parino banner
Home - Project Archive - In production - Resource

Digital video DV, DVCam, HD XDCAM, DVCPro...
So what are all those little letters about? The format shift over the last year or two has been towards HD, Sky broadcast some channels using this and new TV's are going HD, or Hi-def. What does this mean at source for capturing images?

DV, DVCAM & DVCPRO

use the same video codec and, so, offer the same picture quality. The big difference is that the tape runs faster through the camera in DVCAM, which makes it more robust in the event of dropout and less prone to compatibility problems. DV and DVCAM record standard definition pictures (in 4:3 and 16:9).
DVCAM is Sony's professional version of DV and DVCPRO is Panasonic's. DV, DVCAM and DVCPRO all use the same video and audio codec.
DVCPRO comes in three flavours:
DVCPRO 25 (at 25Mbps, 4:2:0),
DVCPRO 50 (at 50Mbps 4:2:2), and is considered by many to be as good as Digital Betacam
DVCPRO HD (at 100Mbps, 4:2:2)

HDV
is a new recording format developed by a consortium of manufacturers including Sony, JVC, Canon and Sharp. HDV is a high-definition version of DV, which can record in either 720 lines (HDV1), at 19Mbps, or 1080 lines (HDV2), at 25Mbps. HDV gives higher resolution pictures but still records them on a miniDV or DVCAM tape.
DV, DVCAM and DVCPRO (25, 50 and HD) record whole frames onto the tape, with each frame being compressed to get all that information onto the miniDV tape. The amount of data created in HDV is far greater than in DV (or DVCAM) but it still has to fit on the same little tape. So, the engineers had to find a way to more efficiently compress all that information. Step forward MPEG2 long GoP encoding.
With this compression, the camera does not record every frame of video. It records occasional key frames and just enough other information to enable it to recreate the rest. The video between one key frame and the next is called a Group of Pictures (or GoP). A Group of Pictures consists of three types of frames - the I-frame, P-frame and B-frame. (Intra frame, predictive frame and backwardly predictive frame.)
A typical GOP consisting of 12 frames will be: I B B P B B P B B P B B
The I-frame holds the most information and is the largest frame size in terms of data.The P-frame is compressed further and holds less information than the I-frame. It predicts what will make up the image from the previous I-frame or P-frame. Then comes the B-frame, which holds the least amount of information. It is calculated by forward and backward prediction to the I-frames and P-frames or even an estimate of the in between value of them.
In DV or DVCAM, a small amount of dirt on the heads when recording could result in dropout problems on a single frame only, because each frame is independent of the other. With MPEG2 a small bit of dirt causing dropout on an I-frame could affect half a second of video, ie the whole GoP.
With MPEG2 only the I- frames hold all the picture information – but when editing you want to edit between any frame. Final Cut Pro 5 edits native HDV - but requires a lot of processing power to do it, as it has to look at all of the GoP at once, and calculate a complete frame for each of them, so that the instant you edit, it can offer you an accurate full frame picture.

SDI stands for Serial Digital Interface (also known as CCIR 601). It is the broadcast standard for interfacing uncompressed video, from cameras, VTRs, etc. HD SDI is the HD version of SDI, and offers a throughput of almost 1.5Gbps.

4:2:0 and 4:2:2 or Y:Cb:Cr?
The first number (Y) represents brightness or luminance (or what engineers call luma), The next number (Cb) is the blue signal minus luma (B-Y) and the third number (Cr) is red minus luma (R-Y). The numbers show how much colour information is left in the recorded pictures.
4:2:2 sampling is used for DVCPRO50, DVCPRO HD (and Digital Betacam)
4:1:1 sampling is used in DVCPRO25 and DVCAM (NTSC)
4:2:0 sampling is used in DV and DVCAM (PAL)
The zero doesn't mean there is no colour info - but it does explain why DV is difficult to use for green/blue screen work - there just isn't enough colour info to work with.

Home - Project Archive - In production - Resource

+353 (0)86 0577061 : tony@parino.net | gongfu.ie | smilebrightly.eu