Why do you need to split up the rendering time? The best way to do this is to render the entire thing in one shot with chapter markers.

I agree with Chien. Mark the chapters and render once. Doing it the other way is actually more time-intensive and not practical at all.

I was just thinking of an easier way to fix mistakes. With a long movie, I always get some small mistakes creeping in due to the time it takes to edit. I thought if a mistake was found, all I'd have to do would be to load the chapter with the mistake and fix it, rather than the whole movie then re-render that again.
Bob, when you say a lossless codec, what do mean please?

You want to have one large video, & 1 large audio file -- while you can do them separately in DVDA, your odds of glitches on playback from a set-top box multiply.

Joining mpg2 can be done in Womble, Cuttermaran, & I think ProjectX [I've mainly used ProjectX for repairs & mpg2 mods, so I'm not 100% sure]. The sort of thing you're talking about works best if you fade to black (or white) at the joins, so if you do it plan ahead & add the fades. If using Cuttermaran or ProjectX, both like std. mpg2 video only [m2v], yet Vegas only works with, & DVDA only *likes* mpg [the Sony version is muxed to add data to the stream, but no audio]. Be prepared and get a muxer/demuxer that will mux without audio -- I use TMPGEnc's mpeg tools.

The alternative of rendering to an all key-frame, near lossless, format requires a codec... which one you use is up to you. HUFFYUV is still popular, Lagarith has more compression, if you're on Intel you might check out UT (haven't tried it on AMD - it's Intel optimized), PicVideo has long been popular, WMV at high bit rates can work, but can also be a slow PITA, Avid I think has some free codecs designed for intermediates, some folks like Quicktime mjpeg etc...

"when you say a lossless codec, what do mean please?"

A lossless codec is one that doesn't throw anything away. You can render using it through an unlimited number of generations and the quality will remain the same.
In this case you would also want to retain as much of the original data in your source as possible. For example if you had shot HDV and were delivering a SD DVD using NTSC DV as an intermediate codec would cause some data (chroma) to be lost compared to rendering directly from HDV to mpeg-2.

For what you are doing the 8 bit Sony YUV codec would be quite good enough. There's two reasons I like this codec as an intermediate despite the moderately large file size. Firstly it's lightweight, that means it's easy to decode. Secondly it supports multiple audio channels. The later lets you mixdown lots of audio tracks into a manageable number for doing a final mix as you join all the scenes of your movie. Thirdly it's compatible with Digital Betacam tape, the industry standard of SD content distibution.

Bob.

I render each chapter separately to split up the render times into smaller chunks, can I then import the mpeg 2 chapters back into Vegas to join them all up? Or can DVDA join chapters together? There are several ways to do exactly what you want to do. Here they are.

1. Render as separate MPEG-2/AC-3s from Vegas. Then, in DVDA, insert a Music Compilation. Put all your MPEG-2/AC-3s into this. DVDA will create a single titleset that can be navigated easily by any set-top player. In essence, DVDA joins together the MPEG-2 files, just like you want. The one downside to this approach is that you can only have chapter marks at the boundaries between the MPEG-2 files. You cannot have chapter marks within an MPEG-2 file that is contained within a music compilation.

This is the simplest method, however, to get what you want. It is fast, completely avoids recompression, and gives you compatibility with all players.

2. Put the MPEG-2 files back onto the Vegas timeline and then re-render. Vegas 8.0 and later CAN do lossless combining of MPEG-2 files, but there are some caveats. First, all the MPEG-2 files must be rendered at the same bitrate. This restriction does not apply to programs like Womble, but it does apply to Vegas. Second, to get no recompression, you must specify Constant Bitrate (CBR) and set this to the exact average bitrate you used when encoding the files. You should see "No recompression required" in the Vegas preview screen when rendering. This gets you exactly where you want to go, but it is a little "touchy." I just tested it on some MPEGs I rendered a few minutes ago and it joined them together perfectly without recompressing. As for audio, render your initial audio as PCM, and don't do the final render to AC-3 until you have joined all the MPEG-2 and audio files back together on the Vegas timeline.

3. Use an external tool like Womble and VideoRedo. This has already been suggested.

4. Put the individual MPEG-2 files into DVDA and then link them with end actions. There will be a pause between the files if you do this, and the chapter navigation can be problematic. There are tools that you can use after you have authored the DVD in DVDA that can trick your DVD player into using chapters to navigate titlesets (PGCEdit is the tool) but doing that is beyond what I have time to describe. If you use this method, you need to pay attention to the titleset order (File -> Order DVD Titles) and the order in which the titles appear in the asset list on the left side of the DVD screen. If you don't, even if your DVD player supports titleset navigation using the chapter control (without the PGCEdit hack), you won't get the navigation you expect. In other words, "end action" only control what happens when a titleset (individual MPEG-2 file) ends, but does NOT control the navigation logic initiated through the remote control.

There are other ways beyond these four, but that's all I have time for now.

Hey everybody, thanks for your input. I have plenty to think about.

>"As for audio, render your initial audio as PCM, and don't do the final render
>to AC-3 until you have joined all the MPEG-2 and audio files back together
>on the Vegas timeline."

Good point John.

Audio with mpg2 can be fun... wav is cool except a lot of tools & even players reject it. AC3 with the exception of Womble AFAIK by it's nature can't be edited, plus like mpg audio loses quality from pcm. For cutting/joining it's probably not the best method, but I save the original pcm (wav) & match it as necessary to whatever accompanied the mpg2 as I mauled it, with the benefit of being able to do L-cuts that way.

OK, I decided to render in one go. So far, the render has been running 26 hours and has got to 32%. Just to clarify:

* Computer is dual core running VP9c and Win 7 and footage is Pal 1920x1080
* FX include; vignette, green screen moving background, some pan and scan, Cinescore
sound track, dialogue, and various stills fading in and out.

From what I've read from others on this forum, the time might be about right, and I could expect to go over 40hours.

I was wondering - is there a way to apply the FX selectively and progressively to ease the burden later? For instance, when using Sonar (DAW work station), I can add an effect then apply the effect to the track, which does the processing there and then so that the processor does not have to deal with it later - it is already done. Keep in mind that Sonar is audio only, and applying FX was needed more in the days of slower computers. Anyone care to comment please?

"I was wondering - is there a way to apply the FX selectively and progressively to ease the burden later?"

Yes but you've still got to render it at some point in time.
The big advantage to rendering either down or along (tracks or chunks) is you can avoid having to redo ALL the rendering just because of a simple change.

The simplest and most productive way to do this (think I mentioned this before) is to render say scenes to something like the 8 bit Sony YUV codec. Then assemble all the scenes on a 'master' T/L and encode from that to mepg-2 and wav files.

I don't know why there's some rumour about .wav audio being a problemo in DVDs. Never, ever had such a problem, it's part of the basic spec. The only thing that could happen is .wav uses more bandwidth so you might need to watch the combined bitrate.

Bob.

Ok, I went to bed last night with the render on 36%. When I got up, the render had stopped with a message about being out of memory. Then I remembered that a 16 minute movie of my grand daughter playing a flute had taken over 7 hours to render using both Sony soft focus FX set to black vignette and Sony sharpen. So I took these out and also dropped the render quality from best to good. So far, small test sections have gone well, but I do miss the nice vignette effect. Are these two FX processor intensive?

Quote from Bob:
"render say scenes to something like the 8 bit Sony YUV codec"

Bob, are you meaning stage by stage, say for the CK to remove the green screen? Then stage 2 for the moving back drop, then stage 3 for pan/scan etc? Or am I mis-reading what you are saying?