'Cancel Render' button

[Graham Bernard]

Start a TICKET with Sony Media Software at Madison.

Tell you what, and with no effort at all, Ian, I'll even hand you the website address:

http://www.sonymediasoftware.com/sup...calsupport.asp

Regards,

G

[Douglas Spotted Eagle]

I too, get a partial file when rendering to avi.
That said...
Grazie, are you rendering to DV first, and then using that DV file to render your MPEG from? I sure hope not, as you're tossing so much away it's very much worth mentioning.

[Graham Bernard]

Originally Posted by Douglas Spotted Eagle

I too, get a partial file when rendering to avi.

Douglas Is that with V6 or V7?

Originally Posted by Douglas Spotted Eagle

Grazie, are you rendering to DV first, and then using that DV file to render your MPEG from? I sure hope not, as you're tossing so much away it's very much worth mentioning.

I sure hope I aint missing anything too?!? How much, Douglas, could I be tossing away when I render to a "New Track" and use the template? Please spell out the actual format I SHOULD be rendering to?

You have my full attention.

TIA - g

[David Jimerson]

Originally Posted by Plamen Petrov

Vegas DOESN'T leave any partial except .WMV!!! All other format files (.avi, .mpg, .mov, etc.) disappear totally if you cancel Vegas rendering.

In my experience, that simply isn't true.

What are you rendering that it takes you 20 hours to do?

[David Jimerson]

Originally Posted by Graham Bernard

Douglas Is that with V6 or V7?



I sure hope I aint missing anything too?!? How much, Douglas, could I be tossing away when I render to a "New Track" and use the template? Please spell out the actual format I SHOULD be rendering to?

You have my full attention.

TIA - g

I don't have ANY projects where I don't do at least one mixdown before rendering to MPG for DVD.

Now, rendering from HDV to DV before going to MPEG might not be the best move, but from DV to DV, I don't have any issues. I don't think I'd do it too many times before the DVD render, but once is certainly nothing to worry about.

[Graham Bernard]

Originally Posted by David Jimerson

I don't have ANY projects where I don't do at least one mixdown before rendering to MPG for DVD.

Now, rendering from HDV to DV before going to MPEG might not be the best move, but from DV to DV, I don't have any issues. I don't think I'd do it too many times before the DVD render, but once is certainly nothing to worry about.

Breath out . . . and relax . . .

Glad to hear I wouldn't be tossing-out much - if anything? Would NOT want to be regarded on the forums as a "tosser" of quality! . . Or any other type of . . . remover.

Still working with SD, I need all the quality I can.

[Douglas Spotted Eagle]

Originally Posted by David Jimerson

I don't have ANY projects where I don't do at least one mixdown before rendering to MPG for DVD.

Now, rendering from HDV to DV before going to MPEG might not be the best move, but from DV to DV, I don't have any issues. I don't think I'd do it too many times before the DVD render, but once is certainly nothing to worry about.

Run a color chart through the system, be sure to put text on top of it. Render it to MPEG 2 from the timeline. Render it to DV from the timeline. Render the AVI master to an MPEG2.
Compare them, particularly the text or graphics.
Then come tell me you're losing nothing.

[Graham Bernard]

Could somebody please tell me what I should I do when I use "New Track" renders? I render up sections of stuff and chain them back up?

TIA - g

[David Jimerson]

I'm just saying what my experience is. Vegas's DV codec is very robust and can handle a generation or two.

But, doing as you ask, here's a composite:

http://img443.imageshack.us/img443/4125/compee9.png

This is made up of both MPEGs -- one rendered from the timeline, the other rendered from a downmixed AVI.

There's a cookie cutter on it, and it runs through both the text and all the color.

If you want to know where the cookie cutter line should be, it's here:

http://img114.imageshack.us/img114/138/comp2fu1.png


Here it is split a couple of different ways:

http://img251.imageshack.us/img251/170/compdiagzl6.png (comp image)

http://img267.imageshack.us/img267/3884/split1kc9.png

http://img339.imageshack.us/img339/4876/split2xm6.png

[Douglas Spotted Eagle]

I'll try again...
Shoot a color chart. With movement, as video would move. This will affect/imact the quality of the encode. Most video subject matter doesn't stand still.
Or, shoot whatever, so long as it has identifiable areas.
Put text or other graphics over it.
Render to both as you've done.

You'll definitely see a difference in the encode. Going back nearly 5 years, this has been a standard part of the Beta process for Vegas and the Main Concept encoder. We do this with both video and stills, and push the color and boundaries of the images as standard routine. The loss has always been, and probably will always be, visible regardless of the robustness of the Sony codec, which is quite good, but math is math/color samples are color samples. No one would ever say I'm not a fanboy of the codec, and was one of those pushing for it in the very early days, but..

Graphics are virtually always 4:4:4 or 4:2:2, uncompressed/lightly compressed
Text/Generated media is 4:4:4 uncompressed
DV is 4:1:1 compressed 5:1 (That's NTSC, it's a slightly different 4:2:0 format in PAL-land, but a transcode still takes place when going from DV Pal to MPEG-PAL)
MPEG is 4:2:0 compressed roughly 20:1

If you don't see the difference, or differently put, if you don't feel the difference matters, then there is no reason for 95% of what exists here on DVInfo.net in terms of color correction information, compression, m2t vs DVCPro, etc.
I know you're really into color, so for giggles, convert HD from 8 bit to 10bit, push the colors hard, and tell me you don't see banding differences? It's essentially the same thing.
I see it, assume everyone else sees it.
Hollywood DVDs look so much better because they're starting with the least compressed source to the encoder. The encoder will *always* do better with the least compressed source whether it's uploading an AVI vs wmv to YouTube or feeding a high end Minerva encoder. Rendering from 4:4:4 uncompressed to 4:1:1 reasonably compressed and transcoding from that 4:1:1 "Master" to 4:2:0 is reducing your chroma values by huge percentages once you add it all up.
Why, if pristine video is the goal, would you convert 4:4:4 to 4:1:1, to lose yet another 50% in going to 4:2:0? You're compressing those graphics twice and neither compression is a very good one.
Put differently, if it doesn't matter, why do most folks doing vid for a living, capture at the highest color sample they can, such as capturing even HDV over a 4:2:2 system vs importing the original m2t?

All of that said, if you can't see the difference, then it likely doesn't matter to you. Both my high end and low end monitoring systems allow me to see the diff, as do scopes.
To close, if you're not working for broadcast where you're working with 4:2:2 masters vs delivering on MPEG for DVDs handed out to friends, none of what I wrote above is worth a cold cup of coffee.

[David Jimerson]

So do a mixdown of everything BUT the text and render to MPEG with pristine text, if you find the text unsatisfactory. For me, this wasn't about the text, anyway.

All I'm saying is that a single DV to DV mixdown doesn't lose you much of anything. And if you're doing the text afterwards, it sure is going to be a lot quicker to rerender the project if you find you made a mistake in the text than it would be if you were to rerender the whole timeline, with color correction, transitions, cropping/panning, whatever.

[Douglas Spotted Eagle]

maybe this explanation is better/has clarity?
Color correction, chromakey, masking, and other processing is done in the 4:4:4 color sample range. Same with graphics, text, and *anything* that isn't DV source.
Even DV source is processed at 4:4:4 when processed, but left alone as 4:1:1 when nothing is done to the footage.
I suppose you could:
~render out any unprocessed footage that contains no transitions, color correction, or any other modification;
~Keep those segments separate from the color corrected/transitioned/masked segments of media;
~render new track for text/graphics only as a 4:4:4 stream, laid over top of your 4:1:1 unprocessed stream, and over the segments of the processed streams;
~render all of that to an MPEG file from the timeline.
And/or:
Render the entire segment to 4:2:2 YUV for the least amount of loss anywhere, and keep it as intact as a master as you can, using the 4:2:2 master for web submaster, MPEG submaster, DV submaster, and any other outputs you might want to use.

Or:
~You can render from the Vegas timeline to MPEG for MPEG delivery and render from the timeline to DV for archiving, and be done with it.

Doing text "afterwards" generally isn't an efficient option in corporate, training, infomercial work, because there are so many graphics, text events, etc. For templated work, this wouldn't work at all, IMO.

I'm sure that the standard workflow for most post/broadcast houses isn't fitting for everyone, but the workflow I've described is pretty standard, or seems to be in the post houses in which we've trained editors.
Sony, Apple, Sorenson, MainConcept, and other encoder developers recommend the same workflow.

This is part of the reason that you're always recommended to let Vegas downconvert HDV to MPEG 2 for DVD rather than capturing it as SD. There are at least 6 streams on the web I'm aware of (2 from us) that demonstrate the superiority of allowing the file to stay native all the way through, or by using an HDI, preserving the data for as long as possible. Recompression, particularly when dealing with already highly-compressed media, is a bad thing.

Whatever works, works. And at the end of the day, that's all that matters.

As I said before...if you can't see it, then it's likely a moot point. But recompression and transcoding combined, virtually always results in a softer picture. You'll also find the MainConcept encoder does much better from the raw timeline than from an AVI with high latitude or high motion, which is why MainConcept recommends encoding from the timeline. If you're working with 10 bit source, or 10 bit HDI's, would you also render those to 4:1:1 before MPEG? I sure hope not!
Graeme Nattress addresses part of this exceptionally well in his comparisons of DV vs uncompressed vs DVCPro. You can clearly see the problems with 4:1:1, and then halve that problem again by transcoding compromised media to 4:2:0.
Most of us try to preserve the source as best we can; that's why all the hubbub surrounding the Intensity, HD SDI capture, Convergent Designs converters, etc. People *want* to preserve the quality of their source. By rendering to 4:1:1 as a "master" you've tossed all that preservation out the window.

And to say it once again...if you're not delivering for broadcast, cable distribution, or other QC/vetted source, none of what I've written has any bearing on you.