If you render to uncompressed .avi (yes, takes lots of drive space) and then encode to DV/MPEG/WM from this "uncompressed intermediate", you get the best possible quality. This can save quite a bit of time if you have an fx intensive project, with lots of render-intensive blurs for instance- you only have to grind that stuff once.

Batch rendering (using the batchrendergui.js script as a starting point) is an alternative- just set it and forget it. Quality will be the same as the above, aggregate render time will be longer , quality identical, less drive space required, the beauty of this method is you can set it up and let it grind through all the export types without any intervention or hd space concerns.

You mentioned uncompressed avi. Is this the same as what Vegas offers when I chose 'Video for Windows (*.avi)' as the 'save as type' in the render as window input box?

Lots of Drive Space is required for the uncompressed. I had a two hour project to render and tried uncompressed. It used 46GB for 18 minutes of video and ran out of space on that drive.

Using the standard Avi set on best quality does the whole 2 hours in about 20GB if I remember correctly - may have been even lower.

Are these types of statistics similar to what others experience?

If I ever decided to use uncompressed avi, how do I instruct Vegas for this format?

Uncompressed AVI in a standard NTSC format will require about 1.7GB/minute or slightly over 100GB/hour. To select this, choose Video for Windows and use the default (uncompressed) template. The advantage this has over the DV template is that it uses 4:4:4 colorspace instead of 4:1:1 so the colors will be sharper and clearer.

If your source material is all DV then you won't gain much in quality by using uncompressed. You would probably be better off rendering to DV .avi which uses about 225MB/minute or 13GB/hour. Keep in mind that this will impose a 4:1:1 color compression which you might avoid if you rendered directly to MPEG instead. But, if your source is DV then it's already 4:1:1 anyway.

As far as speed is concerned, rendering to AVI first and then rendering that AVI to MPEG will always take longer than rendering directly to MPEG. Each of the two separate renders may go faster, but put together they will add up to longer than a single render direct to MPEG. Why do this then? There are situations in which you will want to render to several output formats including printing to tape. An initial render to DV AVI will produce a file that can be printed to tape, and also takes care of rendering all the effects, cross fades, color corrections, titles, etc. once and for all. Subsequent renders from this file to other formats will only have to perform the format conversion and won't have to wast time rendering all the other "stuff" over again each time. If you're only intending to render to one MPEG file then it's not worth going through AVI first.

Thanks everyone who has contributed to my question and helping me understand a little more about NLE.