firewire vs. Component

[Mike Moncrief]

Hello all,

This has probaly been discussed somewhere here before.. I just happened to have a casual discussion with an engineer yesterday who does equipment rack installs and systems setups for a variety of editing houses.. And he mentioned that when using a mini DV deck with a tape that was captured using a 3 chip mini dv camera, it is better to capture the footage using the component output of the deck versus the firewire .. I would have guessed that the end result of capturing the 2 different captures would be pretty comparable, but I would lean toward using firewire as it is a file transfer versus an analog signal ... Of course the analog capture also requires an RS 422 control, versus firewires built in control.. Any thoughts on capturing one versus the other???

Thanks,
Mike M.

[Bill Ravens]

No matter how you slice and dice it, if it's recorded on miniDV tape, it's been compressed to the DV specification. You may save the minor losses of decompressing DV and recompressing to your capture format, however, I think you'll find that capturing via component won't be noticeably better than capturing over firewire, unless you capture to an uncompressed or lossless format. In that case, you save one generation of recompression, and, if you're using a quality DV codec, the losses won't be that noticeable.

[Heiko Saele]

You are absolutely right, component video is analog while Firewire is lossless 1:1 digital transmission.
When using component you have a conversion from digital to analog and another conversion from analog to digital, it definitely loses quality.
No offense but the engineer can't have a clue about how digital camcorders work...

[Daniel Browning]

Originally Posted by Mike Moncrief

I just happened to have a casual discussion with an engineer yesterday who does equipment rack installs and systems setups for a variety of editing houses.. And he mentioned that when using a mini DV deck with a tape that was captured using a 3 chip mini dv camera, it is better to capture the footage using the component output of the deck versus the firewire.

Short answer: No.

Long answer: At the very best, the quality will be the same as firewire. But there are several ways for it to be worse: progressive-to-interlaced (1080p is NOT supported by component), digital-to-analog conversion in-camera, and interference.

It does have the potential to make the workflow easier, but only for crippled software. The capture software should read the HDV coming in through firewire and convert it to an appropriate intermediate codec (Cineform, lossless, etc.) on the fly, all digitally. However, if the capture software cannot read HDV, then it must be converted digitally in a second step; so reading it in by component would save a little time.

[Dean Sensui]

If the signal has been recorded to tape, compression and encoding has already taken place.

Transferring via Firewire then becomes a digital transfer of data and, technically, lossless.

Going via component means the deck has to translate the digital data on the tape to an analog signal, and that gets transferred via cables and connectors. Back in the days of Betacam, there was a loss for every generation passed from one deck to another. It's the same today. If you transfer via component, you'll get losses through the cables and connectors. You'll also expose your video to additional electrical noise.

[Daniel Browning]

Originally Posted by Bill Ravens

You may save the minor losses of decompressing DV and recompressing to your capture format, however, I think you'll find that capturing via component won't be noticeably better than capturing over firewire, unless you capture to an uncompressed or lossless format. In that case, you save one generation of recompression, and, if you're using a quality DV codec, the losses won't be that noticeable.

Bill, at first I thought you meant that capturing via firewire to m2t then converting to an intermediate codec would result in generational loss due to recompression. But I think I understand now that you are just saying that using any intermediate codec at all will result in fewer recompressions. That's usually correct [1], and I would add that it's just as easy to use a quality intermediate with firewire as it is with component.

[1] With the right NLE, project settings, and workflow, you can use HDV source directly with no recompression: it's just slow. Decompression speed is the most important reason to use an intermediate.

[Bill Ravens]

What I mean is really not in reference to HDV, but to DV. Intermediate codecs are usually mentioned in reference to capturing HDV content. My understanding of the original question was in reference to a DV deck, in which there is no intermediate codec, everything is compressed, stored and distributed in the DV codec, which is quite a bit lossier than an HDV intermediate codec. Multi-generational losses in DV are more striking than in an intermediate codec, particularly where color reproduction is important.

[Daniel Browning]

Originally Posted by Bill Ravens

What I mean is really not in reference to HDV, but to DV. Intermediate codecs are usually mentioned in reference to capturing HDV content. My understanding of the original question was in reference to a DV deck, in which there is no intermediate codec, everything is compressed, stored and distributed in the DV codec, which is quite a bit lossier than an HDV intermediate codec. Multi-generational losses in DV are more striking than in an intermediate codec, particularly where color reproduction is important.

I see. While the convention is to capture firewire to DV and component to a higher quality intermediate, that's all it is: convention. One may capture firewire to uncompressed (125Mbps) or component to DV (25 Mbps). The point that we're both trying to make is that a quality intermediate codec is important.

You're right about the OP referring to DV; I read HDV into everything, apparently.

[Glenn Chan]

Uncompressed usually refers to 4:2:2 uncompressed or 4:4:4 uncompressed (with either 8 or 10 bits)... that data rate should be more than 125mbps (for the video portion).

2- You can get different signal processing depending on which route you take.

If you take the SDI out from a deck, you are using the decks 4:1:1-->4:2:2 conversion.
In a NLE, you are using the NLE's 4:1:1-->4:2:2 conversion. Most NLEs upsampled using box reconstruction, which gives a boxy appearance to the chroma and isn't good for chroma key.

3- In higher-end editing facilities, you will see them take the SDI output off a DV/DVCAM deck. One advantage is that their other gear simply doesn't take DV/firewire in (e.g. other VTRs, certain online editing systems).

So by component, they might be referring to component digital / SDI. (Analog component would be stupid IMO.)

[Mike Moncrief]

The deck the engineer was refering to does not have an SDI port.. He was referring to the compnent outputs..And I tend to agree, that it is a bit foolish to go backwards and go Analog..

Mike M.

[Daniel Browning]

Originally Posted by Glenn Chan

Uncompressed usually refers to 4:2:2 uncompressed or 4:4:4 uncompressed (with either 8 or 10 bits)... that data rate should be more than 125mbps (for the video portion).

You're right. 10-bit 4:2:2 480i is 216 Mbps, double what I said. 10-bit 4:4:4 is 316Mbps.

[David Kovalev]

I know it might be a little off subject but I have search this forum near and far and couldn't find an answer to my question. Maybe somebody can help me out. I was wondering if quality differs if you capture the tape (miniDV) through the "good" camera or a small home camera? I would prefer to capture it through the home cam to save a bit of life on my "good" camera....

[Glenn Chan]

David, it's better if you put not-quite-similar questions in their own thread. The discussion will be less confusing and you'll get more responses that way.

To answer your question...
The quality will be the same. In some cases, your other camera will not play back the tape as well as the original camera that shot it due to different head alignments. Cameras are sensitive to that if you shoot LP mode material.

[David Kovalev]

Ohh sorry about that. It makes sense not to go too far off topic. Thanks for the response. I shoot in SP mode so it shouldn't make too much of a difference.