Backing up my film, does it matter if it's printed on cheaper tapes?

H

HKFF

Active member
Hey Everyone,

I am about to print my movie to tape so I can take it to a dubbing lab to transfer it to BetaSP for film festivals.

The question is, does it matter whether it's printed on a Master Panasonic tape, or not? Will it be worse quality or not as good for archive purposes if it's printed on say a cheaper sony tape?

Thanks for your advice!!
 
export your movie as a self-contained movie and have a facility export out of their SD or HDV card directly to beta. You'll lose 20-30% of your quality if you go from Hard drive to mini-DV then beta. You're better off paying extra to go directly to beta from your self-contained movie or whatever it is you export out of if you used compressor.
 
I'm not sure what WaveRider is talking about... perhaps assuming you're working in an uncompressed format? Or perhaps thinking that DV playback to the beta would be analog? If the dub house transfers your video to beta digitally, then it shouldn't be different from an exported DV file...

Anyway, regarding tape - no offense, but, have you invested all your time and effort into this in order to risk the whole thing by skimping out on the tape? If it were me, and I'd spent a lot of time on it and if I wanted to make sure to have a good transfer and good archive, I'd archive to at least 2 or 3 Master Quality tapes, kept in separate locations.
 
What I think he may be referring to,
is that if you have done a lot of CC and compositing, you may see a better result if you render out master as uncompressed timeline with high precission YUV settings.

This - of course - is not the case if it is only cuts.

In this case - to play back this file to a digi master may be a good option - also for a master backup.

gunleik
 
Thanks for all your advice everyone.

gunleik - what do you mean by "This - of course - is not the case if it is only cuts."

Wouldn't printing to tape from fcp to a minidv then to beta transfer if my footage was originally shot on mini dv be the same quality?

Why is taking an uncompressed exported (like a qt self contained movie) to the dub house better? Is it because quality may be lost because I will be transferring from one tape generation to another?
 
Mini-DV tape won't degrade the quality of your video, as it's just digital data on the tape, not analog images. The higher quality Mini-DV tape only ensures longer life and reliability of that digital data you put on it.

The moment you output to a self-contained QT movie, you've re-compressed and "technically" lost quality, I don't recommend it if you HAVE to keep top image quality and are going to DVD.

In your case, I'd go out to Mini-DV.

If it's HD footage, I'd go out to a QT movie file and just not worry about it. The image quality loss when making a self-contained QT movie is very minimal, not much at all, but is there, none the less.
 
what he means is you can print back to a mini dv if you only made cuts and didn't alter the look of your footage. If you did Color Correction or anything like that, then sending a self contained file to them might look better than sending them a tape.
 
A self contained QT won't retain any more clarity than printing directly off to Mini-DV. In fact, it will actually lose some quality in the re-compression. You're best off going directly from your Timeline window to Mini-DV in that situation.
 
Not to steer the thread OT, but I thought when you made a self-contained QT movie you were getting exact copies of the used portions of your master clips. Why is there another round of compression added?


-A
 
But come on everybody, we need to be clear:

This quality issue involves ONLY the final compression step.

1) If you are in a DV timeline, and export a DV file, it's no different from printing to tape. The step of recompressing any modified frames (CC'd, faded, etc.) to DV is done in both cases - the only difference is the medium on which it's stored. And, unless you check "recompress all frames", there's no extra compression going on - umodified DV frames are just copied.

2) If you are in a "better than DV" timeline (uncompressed, DVCPro 50, whatever), and export to a DV file, it's no different from printing to tape. The step of compressing all frames to DV is done in both cases - it's just a storage issue.

3) If you are in a DV timeline, and export a "better than DV" file, you may have a better quality exported file than is possible in DV.
a) Any frames that are not modified (CC'd, faded, zoomed, etc) will be exactly like the DV frame; although if you at some point do compress this file back to DV, it'll have been an EXTRA decompression/recompression cycle that'll make the frames a tad worse than keeping them in DV all along;
b) Any frames that you've done anything to (CC'd, faded, moved, put titles on, sped, slowed, brightened, you name it) will look better in the output file than on DV, and would have gone through a second recompression anyway even if kept in DV all along.

So, if the file you export is DV-format, then there's no quality difference in any case from what you'd send to the tape - to get to DV (file or tape) the same compression has to happen. Only if the file is a "better than DV" format will you see a quality increase. Of course, the "better than DV" file will be much larger than the DV file.

The TAPE TYPE issue is not one of video quality in this sense, it's one of glitches and longevity. Using good tape, and multiple tapes, wiull improve your chances of having glitch-free video on playback/recapture.
 
When you create a "self-contained" QT movie, yes, there is re-compression there, for sure. If it were not self contanied, it'd just reference the masters. But now, you've taken only sections of all those masters and are re-compressing them into a single movie file. Yes, it is re-compressed, that's the fact.

When you print to tape, FCP is feeding a video stream that references the render files, thus, no re-compression.

I have to simply say Ulanta is just not being technically acurate, as any type of self-contained QT movie IS IN FACT re-compressed into a new file. Sorry, that's just how FCP works.
 
BenB - again, the difference is the final format. If you are printing to DV (or miniDV)tape, then the video has to be in DV format. That means the stream, whatever render files it is generated from, is output as DV. Which means recompression of anything that's not DV. miniDV does not allow any "better quality". (Let's not discuss HDV)

If you are in a DV timeline in FCP, all the render files will be rendered to DV. That's recompression.

A self-contained QT file involves no more or less compression than a reference QT file.
The stream of frames produced from playing back either is exactly the same. The self-contained file merely agglomerates all the referenced files into a new file. Any files referenced by a reference file are ALREADY in DV format - either original, or recompressed.

FCP will NOT recompress frames where it's not necessary, in either output type, unless you check "recompress all frames."


Sorry, BenB, THAT's how FCP works.
 
I guess you know more about the techincal insides of the compression engines than I do. But you're still wrong. I'm done here.
 
Hey BenB, I don't mean to tick you off. It'd be better all around for everyone if you can stick around and explain what you mean. If your experience contradicts what I'm saying I'd love to hear it, and especially, to understand it.
 
BenB said:
When you create a "self-contained" QT movie, yes, there is re-compression there, for sure. If it were not self contanied, it'd just reference the masters. But now, you've taken only sections of all those masters and are re-compressing them into a single movie file. Yes, it is re-compressed, that's the fact.
Click to expand...

I think this is the confusion - it sounds like you're confusing "compression" and "consolidation". A Quicktime reference movie does not do compositing or recompression on-the-fly; thus it's not "referencing the masters" for anything that's been modified. It's referencing the render files. And the render files are in whatever format the timeline is in - in this case, say, DV. So, in the process of GENERATING the render files, the DV recompression is taking place.

When you print to tape, FCP is feeding a video stream that references the render files, thus, no re-compression.
Click to expand...

So, this is correct as far as using the render files, and not compressing them further, but they've ALREADY been recompressed to DV.

The difference between a self-contained movie and a reference movie is that the self-contained movie has only the exact material needed for the final stream... not a single extra frame. The reference movie depends on original masters for unmodified frames and possibly render files for modified ones. The consolidation process which generates the self-contained file copies all untouched frames exactly, copies properly rendered frames exactly, and renders any frames that need rendering but for which there's no current render file. In fact, except for time constraints, the consolidation process for a self-contained file is exactly the same as the process of playing a reference movie. So I guess you might say the self contained movie is "compressed" compared to the reference movie, since it takes up less HD space then the reference and all its referents... but that has nothing to do with video compression - each frame is identical.
 
I'm not sure if I'll find the answers I'm seeking as people have bailed from this thread but here it goes anyway.
All of the following is assuming I'm working 100% in DV.

There are a few reasons why I thought File->Export->QT movie->"make movie self-contained" (I'm away from my machine, but I think that's path) didn't recompress the footage.
1. When I read up on it in the FCP manual the idea that I got is that it basically did a 1:1 copy of the required media and strung all the pieces together into a single QT file.
2. There is a "recompress all frames" checkbox. What function does this checkbox have if "make self-contained" already recompresses everything?
3. If I use the QT Conversion option to export a DV file of Sequence XYZ it takes longer than exporting Sequence XYZ via the self-contained method. I took this as an indication that the former was compressing the sequence (thus adding another round of DV compression) while the latter was just copying media (which is why it went faster).
4. I assumed the QT Movie and QT Conversion options operate differently "under the hood" otherwise they seem like very redundant features.


I'm not looking to have a geek-fight or anything, I'm just looking for what's correct.


-A
 
adkimery said:
1. When I read up on it in the FCP manual the idea that I got is that it basically did a 1:1 copy of the required media and strung all the pieces together into a single QT file.
Click to expand...
That's correct.
2. There is a "recompress all frames" checkbox. What function does this checkbox have if "make self-contained" already recompresses everything?
Click to expand...
Your assumption is correct... unless you check this, only modified frames are recompressed. Otherwise they're just copied.
3. If I use the QT Conversion option to export a DV file of Sequence XYZ it takes longer than exporting Sequence XYZ via the self-contained method. I took this as an indication that the former was compressing the sequence (thus adding another round of DV compression) while the latter was just copying media (which is why it went faster).
4. I assumed the QT Movie and QT Conversion options operate differently "under the hood" otherwise they seem like very redundant features.
Click to expand...

You don't need to do "conversion" - there's no conversion necessary in DV to DV. I've never tried it... interesting that it may be doing the conversion anyway (meaning, it's likely expanding each frame to uncompressed and then re-compressing it). For kicks, have you timed this against QT Movie export with "recompress all frames"? Seems like it may be the same, more or less....
 
For a 30 sec section of a DV sequence w/a couple of A/V cross dissolves I got the following times on my 5yr old Mac:

Self-contained - 7 seconds
Self-contained w/recompress frames - 22 seconds
QT conversion to DV - 37 seconds

Now just a straight DV clip (no edits):
Self-contained - 6 seconds
Self-contained w/recompress frames - 19 seconds
QT conversion to DV - 36 seconds


-A
 
adkimery said:
For a 30 sec section of a DV sequence w/a couple of A/V cross dissolves I got the following times on my 5yr old Mac:

Self-contained - 7 seconds
Self-contained w/recompress frames - 22 seconds
QT conversion to DV - 37 seconds

Now just a straight DV clip (no edits):
Self-contained - 6 seconds
Self-contained w/recompress frames - 19 seconds
QT conversion to DV - 36 seconds


-A
Click to expand...

Interesting! Any qualitative difference?
 
You must log in or register to reply here.
Back
Top