Peter, thank you for the clear explanation. It helps me with my editing strategy.

Since you are using avi-the multiple rendering will have a verry little effect.
Different story with MPEG.After few(3) rendering it does show some quality loss.Since I am using NTFS file- I am capable of capture large avi. I store everything on avi and after changes needed to be made - I render in mpeg for final.

I've read that you will loose more from the FX you add then with the rendering. Each FX changes the attributes of the colors, chroma, etc. You've gotta watch your levels close, but that's all.

Be careful de-interlacing, too. This is a destructive process that softens images. Isolate those titles, areas with jitter and render them de-interlaced separately...once.

If you know ahead of time that you are going to be doing multiple passes, you might consider rendering to Uncompressed the first time. Then do not render back to DV until your final pass. This may save some quality. The quality loss is introduced because each time you render an effect, the file is Uncompressed, Rendered and then recompressed. The uncompression/recompression is where the quality loss is introduced. By minimizing how many times this happens, you can save some quallity.

Dave T2

Thanks all for the advice. It sounds like 2 to 3 passes (NTSC DV) would be acceptable and would generally allow me to break a large project down to work on individual segments and then reassemble without much loss in quality. Uncompressed generates files about 10x the size of NTSC DV, so I might reserve it for really critical stuff.

Let me pick your brain for minute...

I'm just curious why you break the project down and render bits of it then bring it back in to render again.

I do some very long and complex projects and I NEVER pre render or render multiple times. Rendering for me is the absolute final last step and I only do it once. Period.

BB,

While it seems he is adding effects at multiple places, I don't do that but I DO create multple renders.

I will work on a small section of the final product, finish that section, and render it. I will then work on the next section and render it. This process is repeated for as many sections as I need.

Once all of the sections are completed, I then create a "final" project that combines all of the sections adding dissolves between the sections as needed. The one difference between my workflow and his seems to be that he then adds additional effects in the final step. I don't do this (although I guess it WOULD be a good place to add a broadcast clamp or something of that nature).

In my case, since only the transitions will be rendered (the rest is merely copied) I will get no loss in resolution with this process. However, there ARE reasons for editing in pieces.

DSE has spoken of doing long-form work like he's writing a book -- chapters at a time. Maybe he never re-renders, but then, maybe his needs are different than others...

Myself? I re-render a LOT. I've got a project right now that has about 12 layers of transparency and a lot of care taken to make sure the final product looks great -- though it will never play flawlessly in the Vegas preview window, so I have to render out segments, bring them back in, and re-render these segments with other parts ontop so as to keep more subtle transitions previewing accurately.

I'm sure someone else has equally valid reasons to re-render.

Which is what I was asking.... WHY render in pieces?

I'm not suggesting one way in better than another, just wondering WHY you're doing it.

Because it easier to keep track of all the pieces of a montage and build that montage if you don't have to worry about other things on the timeline. Build the montage, render it, add it to all the other pre-rendered montages. Makes for a simple procedure and a less confusing timeline.

There have also been times when I HAD to do it because of disk space. Render a segment, delete all of the source files.

I guess it’s because I don’t know a better way; maybe a carryover from my programming days. If the timeline gets too congested and unwieldy, I like to isolate the more complex segments and work on them individually without all the extra baggage on the timeline.

When a project starts getting very long and complex, it can be rather easy to forget to toggle the Ripple Editing mode when necessary and not see the effects for a long time. I might spend an hour zoomed in working on a small section only to discover later that i've destroyed the sync between tracks later in the project. At that point undo is a painful option because then i lose all the current work i've just completed. In a case like this, working on a smaller complex section independantly then bringing the rendered version into a larger project can save a lot of confusion and headaches.

Different strokes for different folks I guess. For me I don't care how long the timeline is, you're only working on a segement at a time anyway so I don't care what's 30 minutes down the track or 2 hours down the track... because that's not what I'm working on at the moment.

BB,
I do the same thing for long projects. First VV project was 12 hours of my travels in China cut down to fit a DVD. I did each leg of the journey as a separate project, rendered them out and then bought them into a final project, joined together with a few seconds of black between. The black corresponded to the chapter points in the DVD.

Big advantage is not having to worry about screwing up somthing an hour down the timline while working on something at the beginning. There is a nasty problem with VV that was driving me nuts and that's why I always work this way. Remember if you split an event VV unlocks the A/V on the RH side of the split. Makes it all too easy to get things out of sync etc.

Having started out with Premiere I'd say that was the only thing I liked about it, it was much harder to get things out of sync and even when you did you could always get it to resync them. I'd say again that was the ONLY thing I liked about it though.