity editing. Snowboarding clips always have to be messed with as far as that goes. In experimenting, I think the 'split' function in vegas must create a new event which starts the DV clip at time t(cursor)-t(first_event_start) regardless of the velocity.

Maybe I'll create a script called velocity_split that calculates where in the DV clip the new event should start based on velocity points.

In a related velocity problem, I often have two events next to each other that are the same DV clip, just at different velocities. In the preview, I make this transition seemless, but then when I render it, I get a 'jump' or 'hitch' in the video, where the seam would be. I've often had to go back and make it a continuous clip with a velocity envelope, rather than two separate 'stretched' events. (when I say stretch, I mean holding down ltctrl and click+dragging the right of the event).