Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
IEC 62447-3 — Helical-Scan Compressed Digital Video Cassette System Format D-12: Data Stream Format
Technical specification of the data stream format for DV-based compressed video and audio transmission over SDTI interface
IEC 62447-3:2007 defines the format of the data stream for synchronous exchange of DV-based compressed audio, data, and video over the Serial Data Transport Interface (SDTI). This standard is part of the IEC 62447 series covering the D-12 helical-scan compressed digital video cassette system using 6.35 mm magnetic tape, and focuses on the data stream format for 100 Mb/s, 270 Mb/s, and 360 Mb/s SDTI systems.
1. Standard Scope and System Context
IEC 62447-3 specifies the data stream format for transmitting DV-based compressed video and audio over SDTI per SMPTE 305M. It covers transmission of audio, subcode data, and compressed video packets for 525/60 (NTSC) and 625/50 (PAL) SDTI systems. Part 1 of the series covers VTR specifications, Part 2 covers encoding and data format for 1080i, 1080p, and 720p systems.
Space within SMPTE 305M not used by the DV data stream may be used for transmitting other data types, enabling hybrid systems that carry both compressed video and metadata or auxiliary data over the same SDTI link.
2. Stream Block Architecture
The fundamental building block of the D-12 data stream is the stream block, which is 170 words in length. Each stream block contains a secondary header, two DIF (Digital Interface) block IDs, two DIF block data areas, and an Error Correction Code (ECC) block.
| Component | Word Count | Description |
|---|---|---|
| Reserved data words | 3 | Default value: 200h |
| Signal type (ST) | 2 | Video frame ID, field/frame frequency |
| Transmission type (TT) | 1 | Transmission rate and structure flags |
| DIF block ID 1 | 3 | Identifies DIF block 1 |
| DIF block data 1 | 77 | Compressed video/audio data |
| DIF block ID 2 | 3 | Identifies DIF block 2 |
| DIF block data 2 | 77 | Compressed video/audio data |
| ECC | 4 | Error correction code |
| Total | 170 |
3. Signal Type and Transmission Type Words
The Signal Type (ST) words, consisting of two words (word 3 and word 4), encode critical information about the video stream:
- STVF ID (word 3, bits B7-B0): Identifies the specific type of video frame, including 3:2 pull-down converted progressive pictures from 480-line/29.98 fps sources.
- FF flag (word 4, bit B7): Indicates the field frequency — 0 for 60 Hz (59.94 Hz), 1 for 50 Hz.
- DIF structure (word 4, bits B6-B4): Defines the DIF structure format.
- TRF flag (word 4, bit B2): Indicates the transmission rate flag.
The Transmission Type (TT) word provides additional information about the transmission structure, including the data rate (100 Mb/s, 270 Mb/s, or 360 Mb/s) and the frame structure format.
4. Channel Unit Mapping
The standard defines channel unit mapping rules for three SDTI data rate structures: 100 Mb/s, 270 Mb/s, and 360 Mb/s. For the 100 Mb/s structure, separate mapping tables are provided for 525/60 and 625/50 systems. The mapping defines the start lines of channel units and the precise arrangement of DIF sequences within each channel unit.
Engineers implementing D-12 decoders must handle the different channel unit mapping for 270 Mb/s and 360 Mb/s systems separately. The mapping is not a simple scaling of the 100 Mb/s structure due to differences in the underlying clock rates and DIF sequence alignment.
5. Error Correction
The ECC block consists of 4 words providing Reed-Solomon error correction for the stream block. The ECC is calculated over the entire 166 words of the stream block (excluding the ECC words themselves). The block type word in the SDTI header distinguishes between streams with ECC (value 173h) and without ECC (value 233h).
6. Engineering Design Insights
- Legacy format considerations: The D-12 format was designed for professional broadcast applications. While largely superseded by file-based workflows, understanding the data stream format is valuable for archival access to legacy D-12 recordings.
- Interface compatibility: The standard’s reliance on SMPTE 305M (SDTI) means that D-12 streams can be carried over standard SDI infrastructure (270 Mb/s for SD, 1.485 Gb/s for HD), simplifying integration with existing broadcast plants.
The reserved data words (default 200h) and the parity protection scheme (even parity plus complement on all control words) provide robust error detection. The even parity bit and its complement on bit B9 enable single-bit error detection on all control words.
7. FAQs
Q: What tape width does the D-12 format use?
A: The D-12 format uses 6.35 mm magnetic tape in a helical-scan recording configuration, similar to the DV format but optimised for professional broadcast applications.
A: The D-12 format uses 6.35 mm magnetic tape in a helical-scan recording configuration, similar to the DV format but optimised for professional broadcast applications.
Q: Can D-12 streams be transmitted over IP networks?
A: The standard defines SDTI transport only. For IP transmission, the D-12 stream would need to be encapsulated in an appropriate IP transport protocol such as RTP, with appropriate timing and synchronisation mechanisms.
A: The standard defines SDTI transport only. For IP transmission, the D-12 stream would need to be encapsulated in an appropriate IP transport protocol such as RTP, with appropriate timing and synchronisation mechanisms.
Q: What is the relationship between IEC 62447-3 and SMPTE standards?
A: IEC 62447-3 references SMPTE 305M (SDTI) for the physical transport layer and SMPTE 370M for the DV-based data structure. The standard is complementary to these SMPTE standards, adding the specific D-12 stream block format definition.
A: IEC 62447-3 references SMPTE 305M (SDTI) for the physical transport layer and SMPTE 370M for the DV-based data structure. The standard is complementary to these SMPTE standards, adding the specific D-12 stream block format definition.
Q: Is ECC mandatory for D-12 data streams?
A: No. The SDTI header block type word distinguishes between ECC-protected (173h) and non-ECC (233h) streams. The standard supports both configurations to allow trade-offs between error robustness and bandwidth efficiency.
A: No. The SDTI header block type word distinguishes between ECC-protected (173h) and non-ECC (233h) streams. The standard supports both configurations to allow trade-offs between error robustness and bandwidth efficiency.
