SMPTE Time Code Scenarios

[ LiB ]

Now that you have the technical background, let's examine potential project scenarios and see how the audio and time code would be handled. With the advent of HDTV, acquisition and delivery formats have expanded, creating many varied possibilities.

NTSC Video to NTSC Video

The simplest situation is a project shot on NTSC video that is destined to be seen only on broadcast TV or on VHS tapes and DVDs. In this scenario, the time code will run at 29.97 throughout the project. The choice to use drop-frame or not depends upon the requirements of the delivery format or the working practices of the video post facility.

When shooting video there are two choices for time code. The first is time-of-day code, which simply means that at the start of the day, the SMPTE generator is set for the time of the day and that it will run constantly until shooting is over, displaying the hours of the day. Time-of-Day should be drop-frame time code so that it would stay in sync with real time. The advantage here is that the time of day at which each take was recorded is conveniently stored on tape in the SMPTE time. For any given day, there is also a unique SMPTE time for every take filmed. This can help eliminate confusion during post-production.

NOTE

TIME CODE LIMITATIONS

Beware that Pro Tools has an issue with time code beyond 13:00:00;00 due to limitations in operating systems. There are workarounds for this bug that will be discussed later.

The second choice is called runtime SMPTE . In this situation, SMPTE runs only when the camera is running. When filming stops, so does the SMPTE. Typically, whomever changes the tapes in the video camera will also change the starting hour of the time code to reflect the tape number currently being used, making it easy to identify each tape just by looking at the time code's hour number. This technique is most helpful on shoots that last less than one day. If shooting runs into more days, the chances of creating a tape with the same hour number increases , leading to more possible confusion later.

During post-production, video is simply transferred into a video editor workstation without any time code or frame rate changes. Audio will remain at 48k and use the same time code as video. Layback and delivery are just as easy. The only choice is whether to use drop-frame or not. If the program material exceeds a minute in length and is destined for broadcast, the final master should use 29.97 drop-frame in order to keep the time code in sync with the clock on the wall, giving an accurate indication of how long the material runs.

OK, that was simple. Let's look at a more complicated process.

Film to NTSC Video

Many projects are shot on film for the look but edited on video for ease, then are delivered to the public on video, whether on television or on videotape and DVD. While the visual process is standard, there are a couple of different ways to handle the audio.

Film cameras typically run at 24fps when filming scenes in which the audio is also being recorded. When 24fps is used for film, audio time code is recorded at 30fps. Film cameras are capable of filming at 30fps and 29.97fps while audio is being recorded as well. In these situations, audio time code will be 30 and 29.97, respectively. Usually, 24fps is the frame rate for the film camera and 30fps is the rate for the audio deck.

Why have two frame rates (24 and 30) running at the same time? Because once the film has been slowed down during the 2-3 pull-down process, as in Figure 3.5, the audio will be slowed by the same amount, resulting in the NTSC standard rate of 29.97fps. Here's where the choice comes in. In order to slow the film down, the audio must slow down as well. Otherwise, the audio will not be in sync with the transferred video. The only way to do that with a digital recording is to slow the sample rate down. If 48k was used while filming on the set, it will have to be slowed down to 47.952 during the film transfer in order to remain in sync with the transferred film on video. This results in an analog transfer or D/A-A/D conversion to the videotape copy. The other choice is to record using a "pulled-up" sample rate of 48.048k while filming, so that during transfer the sample rate will slow down to 48k, making a digital transfer of the audio possible. In either case, the videotape of the film transfer will have audio that is in sync with the film footage, uses 29.97fps SMPTE, and has a 48kHz sample rate with digital video formats.

The rest of this scenario plays out just like the video-to-video version, in which the only choice is whether or not to use drop-frame SMPTE during post-production and final delivery.

Film to Film

When a project starts on film and ends on film (such as a feature film in theaters) the choices for time code increase. Since the final format will be film running at 24fps, there is no need to record audio during the shoot at 48.048k (called 48k pull-up), as eventually the 2-3 pull-down process will have to be reversed back to 24fps for the final film release. Reversing the pull-down would result in a non-standard sample rate (48.048k) for the final mix. Typically, film will be transferred to video for editing, effects, and computer manipulation. Then, once the film is in final form, it will be transferred back to a film print for presentation in theaters.

There are two ways to deal with the pull-down issue. First, you could follow the example in the previous section of a film-to-video project including the 2-3 pull-down process during transfer to video. In this case, the process would include a final step to reverse the transfer process when going back to a film master. This final step involves "pulling-up" the sample rate of the final mix in order to match the original speed of the film. In this scenario, another sample rate conversion must take place in order to create a 48k digital master. If the production audio was recorded at 48k and transferred to video using a 2-3 pull-down process, then large portions of audio have undergone two digital to analog to digital conversions by the time the final mix is done. This is not a preferred method, although it will work and can be of high quality.

The second and preferred method, although it might be monetarily out of reach for lower-budget productions , is to transfer production audio directly into Pro Tools and then pull down the operating sample rate of Pro Tools during post-production in order to match the timing of the reference videotapes. As the film editor will most likely be editing on video, the film will be slowed down due to the 2-3 pull-down process. You can pull down Pro Tools while syncing with edited videos of the film. When it comes time to print the final mix, Pro Tools can be run at the normal sample rate of 48k and will be in sync with the original speed of the film. This way, no additional D/A-A/D conversion is necessary during the final mix, allowing the full quality to be maintained . It is a little more confusing to work this way, but the results will be superior due to the fact that no analog transfers were made.

NOTE

IMPORTING MUSIC WHILE IN PULL-DOWN MODE

Be careful when importing score music or any other audio that has been created in sync using videotape transfers of film material. If the composer or other audio editor created this material without using the pull-down mode, you will have to convert the sample rate or do an analog transfer to match your pull-down condition. Also, incidental music used by the director during the video edit will have to be conformed to pull-down in order to maintain sync in the final mix.

24p HD to Video

Starting a project in 24p HD video and finishing in regular NTSC video is very similar to a film-to-video project, with some exceptions. When using 24p HD as a source for NTSC video, the video will always be shot at the pull-down frame rate of 23.976fps. This allows an immediate transfer to NTSC video without any special equipment. Audio can be recorded either directly to the HD camera at 48k or to an external DAT machine at 48k using 29.97 SMPTE, as the video frame rate is already pulled down. Audio can be transferred directly to video without any speed changes. This means that you can capture production audio digitally into Pro Tools without any fuss.

Once all of the audio is in Pro Tools, you can proceed exactly as if you were working on a film-to-video project working at 48k and laying back directly to a VTR using 29.97 SMPTE. One of the great advantages of certain HD video cameras is the ability to resolve 24fps to other SMPTE frame rates. For instance, you can play back 24p video at 23.976fps and still output 29.97 SMPTE. This is simply not possible with film, as the telecine process is first required to resolve to 29.97.

24p HD to Film (or 24p HD)

Many movies these days are being shot on HD and then projected from film in theaters. In this scenario, 24p HD would be the original format and film would be the master format. For audio, the time code scenario is similar to a straight film-to-film project. You would start by using 24fps on the camera. If audio were recorded to another deck besides the camera, then it would be at 48k using 30fps SMPTE time code. The audio would have to be pulled down in order to maintain sync with transferred footage onto video. You could complete post-production in Pro Tools using pull-down mode. When the final mix was completed, returning to 48k would bring the audio back up to the original 24p speed.

Without using pull-down sample rates during editing, each transfer from 24p to video and back again would require either a sample rate conversion or analog transfer for the audio. This is the less desirable path , of course, but it does allow users without the extra hardware (Sync I/O, USD, and so on) to work with 24p and film sources. The reality of budgets must be taken into consideration, and Pro Tools has the flexibility to work under many constraints.

Film to 24p HD

I think we will be seeing more of this type of project. As film captures the light so well, many moviemakers will still find film to be the way to shoot. HD affords the ability to endlessly edit and create all sorts of effects and CGI (computer generated images), all while remaining in a high-quality digital medium. It only makes sense that many projects will begin on film and end up on HD in their final form.

HD TV is currently broadcast in one of three formats. The most popular is called 1080 60i (meaning 1080 horizontal lines of resolution with 60 fields per second, interlaced) and is used by NBC, CBS, and PBS. Another format, used by FOX and ABC, is 720 30p (720 horizontal lines at 29.97fps progressive scan). HBO HD uses 1080 24p, which uses less data than 1080 60i and can be easily transmitted via satellite. HD receivers perform a 2-3 pull-down of the 24p signal to create a 29.97fps interlaced image for your TV. All of these formats retain the original film speed and will not require sample rate pull- downs or pull-ups. Time code will either be 30fps or 24fps depending on the final format.

HDTV is a data specification. The video frame rate and audio sample rate can be changed. That is a great part of the system. Different resolutions are supported within HD TV. These broadcast formats can change and will do so at some point in time. Keeping up with the technology will help you stay ahead in the industry.

As you can see, there are a lot of options for handling time code and sample rates. You must have a concrete idea of where the project will end up in order to create the highest quality mix at the proper speed and sample rate for any of the media discussed in the above sections.

NOTE

HD TV STANDARDS

Since the HD TV broadcast is a data stream, the format of the video in that stream can be in a variety of frame rates and resolutions. Networks may change their standards at any time, so any information provided here about network broadcast formats could be different by the time you read this book. Be sure to have the most current information before deciding on HD formats.

Audio for HD could end up at 24fps, 23.976fps, 29.97fps, or even 25fps in the case of PAL. Your mixes should be at 48kHz relative to these SMPTE formats. Basically, any of the above scenarios could apply to a project ending on HD.

SMPTE time code can be a confusing subject, but as you gain experience working with different video formats, the common practices will become apparent. Now that you have a basic understanding of how SMPTE and synchronization works, let's get Pro Tools set up to work with video.

[ LiB ]