So I guess translated to so my 4 brain cells can handle it this morning that means it doesn't matter what I just rendered - it's always gotta do it again?

Okay. I think I can handle this.

Thanks.

You have to render to DV to print to tape, yes.

But since you already rendered uncompressed you can render from that file to DV and maybe save some time (no need to render fades and effects again).

If you had rendered to the default DV template then it wouldn't have to be rerendered for printing to tape. The printing operation requires a file that is in exactly the same format as what will be recorded on the tape. This format is DV. As a side bonus, DV files are only about 1/5 the size of uncompressed .avi.

Of course, in order to use this DV render, you'd have to load that onto the timeline of a new project and print from there instead of printing from your current project. Vegas will still rerender the audio track though, but this is usually a very fast operation. Alternatively, use the VidCap program to print it to tape instead of printing from the timeline.

Chienworks wrote:
"... DV files are only about 1/5 the size of uncompressed .avi."

Is there any appreciable difference in the final images?

Jay

There is no difference between the DV and the uncompressed AVI--> if the file was originated/captured DV through your system. You can never recreate, regenerate or encode a file any better than where it started!So rendering uncompressed AVI is a waste of time and disk space if your project stated in the DV format.

""... DV files are only about 1/5 the size of uncompressed .avi.""

"Is there any appreciable difference in the final images?"


Technically any image compression that is lossy, throws out some of your image data, is going to be inferior to the original. Visually the result depends on your source video, what does the conversion to DV, and of course your eye - whether "you" will see a real difference or not.

In practice the question is normally moot, as few people or shops can deal with uncompressed video, and those that can, usually opt not to. Storage is cheap, but not that cheap, and the hardware cost to work with uncompressed can skyrocket.

DV (&/or mini DV) itself was developed for cameras, to allow them to capture video on tape, but in a digital rather then analog format. Because of this, the requirements were/are different then say for video archival, distribution etc., and other formats are often used once the video is off the camera so-to-speak. Beyond that, things can get fuzzy...

Trivia: DV itself has a somewhat different scan matrix and a slightly reduced color range, and different cameras, hardware, and software handle this differently, some doing better then others. The original idea was to have a set data rate, and vary the compression as nec., so the subject matter has a bit to do with it, as does stuff like the different lighting requirements for DV.

One of the ways I handle this issue is to render a new track (at the end of the project) using the DV template. If I find any problems I fix them, slice the rendered DV track, re-render the fixed portion and slip it into place on the top track which is the DV render.
On short shows (under 15 minutes or so), this is all not too much of an issue, but when you get to longer shows, you hate to rerender the whole thing, every time you want to make a DV print.
By keeping the top tracks the "finished final" (is there ever such a thing?) show, you can reboot the computer a week later and lay off to DV pretty quickly.
Plus, you always are certain that what you are working on is truly the "last" revision.

I too would like to hear, in "Ordinary language", an explanation of how DV, already compressed ~5:1 in camera, get's rendered as "Uncompressed"

Does it mean that whatever information was "left out" during the camera hardware compression is "recreated" and put back?

"Uncompressing" can't replace the information that was lost. It has to make up new information to fill the holes. This usually isn't as bad as it sounds though. Typical compression schemes look for areas in the video where a description of part of the frame can accurately replace the original video data, and where this description takes fewer bytes than the original. For a terribly naive and basic example, suppose that a large rectangular portion of the frame was a solid red area. The compression scheme could simply record that there was a group of red pixels, a certain height and width. This would only take a few bytes whereas uncompressed video would require three bytes for every pixel involved. Decompressing this description would restore the original red area rather faithfully.

With more real world examples, there have to be more complex descriptions. The area that is being described may be some other shape than a rectangle. It may fade from one color to another across the width of it. It may contain a smaller area that needs a different description. Formats such as MPEG also include temporal compression, which means that future frames can contain descriptors such as "this area is just like what came in the last frame, except moved slightly here and changed in this way ...".

With higher compression ratios, the compression scheme must find more data that can be thrown away and therefore preserves fewer accurate descriptions. The decompression then varies more and more from the original. Take a look at some very low bitrate media files, like Real or WMV at 56Kbps. You'll see that the compressor often smudges out details so that there are more smooth colored areas instead of sharp lines and contrast. These smoother areas compress better and require smaller descriptions. The decompression process has less information to work with and can only approximate the original to a poorer degree as the compression ratio goes up.

One interesting thing to consider is that you never watch compressed video; you always seen uncompressed! I suppose someone could write a player that would show you the compressed version, but believe me, you wouldn't want to see it. It would look worse than static on drugs. All players decompress the video before showing it on the screen.

Dratme, to answer your first question, to render DV as uncompressed all you have to do is plunk the DV clip down on the timeline, click File / Render As, and choose the uncompressed template. This is a very simple operation. For that matter, Vegas does it every time it shows a DV clip in the preview window (see previous paragraph).

"It would look worse than static on drugs" LOL - now, THAT's ugly...

There is a place for uncompressed video, altho' MJPEG might be better. Except for the file size, uncompressed makes for an OK data transfer medium since re-rendering doesn't introduce degradation to the uncompressed file. Every time a compressed DV file is re-rendered, even the best codecs suffer some losses. MJPEG offers a means of "lossless compression" that can be decompressed ad infinitum without quality loss. For all those intermediate re-renders leading to the final master, uncompressed or hi quality MJPEG may be the method of choice.

Generally, uncompressed isn't much use for DV footage. Uncompressed artifacts aren't better than compressed artifacts.

However, If you have nice sharp graphics and generated media you may find that rendering that section as uncompressed is useful, especially if you plan to do more processing on that clip in Vegas or another program.

As far as the render issue goes, consider printing to tape as a straight data dump. In that light you would of course have had to render the file in a format that allows it to be dumped back to tape. That would be the DV template, not anything else.

Rob Mack

VideoCurmudgeon:

DV is a very good image compression format, and the amount of compression (5:1) is very mild. There are differences between a DV image and a raw uncompressed image, but the differences are slight.

If you want to dig deeper, try spending a few minutes surfing around this site:

http://www.adamwilt.com/DV.html

Thanks for all the replies....you guys are great!!!