Advantage to Capturing Uncompressed?

T

toddburch

Active member
Hi all,

Is there any quality advantage to capturing DVX100A footage with an uncompressed codec rather than the DV/DVCPRO codec?

I'm capturing with a Pinnacle Systems Cinewave system, which I suppose is overkill for 4:1:1 Mini-DV video.

I've noticed when trying to capture to different codecs via Firewire that the only codec that directly allows for 2:3:3:2 pulldown removal in FCP is the DV/DVCPRO codec. (I guess this is due that the DV/DVCPRO codec is the only codec that "recognizes" the media as being flagged with 2:3:3:2 pulldown). If I attempt to capture using an 8 bit uncompressed codec or the Targa Cine YUV codec that it won't remove the 24pa pulldown, and I have to do it manually using Cinema Tools.

So will my footage look better if captured uncompressed, or is the use of the DV/DVCPRO codec the best you can get with 4:1:1 Mini-DV footage?

Thanks,
Todd
 
Re: Advantage to Capturing Uncompressed?

Firewire to the DV codec allows you to get all the data exactly as it appears on the tape. It is, theoretically, the highest quality transfer you can do from a DV source.

Others have had good success with digitizing the footage in through an SDI connection. While theoretically it should make no difference, the possibility exists that whatever hardware is doing the conversion to SDI could be enhancing the signal as well: smoothing out the chroma, up-rezzing to 10-bit or 12-bit video, etc. Once you have your footage in 10-bit mode, if you were to continue editing in an uncompressed codec with a higher bit-depth, the cumulative results could result in better-quality footage.
 
Re: Advantage to Capturing Uncompressed?

Thanks Barry, that's exactly what I wanted to know!

But how do I capture SDI from the DVX100? I have the SDI digital tether for the Cinewave system, but is there a Firewire to SDI converter, or an SDI output hookup available for the DVX100?

-Todd
 
Re: Advantage to Capturing Uncompressed?

There are firewire/SDI bridges out there... but mathematically it doesn't seem to make any sense to do so... if you can go firewire, you'll be getting the raw digital data. Any up-rezzing to 10 bit, or chroma smoothing, or anti-aliasing or whatever, should be possible to do once you have the data in the computer...
 
Re: Advantage to Capturing Uncompressed?

Todd,

As an editor i can tell you that one of the biggest advantages of uprezzing footage is for color correction. The more info(pixels) that you have to work with in color correction the more even your adjustments will look. So, whenever you have the hard drive space and hardware to do so, the better result you will have.
 
Re: Advantage to Capturing Uncompressed?

Technically, when you roll footage into an NLE, at least one like FCP (not sure about Pinnacle) it is not a capture, but a clone (apology for being an engineering grog). The material you get in the NLE is exactly what you had on the tape, and there is no need to re-encode.

The rule of thumb I teach to students, both professional and college, is that you can always loose quality (for example, each time you have to break your code and recode it you loose quality, so avoid DV - MPEG2 - DV traps, not that this should be an issue with a normal work flow) but rarely can get it back. There is one exception.

It matters little if you send DV25 in on SDI or through firewire, the codec did its damage in the camera. Same with DV50 or even the new DV100 codecs. The process of recording to tape did you in right from the start and no matter how you clone to your NLE you still remain the same. If, by moving through the SDI connection, there has to be an re-encode, even one to a higher quality format, you may even end up with a lower quality product (simple test of this theory - take a jpep medium image in photoshop, and resave it as jpeg - high, and you will notice a slow loss of actual resolution and a "blockiness" that the raw image saved as jpeg - high does not have, also you will notice the jpeg ghosts getting stronger).

The one exception is that, as you clone all of your product into your computer, if you have the hard drive array to handle the speed and the time / card (rendering will be slower), you can covert each video clip to an uncompressed format and then edit using FCP or another high end editor with an uncompressed timeline. The advantage of an uncompressed time line is that key clips are smoother, graphics and special effects are cleaner, and rendered product from 3D applications, which can also be done in an uncompressed codec, comes it at its cleanest state. When you color correct and match your clips, the uncompressed codec you are working in will also produce a better result, as DV25 is a rather harsh compressor and was never really designed to do with it what is now standard practice. If your footage was lit well and clean in the first place, it will give you probably the most out of the DV25 originals that you can get.

The caveat is that if you are simply returning to DV25, all that is not a good idea, since in the long run the DV25 footage will need to be rerendered back into DV25, loosing quality. If you are planning on rolling your stuff to film, or even sending it to a harsh compressor like MPEG-2 or MPEG-4 for distribution on DVD or the Internet, then it is a winner. You would still have had to convert your DV25 over to MPEG-2 with the loss of quality that entails, and the product that is usually hardest hit in a strong compressor, your graphics, starts off clean for the compression.

If all this does not make sense, e-mail me and I will send you my hand out "Understanding Compression for Artists" (have to find it, it is packed away for summer) which tries to discuss compressors and their use in production.

Steve
 
You must log in or register to reply here.
Back
Top