Preferred Alignment for Dolby E in HD Systems

System designers, content creators and equipment manufacturers may have questions on how to eliminate the ambiguity of Dolby® E alignment measurements with reference to various video formats. To ensure error-free editing and switching of Dolby E signals, we have developed a recommended approach to alignment issues.

Dolby E streams include audio and metadata encoded into blocks of SMPTE 337M data (initially placed on a discrete AES/EBU carrier, or subsequently embedded into SDI). These data blocks are output at a rate associated with a chosen video frame rate, as the Dolby E bitstream is normally carried along with video.

To allow error-free cut-style editing and video switching, a small section of null data (or silent samples) exists around the video switch point (also referred to as the “guard band”). The guard-band location is designed to allow a Dolby E bitstream to be delayed, up to a point, as a safeguard against frame errors that may occur in a bitstream upon switching or editing. The measurement of where the guard band ends and the SMPTE 337M Dolby E data burst starts is often called the “line position.” This relates to a specific video line to which the start of the Dolby E data burst is aligned.

Placement of Dolby E Line Position

The best placement of the Dolby E line position is relatively simple to deduce when there is a limited number of video standards to which the signal must adhere. For example, if standard-definition video is the only format to be used, then determining the preferred line position is straightforward. The placement becomes more complicated when multiple formats come into play, or when there is cross-conversion taking place. For example, if the video format was originally standard definition, and was then upscaled to high definition, the line position measurement of Dolby E against the SD video format would not correctly reflect the Dolby E alignment against the HD video format.

Determining Dolby E Alignment

In such cases, the preferred method for determining the alignment of Dolby E with respect to video is to measure the time elapsed from the SMPTE RP168 reference point (the point in time indicating the start of the video frame). According to the SMPTE standard, this reference point is further defined as the point in time where line 1 for 720p 59.94, line 1 for 1080i 59.94, and line 4 for 480 29.97 fps (525 NTSC 29.97 fps) signals align, as well as the point in time where line 1 for 720p 50, line 1 for 1080i 50, and line 1 for 576 25 fps (625 PAL 25 fps) signals align.

In general terms, Dolby E content is in a safe position if the start of the SMPTE 337M data burst occurs after the switch point and the end of the SMPTE 337M data burst occurs before the start of the next video frame.

The following charts provide the recommended alignment locations for Dolby E frames with respect to SD video lines, HD video lines, and the RP168 reference point, as would normally be presented from a Dolby E encoder. This is indicated by the “ideal” positions. The closer to this alignment a signal is, the more likely it will continue to be error-free upon switches downstream.

Dolby E Line Position Table - NTSC / PAL
Dolby E Line Position Diagram - NTSC / PAL
Dolby E Line Position Table - all rates

There is an allowable skew to take account for production and head-end functions like embedding. For example, embedding can add a few lines of delay to the audio samples, thus skewing the Dolby E position. The limits of this allowable skew are indicated by the “potentially earliest” and “potentially latest” positions, beyond which the Dolby E bitstream will likely be corrupt upon switching or editing. It is possible to embed / de-embed several times, and still remain safe, before an ideally positioned Dolby E signal reaches the potentially latest position.

DM100 v displays 23.98, 24 and 30fps Dolby E line positions with a 5% error margin.