DV or DVCam signals?

[Saul Martinez]

I will appreciate a lot to whom explain me the differences between DV and DVCAM signals.
Thanks

[Andy Tejral]

Well, they ain't got nothing to do with JVC Pro HD Acquisition.

Ain't electronically, there ain't much difference between the two. In DV, the audio has a range of samples per video frame and in DVCAM there is a fixed amount of samples per video frame. DV is "unlocked" and DVCAM is locked.

As a tape format, DVCAM moves faster (a 60 min DV tape = 40 min DVCAM) and the pitch or distance between swipes of the head is greater and the size of the swipe is larger.

That's it.

[Jack Walker]

From Wikipedia.com:
DVCAM

Sony's DVCAM is a professional variant of the DV standard that uses the same cassettes as DV and MiniDV, but transports the tape 33% faster. This leads to a higher track width of 15 micrometres. This variant uses the same codec as regular DV. However, the greater track width lowers the chances of dropout errors. The LP mode of consumer DV is not supported. All DVCAM recorders and cameras can play back DV material, but DVCPRO support was only recently added to some models like DSR-1800, DSR-2000, DSR-1600. DVCAM tapes (or DV tapes recorded in DVCAM mode) have their recording time reduced by one third.

Because of wider track DVCAM has the ability to do a frame accurate insert tape edit. DV will vary by a few frames on each edit compared to the preview. Another feature of DVCAM is locked audio. If several generations of copies are made on DV, the audio sync may drift. On DVCam this does not happen.[6]

The link to the entire article:
http://en.wikipedia.org/wiki/DVCAM#DVCAM

[Tim Dashwood]

Incidentally, the ProHD cameras and decks can play back both DV and DVCAM formats.

[Saul Martinez]

Thanks to all for the lessons.

[John Miller]

The locked vs unlocked is marketing nonsense. This is very apparent if you read the official DV specification IEC 61834(2) and understand what the actual definitions of locked and unlocked are...

The claim in the Wikipedia article that drift will occur over multiple generations of copying is rubbish! DV-to-DV dubbing is done on a frame-by-frame basis. Each frame contains 120,000 or 144,000 bytes (NTSC or PAL) which, in turn, contain the audio samples, compressed video etc etc. No unpacking and repacking of the audio occurs. It is a direct, bit-for-bit copy.

Wikipedia is a great at-your-fingertips resource but the content isn't 100% accurate.

(I say all this as someone who has written his own DV encoder/decoder from the ground up.)

[Shaun Roemich]

I beg to differ on the "marketing nonsense" statement. Try editing an unlocked audio stream to a locked audio video recording device suchas a DVCam VTR (ie a dv tape digitally transfered to DVCam). You can't. The recording device stops dead. Professional DV is a different animal. It has locked audio.

[Jack Walker]

Originally Posted by John Miller

Wikipedia is a great at-your-fingertips resource but the content isn't 100% accurate

In comparisons a greater percentage of inaccuracies have been found in the major encyclopoedias (e.g. Britannica, World Book) than in Wikipedia.

However this is not a comment on the specifically referenced content.

[John Miller]

Originally Posted by Shaun Roemich

I beg to differ on the "marketing nonsense" statement. Try editing an unlocked audio stream to a locked audio video recording device suchas a DVCam VTR (ie a dv tape digitally transfered to DVCam). You can't. The recording device stops dead. Professional DV is a different animal. It has locked audio.

For NTSC, even "locked" audio can't permit frame accurate editing of two streams without unpacking and repacking it. The number of audio samples per frame is exactly specified for locked audio and it repeats on a 5 frame cycle. i.e., audio frames have to align to a 5 frame boundary.

For PAL, each audio frame contains the same number of audio samples and can be aligned to a single frame boundary.

Both locked and unlocked *must* have exactly the same average number of samples per frame.

Within an NLE, locked vs. unlocked is irrelevant because the audio is regenerated.

There's a completely false myth that unlocked audio leads to drift and is why some playback software exhibits out-of-sync audio. Unlocked simply means there is flexibility in how the required average samples per frame is achieved. Somehow, people believe that the audio slips out of sync from the beginning of the tape to the end....

[Shaun Roemich]

Sorry John, I should have clarified: I wasn't suggesting that the statement of drift was correct from the previous post, I was just pointing out that Locked and Unlocked are NOT the same. And I am talking about the actual "to tape" signal, not something that has been regenerated by one's NLE.

In short: If one was connected from a Sony VX1000 camera (11 year old miniDV) to a DSR40 DVCam deck via Firewire to make a copy TO the VTR, the transfer would fail as a result of the locked vs. unlocked audio issue. The transport on the deck would leave record mode when it found the first frame that didn't have the exact number of audio bits it was looking for.

The line blurs significantly when one considers what happens to the stream once the NLE captures it but the raw tape information is not UNIVERSALLY interchangeable.

[Shaun Roemich]

Audio slips out of sync over a tape CAPTURED into an NLE as a result of incorrect audio sample rate settings. Under Final Cut 1.2.5, if I captured 32k audio to a 48k project/sequence, when playing back from the timeline, audio would slip. I don't ever use 32k audio anymore so I have no idea if this is still an issue or not. This had nothing to do with locked or unlocked; this was a bit quantization issue.