Think of file formats as tools with some doing certain jobs better than others.

AVI encompasses a few types of encoding, most of which use lossy compression. If you encode to a lossy AVI format and then from that to MPEG2 then you're compounding the compression artifacts.

However, If you want to encode an AVI master and then do several compressed encodes based on that (say you want to make quicktime, wmv, and mpeg2 encodes) you could make an uncompressed AVI encode. This format is lossless but takes a ton of disc space.

The main reason to do this is if your project takes a while to render. You don't want to have to do it over and over and over just to render to these delivery formats. So you just do it once to a good looking format and then do the final renders from that file.

DV25 is an aquisition format. It's actually not such a good choice to render final product back to DV25 because it tears up titles and graphics.

Rob Mack

thank you

logan5

I could be wrong, but I thought that when you went from the timeline to MPEG that there was an intermediate AVI, at least GOP by GOP.

The ideal workflow, if your ultimate destination is MPEG-2 to put on a DVD, is to NOT render to any other format (unless it is uncompressed AVI).

However, that said, if your source files are DV, and you render to a DV AVI, Vegas' DV compression is so good that you are simply not going to be able to see any resulting loss. That has been confirmed by countless tests. Therefore, if this is all you are doing, and rendering first to a DV AVI somehow helps your workflow, then go ahead and do it.

But there is one more part to the story. Any graphic created by Vegas (like titles) and any still image you import, is going to have full color fidelity, unlike the limited color space of DV AVI. When you render these to DV AVI, color information is lost, and that loss is noticeable. By contrast, MPEG-2, despite the additional compression loss compared to DV AVI, actually renders colors more faithfully. Thus, if your project has still images or has lots of things created by Vegas itself (backgrounds, titles, etc.), then you should skip the creation of the intermediate DV AVI, if possible, and simply render directly to MPEG-2.

Yep. Going straight from the timeline to MPEG2 will give you good quality, just as good as you'd get by going to uncompressed AVI and then to mpeg2.

The point I was making about going to uncompressed as an intermediate is just that if you have to make multiple renders for delivery in several formats then making an uncompressed intermediate may save some render time overall. You have to use your best judgement, of course.

Vegas does a great job of rendering DV25 to DV25. It'll hold up to way more generations than you'd ever reasonably make. In fact, Vegas is better at this than many other NLEs and this is mainly because Vegas does a straight copy of the DV25 data whenever it can. Those parts of a render go by very fast too.

The problem, as John points out, is that added graphics and text don't hold up when encoded to DV25. The issue here is that graphics, titles, and photos all use 4:4:4 color sampling. This means that every single pixel has a sample value. When you encode these to a DV25 file Vegas must throw away three out of four color values because DV25 uses 4:1:1 color sampling. That is, for ever group of four pixels, the first one will have a brightness value and two color values. The other three will only have a brightness value and the color values are calculated based on the last and next known values.

The result here is that DV25 must calculate an average color in these three-pixel stretches. Uncompressed encoding doesn't do this, of course. Also, going straight from the timeline to whatever your final format is, whether it's mpeg2 or Window's media, or one of the many Quicktime formats, skips any sort of DV25 downsampling. So you get better quality than you would have by rendering to DV25 first.

Rob Mack

I've done it both ways. Making a master AVI file file to take over to DVD Workshop for the encode, and making the MPEG2 from the Vegas timeline to take over to DVDWS (I don't have Vegas DVDA). I liked the encode from DVDWS better than Vegas' MPEG encode, even though I believe they're both Main Concept (not sure).

As far as the resulting encode quality on the finished DVD, I honestly can't tell the difference.

Neither can I. I have sometimes eliminated flash frames by going through the intermediate step first.

John