Geoff, you can trim off the beginning or end of a compliant video without causing recompression. Double click the thumbnail to open it in the Timeline pane and move the yellow flags. I occasionally have to do it when compiling DVDs with other people's files that contain too much black at beginning or end.

Tried that.

It seems to change the amount playable on the resultant DVD, but not the total data size written to the disc.

... which is a pity. Sure, reducing the source media is an 'editing' function, but such a feature you would think would be trivial to implement . Or maybe not ?

geoff

The easiest solution is to render to a folder, ignoring the warning that it's too big. Then open the VIDEO_ts folder in DVDShrink and allow it to automatically reduce the size to fit a DVD.

The process is fairly quick and painless an will allow you to keep all the content.

I want to avoid re-encoding, which DVD-Shrink does, or DVDA does if you tell it to.

geoff

If your project, when created to a HDD folder, is marginally bigger than can be fitted onto a DVD, DVD Shrink will create a standard size DVD folder with no noticeable degradation. Before embarking on Blu-Ray I used DVD Shrink whenever my projects were too big. Try it. You'll be surprised at the resulting quality.

How can it do that - a better encoding algorithm than anybody else has been able to achieve ?

I'll give it a try then !

geoff

From memory, DVD Shrink doesn't re-encode, it compresses. Can anyone confirm?

The impression i got from the description a few people in here have given is that it's smart enough to find wasted space in the MPEG2 encoding that isn't necessary, and copy only what is necessary. Sounds a bit like voodoo to me. I can't imagine any decent MPEG2 encoder putting much wasted space in the output file.

In any case, if the needed reduction is more than a few percent, it will decode and re-encode, which will be an additional compression process.

That being said, i've seen stuff 'shrunk' with it, and it is pretty impressive.

Geoff, regarding trimming in DVDA, you are correct in that all it does is restrict the playable portion of the file. Then entire file, including the trimmed off portions are still physically burned to the disc.

For estimating project size i've never gone wrong using the formula 600 / minutes - audio bandwidth.

For example, if i have an 85 minute project, 600 / 85 = 7.059. Subtract about 0.192 for my usual audio template of 192Kbps = 6.867ish. Encode at an average bitrate of 6,866,823 and it'll fit. I usually subtract an additional 1% or so for space for menus and wiggle room, but when i do that i always end up with an image that occupies 99% of the disc, so i'm just being paranoid.

Never pay attention to any numbers that DVDA shows on the screen.

With VBR, It's impossible to estimate the compressed file size without compressing the video. This is because the compression is a moving target. Scenes with little motion will compress easily, and scenes with a lot of motion and changing background need a lot more compression.

If your VIDEO_TS folder in Windows is 4.3Gb (as reported by Windows Explorer) or less, then it's fine to burn your DVD.

Steve, that's why the compressors offer a 2-pass option. That does allow an accurate estimate. Using 2-pass i've never had a final render be more than a 10th of a percent different from the estimated size.

The render as 2-pass option still doesn't know what your video contains. A one-hour talking head interview or panel presentation will render a smaller MPEG file than a one-hour sports video with lots of pans and changing backgrounds.

The render as 2-pass option still doesn't know what your video contains.

But surely after the first pass it does so on the second pass it uses up every bit giving you the same file size as if you'd used CBR?

[i]"The render as 2-pass option still doesn't know what your video contains."[/]

That is exactly the point of a 2-pass render. Vegas experiments with the contents of the video on the first pass, seeing what it can compress by how much. Then when the second pass starts, it does know what the video contains.

I want to avoid re-encoding, which DVD-Shrink does, or DVDA does if you tell it to.There are several slightly incorrect statements about DVD Shrink in this thread.

While it is true that DVD Shrink does not re-encode, it DOES change and degrade the video. The guy who actually wrote DVD Shrink provided a detailed explanation of how it worked, over a decade ago. The original post no longer exists, but it is quoted here:

DVD Shrink Explained

The key paragraph from the long, but excellent explanation:

DVD Shrink achieves compression by removing some of the residual data. This process is called requantization. Selected DCT coefficients are scaled down (thus reducing the number of bits required to store them) and a corresponding scale value for these coefficients (quantizer scale) is scaled up. The result is a less accurate description of the same residual data, which takes up less space. Note motion vectors are left unmodified by this process."