A PRIMER ON DIGITAL BETACAM AND RELATED TECHNOLOGIES
June 1994

What is Digital Betacam? Digital Betacam is a video tape format that is mechanically based on the Beta tape transport. The recording on a Digital Betacam is component digital video with four channels of digital audio. Digital Betacam has been available in Europe as a 625 line (PAL) machine since early 1993 and became available as a 525 line (NTSC) machine later that year. The first machines delivered in the world were to Henninger Video in Arlington, VA on 18 Mar 93.

Where can Digital Betacam be used? The best use of Digital Betacam is for long-form video mastering that is currently targeted for D1 mastering. It is also a good tape format for studio recording of camera input material, graphics material and telecine (film-to-tape). Field recorders were shown at the 1994 NAB Convention and became available in the Fall of that year.

What is the difference between D1 and Digital Betacam? The first D1 tape transport was a specialized high precision computer data recorder that had been in use since 1982. The medium is cassette loaded 19 mm wide (3/4") oxide tape. In 1984 an electronics cabinet was introduced that adapted the same data recorder to record and play digital component video and four digital audio channels.

Advances in digital data recording allowed recorders to use half inch metal tape and less expensive Beta tape transports. The recordings on Digital Betacam use a system called DCT (Discreet Cosine Transform) BRR (Bit Rate Reduction). Redundant and unnecessary data is eliminated from the data stream recorded to tape, thus saving space on the tape. On playback, the data is reconstructed using complex mathematical algorithms.

The D1 tape format requires large amounts of tape stock to record the continuous data stream derived from digital video and audio. The Digital Betacam economizes on the amount of tape with a less expensive transport to achieve essentially the same results.

Since the Digital Betacam uses a similar transport to the Betacam SP format, it is possible to play back analog Betacam SP tapes on a Digital Betacam machine. With over 200,000 analog Betacam machines in the world making recordings, the advantage of a Digital Betacam machine in an existing facility becomes fairly obvious.

What circumstances warrant recording to D1 or Digital Betacam? The specific exercise in choosing a component recording format over a composite (NTSC) format, such as D2 or 1 inch, is to avoid the shortcomings of an NTSC recording. These shortcomings may be visible as defects in an otherwise perfect picture.

Composite NTSC video was designed as a distribution method for all broadcast and cable television transmissions. It is not perfect for effects intensive post production functions but it has been and still is quite acceptable for most applications. Due to the relative simplicity of composite equipment, it has generally been more cost effective than similar component equipment. Composite equipment has also developed faster than the same type component equipment because of its simplicity. Component digital recorders have recently gained feature sets that had been commonplace on composite digital recorders for the previous five years. Affordable digital post production systems, switchers, machines and peripherals, arrived as composite systems well before affordable component digital systems were available.

If money is less of an object than picture quality, then component digital systems are generally more desirable than composite digital in some applications. Now that the cost of composite and component digital systems are coming in line with each other, the choice to use component digital systems becomes easier to make.

The areas where component recording have an advantage over composite are:
· Effects and Graphics Compositing: Graphic or live images that are painted, moved, keyed or otherwise manipulated are best done in a digital component environment. Digital component is the native signal format for digital effects devices and most paint systems. Any full bandwidth digital system provides the operator with a transparent signal path that will survive unlimited multi generation work. Component digital systems allow free exchange of images between paint, effects and editing devices without loss of quality or regard to NTSC signal anomalies.
· Standards Conversion: The recording on a D1 or Digital Betacam tape is digital component which is the native video signal format of standards conversion equipment. It presents the least adulterated signal to the converter for the best conversion quality.
· Color Correction: The native video signal in color correctors is component video. D1 and Digital Betacam record and play back component signals and present the color corrector with the best signal for correction.
· Projection: Video projectors that produce large screen images are best served by component a video source. Large screen images would make the inherent shortcomings of an NTSC source very noticeable.
· Film recording: Video programs targeted for film recording are best done on component recorders. Film recorders are presented with an RGB (Red Green and Blue) image that would be degraded if they had been composite at some point along the way. The film image is still limited by the resolution ability of the video signal.
· Computer Images: Video intended for display on computer equipment is best presented in component form since the native internal format of the computer is RGB.

How do you select a tape format or video standard for editing? The decision on which format or standard to use should depend on what you are doing from all aspects of the job. What are the sources? What is the end product? What is the content? What are the technical difficulties introduced by using one system or another? What are the costs involved? Will there be a reasonable difference in quality based on the cost and production expectation? These questions are both technical and production value oriented. The two become linked when the technical process affects the production value or quality of the product.

Any original program, where all production elements can be generated as digital component, will gain the most from a digital component post production environment. Without examining the range of options, however, using component digital equipment could nearly double the cost of post production over other methods without improving the end product. Part of the consideration of selecting a system must be whether there will be a difference in the end product compared with the costs associated with the process. There is no single answer on tape formats or video standards for all production or post production needs. There must be an appropriate application of technology in every step of the production path.

To provide a pure component digital operating environment for all production and post production has been prohibitively expensive until the introduction of Digital Betacam and other affordable infrastructure components. It could always be done but was met with questions of "can I afford it?" and "do I need it?". Component digital systems were reserved for short duration, high intensity graphics production where the quality level was absolutely essential. In contrast, the strengths of component digital are barely tapped when used for long-form editing and can be classed as something of a waste, if cost is an issue, without other justifications. In some cases, a component system will be a detriment to the product quality.

Assuming an NTSC final product, it is ironic that the component environment will damage all NTSC source material, such as D2 or 1 inch. On the other hand, the composite environment will accept encoded component sources without causing additional harm to the signal quality. The specific technical exercise in any post production environment is to get the best looking NTSC picture possible.

It is important to note that most of the finished edited master tapes that exist are composite masters for distribution to broadcast operators or other post production facilities, even though the elements may have been component sources. The edited masters themselves are never distributed. It is more likely that a composite duplicate will exist in the distribution pipeline and will become the source used for subsequent program modifications. It is unlikely that the elements will be used, or even be available, to rebuild an entire program to perform program modifications. It is technically most appropriate to stay in a composite environment for these functions.

Over a period of time, component digital masters and duplicates will become more commonplace, even in broadcast environments. Component digital systems and infrastructures will proliferate now that costs for systems are more closely aligned with composite counterparts. Digital Betacam and it's related systems is the first component digital format that can challenge the composite digital environment for cost and functionality. Composite video systems and edited masters will continue to be a factor until the day you are able to go home and turn on your component digital home television receiver.

Is there an advantage to assembling an NTSC product with component video?
A case can be made for either a "yes" or a "no" answer. In a practical sense, there will be absolutely no quality difference between an image that is encoded and edited on digital composite equipment or the same image edited on digital component equipment with encoding saved for the final step. The differences will be in the cost of either method and the flexibilities available.

In graphics compositing, the native format of systems that generate and manipulate images is component digital. The nature of graphics compositing is to modify and refine the graphics gestures and be able to pass the image from one system to another many times without degradation. This requires digital component systems until the graphics elements are finalized. You have the choice to encode the elements and perform a long-form editing process on less expensive digital composite equipment instead of digital component. Once encoded, you will lose the flexibility of transparently returning to the graphics environment for corrections or modifications. If there is a desire to keep that option available, there will be additional costs in staying in a digital component environment through the editing process. The economies of digital composite in this case are reserved for the more organized client.

In long-form editing, there is usually a straight line between production elements and NTSC product. A composite or a component environment will work exactly the same for cuts, wipes, dissolves and titles typical of a long-form project. If there are more steps in processing that will benefit from a component process, like subsequent color correction or standards conversion, a component digital product will have an incremental advantage over a composite one. The inclusion of digital effects in the editing process will be incrementally better if the effects device uses a component source. The output of the effects device will be identical if it were handled as component or composite. Other functions such as off tape chroma keys are typically handled by component sources and keyers. It is common to find component chroma keyers nested inside composite switchers.

The composite NTSC signal should be monitored in any component environment where work is being done since it is possible to make a good component picture that does not look good in composite. This is especially true of graphics elements. If the eventual distribution of the end product is NTSC, all NTSC anomalies will be present whether the program was edited in composite or component.

How are the composite and component devices linked together? The term "composite" describes things that are mixed together which, in this case, is a monochrome picture with the associated color information mixed with it. Composite video is the same signal that is used by all home television receivers. The term "component" describes elements that are held separate which, in the case of component video, the monochrome picture is not mixed with its color information.

To use a component video source in a composite environment, the separate elements have to be encoded through a device called an "encoder". The monochrome video image is mixed with the color information to form a "composite" of the two. The elements of a composite video signal are known as the "luminance" for the black and white image and the "chrominance" for the color information. This encoding gives us NTSC or PAL composite video.

To use a composite device in a component environment, the composite (encoded) video must be decoded into separate elements through a "decoder". The chrominance is stripped from the luminance signal. The chrominance is further processed into separate color signals used by the target equipment.

What do encoders and decoders do to the television image? Color television was adapted to existing black and white broadcast systems by encoding color information into the monochrome video information. Black and white receivers pass only the base monochrome picture to the screen and do nothing with the color information. Color receivers detect the color information and decode it for application to the black and white image as an overlay.

For television broadcast, color information has to be applied to the black and white video in such a way that the entire signal fits into the existing broadcast television channel. As part of this requirement, the resolution of the color information must be limited and interleaved with the black and white image. This encoding process can be done with relatively simple hardware to achieve good results.

The home color television receiver contains a decoder that disassembles the encoded video into red green and blue video signals. They are applied to the red, green and blue inputs on the television's display device. The decoding process is difficult to do well with inexpensive equipment. Extracting the chrominance information without damaging the luminance requires a precision decoder not found in home television receivers.

For any given encoded (composite) video signal, a variety of quality levels can be had depending on the precision and complexity of the decoder. Simple decoders will blur the luminance as well as introduce undesirable artifacts during the decoding process. High quality decoders will pass the luminance signal almost unharmed as well as suppress undesirable artifacts. The quality of the encoder is also a factor but not as large as the quality of the decoder.

How often can a signal be encoded and decoded? If the program material is destined for a composite distribution system (broadcast, laser disk, etc.) it should only be encoded once. Decoding will incrementally degrade the video even if it goes through a high quality decoder. Decoding a signal should only be done if necessary such as providing a graphics workstation (component) with video from a D2 or 1 inch tape machine. Decoding should never be considered a transparent function. Once a signal is encoded to a composite environment, it should stay there for any subsequent processing.

The point at which encode/decode cycles become damaging to the picture depends on the quality of the encoders and decoders and the performance expectation of the entire process. Encoding and decoding once will damage the picture but depending on the picture content it may be subjectively undetectable. Doing it several times will accumulate undesirable artifacts to the point where the picture becomes objectionalble.

Where are encoders and decoders located? Any component source that will be used in a composite environment must have an encoder. Component video devices intended for use in composite systems have encoders built in. Such devices include Betacam SP and Digital Betacam recorders, character generators, video cameras, digital effects devices and standards converters. Of these same devices that accept a composite video input, some may have built in decoders such as Betacam SP machines.

A component recorder, such as Betacam SP machine, used for editing or duplication in a composite environment requires that the video signal be decoded at the machine's input then encoded again on the machine's output. The damage to the signal by encoding and decoding is added to the degradations caused by recording on tape. Even if the machine is a digital recorder, the presence of an encoder and decoder will destroy the normally transparent digital signal path and introduce degradations.

Devices that are not intended for use in composite environments, such as D1 recorders, will not have built in encoders or decoders. They must rely on outboard encoders and decoders to connect to composite environments.

When would editing to D1 or Digital Betacam not be advantageous? There are cases where editing in a composite signal format are more appropriate than component systems.
· Broadcast Masters: Broadcast video programs are sent over the airwaves and through cable systems as NTSC composite video. Editing to a component format, such as D1 or Digital Betacam, will require the program to be encoded somewhere before broadcast. If it is broadcast directly, it is best to supply a composite (D2 or 1 inch) edited master to the broadcast system where any encoding, and subsequently the picture quality, can be controlled by the post production company. Most broadcast video is long-form video that contain few or no special effects that would benefit from an all component post production chain. The client may be served better if a D2 master is created for a broadcast product since the composite picture, particularly digital composite, will not degrade through multiple stages of editing. An excellent composite air master can be made from a component digital edited master at additional expense (duplication time and tape stock), but if no further processing like extensive graphics or color correction is involved after a master is completed, editing a component master holds no advantage in this application. D1 tape masters will cost the client more to generate than an equivalent D2 master with no broadcastable advantage. No broadcasters are known to air directly from D1. Only broadcasters that air directly from Digital Betacam should be considered for native Digital Betacam editing and assembly. Broadcastable video programming comprises the bulk of our video output.
· Composite Sources: If the majority of source material is composite, such as 1 inch or D2, the mastering format and editing environment should be purely composite. If the editing system were component, the composite signal would have to be decoded before the component system could accept the signal This decoding will degrade the video signal. The type of decoder used will determine the amount of degradation, roughly correlating to the price of the decoder. It is not appropriate to do an edit with primarily 1 inch or D2 sources as a component edit. If the finished product needs to be component because of delivery requirements, the client will be better served by an all composite edit with a component copy made at the end through a high quality decoder.
· Video Disk Mastering: Standard video disks, or laser disks, are a composite video format with the same performance expectations as a composite broadcast master. Using the same rules as applied to Broadcast Masters, a master tape intended for video disk distribution can be edited to a composite tape format without disadvantage over a component tape format.

What are some gray areas between component and composite mastering? The type of source material, the amount of graphics compositing, the complexity of the product, the eventual destination of the product and the budget constraints on the project will all determine how the final master will be assembled.

Keeping in mind the circumstances discussed earlier, pieces and parts of the program material can be handled in several different ways before final assembly. For instance, complex graphics can be composited (a term not connected with composite NTSC video) in a component environment and be included in a composite edit where final assembly takes place. Alternately, simpler graphics can be composited (that is, layered) in a composite environment, particularly composite digital, along with the main program material. Degradation will typically be, at worst, minimal or undetectable if the output product is intended to be composite.

As the weight of the job becomes more graphics and effects intensive, the sources become more component oriented, subsequent processing becomes a factor and the distribution medium are less in line with NTSC broadcast requirements, the product can be considered as making more sense being done in a component environment.

Why should Digital Betacam be considered instead of D1 for mastering? There are many factors that speak to the functional and practical advantages of Digital Betacam over D1:
· Cost Of Operation: The chief difference between a Digital Betacam and a D1 machine is cost. A typical Digital Betacam machine is around $55,000. A similarly outfitted D1 machine would cost about $150,000 and still not do what a Digital Betacam will do. This translates to savings for the customers over D1.
· Pre-read: None of the D1 video tape machinery currently available will do pre-read editing. Digital Betacam, in a digital component environment, will do pre-read functions as easily and transparently as the D2 format can do in a composite digital environment. This function will speed the edit by making quick digital compositing possible with one recorder and will also reduce the necessity of driving additional source machines or disk recorders in an edit to do dissolves, an additional savings for the client in terms of tape stock and machine time.
· 10 Bit Video Signal: The resolution of the D1 video signal is only 8 bits while the Digital Betacam is 10 bits. This translates to the number of possible brightness values available (about 250 for D1 and over 1000 for Digital Betacam) which will eliminate problems associated with subtle brightness changes in the picture. Footnote: inserting an 8 bit device anywhere in the signal path will permanently convert the video signal to an 8 bit signal. It is important to have 10 bit companion devices (switchers, etc.) in order to gain maximum advantage of Digital Betacam.
· 20 Bit Audio Signal: The audio resolution of a D1 machine is only 16 bits while the Digital Betacam is 20 bits. This translates to reduced distortion at low audio levels when using like minded outboard digital audio equipment.
· Stunt Modes: Digital Betacam has improved slow motion performance over D1. It will go from 3x normal forward, down through stop to full speed reverse. D1 is now only capable of 2x normal forward through stop to full speed reverse on the Sony DVR-2100. Most other D1 machines are capable of very limited slow motion. The Sony DVR-2000 can vary plus or minus 25% or display a still image. The Sony DVR-1000 can vary plus or minus 15% or display a still image. Other non-Sony D1 machines, such as BTS, have similar performance restrictions.
· Video Level Adjustments: Digital Betacam allows for playback video level adjustments exactly like any other Betacam SP, D2 or 1 inch machine. Adjustments of this sort are not available on D1 equipment.
· Tape Transport Speed: Shuttle times are faster for Digital Betacam which speeds the edit process and makes everything seem to work better. One reason is that there is less tape to pull through the transport than D1 for each program minute.
· Tape Durability: Experience shows that D1 tape stock is not tough enough for many passes through a tape machine. After 12 passes, significant errors accumulate on the tape stock. After 20 or so passes, the errors have been known to be unrecoverable. The Digital Betacam format has been tested to beyond 1000 passes before we gave up on the test.
· Tape Stock Cost: The cost of a 90 minute D1 tape is $288.00 and a 90 minute Digital Betacam is $127.00. Add another 30% for edit stock.
· Tape Run Times: The maximum running time of a D1 tape is 94 minutes. Maximum run time on Digital Betacam is two hours,
· Tape Storage and Shipping: A 90 minute D1 tape weighs seven pounds and is the size of a small briefcase. A two hour Digital Betacam is the size of a large Betacam SP cassette. We have seen occasional major problems during shipping of large shell D1 and D2 tapes due to the size of the tape spool inside.

What is "Pre-Read"? Sony and Ampex both introduced the D2 Digital video tape format to the professional video market in 1988. The Sony machine had an additional feature called "Pre-Read". This function has since been retrofitted to Ampex D2 machines and is also featured on Digital Betacam machines.

Normally, a machine used for mastering will erase the video and audio it is replacing during an edit, never to be seen again. While mastering to a D2 or Digital Betacam machine, the function of pre-read is to play back video and audio that exists on the master tape while the machine is recording new video or audio in an insert edit mode. An extra head in the machine that precedes the recording head recovers video and audio from the master tape just before re-recording. This is the pre-read head.

Functionally, a recording that exists on the master tape can be included as part of the video for subsequent edits on the same part of the tape. A dissolve between two pieces of video can be performed with pre-read by recording on the master slightly beyond the dissolve point with the first piece, then using the same video that now exists on the master as the source to dissolve to the second piece. Multiple continuous dissolve sequences can be performed using only one source machine and a master recorder with pre-read. Extending this, pre-read can be used very successfully for layering applications in the edit suite. If digital switchers are used, there is absolutely no generation loss or quality change as layers accumulate on the master. Layers may include any effect available in the online suite such as digital effects to place multiple images on the screen with only one source machine, or a character generator for adding a title over finished video without running any sources at all. Also, audio may be mixed from any of the four digital tracks and simultaneously re-recorded on the same tracks using the same pre-read conventions as video.

Layering with pre-read is not as forgiving as using dedicated graphics compositing systems with disk recorders. Once a recording is made on the master tape, the pre-read playback will reflect exactly what is on the master without the opportunity to go back to the previous layer. This is true for audio as well as video. It is common practice to incrementally archive the master tape onto another digital recorder if many passes invite the opportunity for error. In our experience, layering in this manner is still economical, quick and is used regularly in the online suite as an available tool for the editor.

No analog video formats, such as One Inch and Betacam SP, offer pre-read. The D1 and DCT digital component video formats do not offer pre-read. Only D2 and D3 in composite digital systems and Digital Betacam and D5 in component digital systems offer pre-read. When using Digital Betacam in a multiple generation effects sequence, it is important to use a 10 bit video switcher and signal path. This will maintain image quality to beyond 100 generations in pre-read. Using an 8 bit device will degrade the image after only a few generations.

Where should Digital Betacam be avoided in favor of D1? Sony does not recommend the use of Digital Betacam in critical graphics compositing. There is a very small degradation when a picture played back from a Digital Betacam is repositioned slightly with a digital effects device and re-recorded. The degradation would only be noticeable after several iterations through the digital effects device. In a practical sense, this degradation is minimal compared to what the digital effects device alone does to any picture offered to it regardless of the tape format. For layering in an edit suite, the background video typically will not go through the digital effects device more than once.

The compression scheme in Digital Betacam can be overrun if the picture information is too complex to compress and record. If more than half the screen is filled with high amplitude random noise (snow) which is a highly complex video pattern, any picture elements included in the same picture will show errors in the compression scheme. The errors appear as blocks that are 8 pixels wide by 8 lines tall with reduced color resolution inside each block. In a practical sense, random noise is not a typical video source and is an extreme example of how the compression scheme can be broken. Also, the compression errors are typically so mild that it would be missed in standard program material by the casual viewer.

The Digital Betacam picture will degrade faster than a D1 recording with repeated passes through external processing. The external signal path should be 10 bit instead of 8 bit. Some of the data that Digital Betacam uses to make an image survive multiple generations will be lost if the full 10 bit path is not used. It has been tested to beyond 100 generations in a 10 bit path without signs of degradation which goes counter to Sony's claim.

One manufacturer that insists upon 8 bit signal processing is Quantel, makers of high-end nonlinear editing and compositing systems. They use their patented "Dynamic Rounding" which essentially "dithers" an 8 bit picture to subjectively look like a 10 bit picture. The dithering process actually adds a little noise to the picture that is not detectable by humans but will electronically degrade a picture originated from and recorded to a Digital Betacam set for 10 bit operation. The Digital Betacam has a feature that allows the output to be set to 8 bit and allows successful 8bit quality multigeneration recordings with Quantel equipment.

Digital Betacam currently is available as a 525 machine (NTSC) or a 625 machine (PAL). They are not available as a standards agile machines (switchable between 525 and 625). D1 machines are standards agile. They can record or play either a 525 tape or a 625 tape. The D1 machine will not act as a standards converter. The machine must be set up to record or play the correct television standard.

Note: The numbers 525 and 625 describe the number of television lines on the screen for the respective television standards. It is a misnomer to call a digital component suite either PAL or NTSC since these are composite encoding schemes and component environments are not encoded. A 625 product could become either PAL or SECAM and a 525 product could become PAL-M or NTSC after encoding.

Technically, where can Digital Betacam be used? The Digital Betacam format can and should be used in an environment that D1 machines can connect to which is digital component. Sometimes an edit suite or a digital component production switcher is improperly called a D1 edit suite or switcher. D1 is a tape format, not a name that describes the type of video signal that is being handled. The analog compatible Digital Betacam machines can physically replace Betacam SP source machines in a composite or component editing suite.

Operationally, where should Digital Betacam not be used? Digital Betacam mastering should be avoided if the client expects to take the master and play it on Betacam SP equipment. A duplicate of a Digital Betacam master can be made on Betacam SP for this purpose, but the extra step and expense must be planned for. Clients will be able to carry the Digital Betacam tape to other facilities when the format becomes an exchange standard.

Technically, where should Digital Betacam not be used? As is the case with all component recording equipment, Digital Betacam does not connect transparently to composite analog or composite digital environments. As an option, an internal decoder may be purchased for the Digital Betacam to allow it to accept a composite analog video input and be used as a recorder. The internal decoder is designed to convert the composite signal to a component digital signal with modest performance expectations compared to some external decoders available.

Graphics compositing that requires multiple generation performance beyond 150 generations may be better served by component digital disk recorders. Using Digital Betacam during effects compositing in an 8 bit video switcher will degrade the image much sooner than using a 10 bit video switcher.

Can a Digital Betacam tape be played on any Betacam machine? No. The cassette shell of the Digital Betacam is compatible with Betacam SP machines but the digital recording will not play back on an analog Betacam SP machine.

Can Betacam SP tapes be played on a Digital Betacam machine? Yes, if the Digital Betacam machine is analog compatible. The model number of the Digital Betacam will have the letter "A" in it if it is the analog compatible variety, such as "DVW-A500". The "DVW-500" is the same machine without analog playback capability.

What is the advantage of Digital Betacam over Betacam SP? The advantages of digital recording for the video portion are immediately apparent with the absence of noise, smearing and loss of resolution. The four digital audio tracks on Digital Betacam may be edited individually which is not the case with Betacam SP. Channels 3 and 4 (AFM) channels on Betacam SP are not editable without destroying the existing video recording. Similarly, replacing the video recording on Betacam SP will destroy the AFM recording on channels 3 and 4. The availability of pre-read editing on Digital Betacam is also a major advantage over Betacam SP. The Digital Betacam's digital audio tracks are much higher quality than the Betacam SP audio tracks, especially when Digital Betacam is connected to digital audio sources and mixing equipment. There are no audio noise reduction processes involved with Digital Betacam as there are in Betacam SP except for decoding the Dolby-C audio on Betacam SP playbacks.

Describe the recording on a Digital Betacam tape. The video recording is digital component. There are four digital audio channels, each of them individually editable, similar to D1 and D2 machines. The recording has longitudinal (LTC) and vertical interval (VITC) time code. The LTC can be restriped. There is also a cue audio track similar to D1 and D2.

Describe another machine that claims to be "Digital Betacam". Sony builds a Betacam SP machine identical to the BVW-60/65/70/75 series that has the letter "D" in it's name, such as the BVW-75D. The machine has connections on the rear panel that target the device for installation into an all digital environment. The goal is to allow standard Betacam SP recording and playback to be done in systems that have digital electrical connections as the native signal path. Audio and video inputs and outputs on the machine are digital connections into a purely analog Betacam SP recorder. The recording and playback are incrementally better than the standard analog connected counterparts because some of the internal analog circuitry can be bypassed. Otherwise, the recording suffers all of the analog tape induced deficiencies present in all other Betacam SP machines. The audio as well as the video recordings are analog and cannot be classed as the same quality as Digital Betacam.

Describe other available digital component recorders. The field is split between tape based and disk based recorders. Disk based recorders at this point do not have removable media and are therefore fixed media recording and playback devices. Material recorded into the machine may not be ejected like a tape cassette and carried off. Recording times vary from 25 to 55 seconds for the Abekas A60 and A66 recorders to several minutes on the Abekas Hexus and Accom DIS. These recorders are usually used as utility or temporary storage devices for graphics production or post production editing. Disk recorders cost approximately $1000 per second of storage with greater efficiencies at longer recording lengths.

Longer times, up to several hours, can be found on the BTS Media Pool and Quantel Clip Box. These longer format recorders are targeted for the post production mass storage and television spot playback markets. Also available are disk recorders that utilize compression technology that trade a slight amount of picture quality for dramatically increased recording time on the same media. The Tektronix Profile recorder uses Motion JPEG compression that is comparable to Betacam SP recording quality; honorable but not perfect. The BTS Media Pool and Quantel Clip Box are also capable of variable compression ratios that trade economy of disk storage space for picture quality. Many of these disk recorders can be connected directly to computer work stations, such as Macintosh or Silicon Graphics, to be used as image storage devices. Images in a 3D sequence, for instance, are typically rendered on the computer at much less than real time and recorded to the disk recorder as they become available. Once the sequence is completed on the disk recorder it can be played back in real time.

In D1 long form edit suites, disk recorders are used to answer the advantage of Pre-Read editing that Digital Betacam offers. Material that would ordinarily require a second D1 source machine is laid off to the disk recorder instead. It then becomes available to perform effects like dissolves. With Pre-Read editing on Digital Betacam or D2, an effect can be performed directly using only one source and the recorder. Using a disk in this application requires extra time to lay off the material and extra hardware attached to the room.

Panasonic has had a long standing feud with Sony since the days of Betamax vs. VHS that extended into the broadcast/production arena. Panasonic introduced a string of extinct or nearly extinct tape formats that have all been answered by Sony products outselling the Panasonic products by 100:1 ratios. The latest is the Panasonic D5 component digital tape format that uses half inch metal tape. It has all the functionality of a D2 or Digital Betacam machine with four digital audio tracks, pre-read and the like. It was intended to be sold against the Sony D1 format but arrived too late to be effective. It is now being marketed against Digital Betacam. The recording does not use any form of compression and is completely transparent to a CCIR-601 (ITU/EBU-601) component digital image. The machine will accept and play tapes from the Panasonic D3 composite digital tape recorders. The D5 machine must decode the D3 recording internally before outputting it as a component digital signal. The format also supports a proposed wide screen 601 image standard that has not been popular with post production houses. Wide screen images are possible with current 601 standards with a slight resolution penalty. The proposed standard would correct the slight resolution loss but would completely obsolete current 601 hardware. The D5 machine is also adaptable to record compressed HDTV images. Even though the D5 machinery is beautifully designed and executed, almost nobody in the production or post production arena is supporting the D5 tape format at this time. Had the machine come to market two or three years earlier, it would probably be a different story.

Coming to the market are the Panasonic DVCPRO tape format and the Sony Betacam SX format. DVCPRO is an extension of a consumer digital tape format called DVC that uses 4:1 compression. The cassette is loaded with 1/4" metal tape and is about the size of an 8mm video tape. It is being offered in a variety of attractive news and production systems that are small, light and portable. Betacam SX is also compressed 4:1 with a proprietary Sony compression scheme. It uses standard Betacam SP cassette shells and will also play back current Betacam SP tapes. The format is intended for news systems and is capable of playing and recording Betacam SX tapes at 4 times real time (great for microwave or satellite news feeds).

Ampex has had their DCT tape format available since 1992. It uses a tape cassette shell similar to D1 or D2 and a transport identical to their D2 and data recorder products. The recording is component digital with variable compression. Depending on the complexity of the image, the recorder may choose to apply compression before recording. Compression will only occur with extremely complex picture information, like if more than half of the screen area is random noise. The DCT format has gained some acceptance as a less expensive D1 recorder alternative in New York and some other markets such as Baltimore.

What are the affects of various tape formats on film material? For all the quality that film brings to the television screen, most of it's advantage lies in film's ability to compress 1000:1 lighting ratios on scenes into the relatively narrow acceptance range of video. Otherwise, film can be a relatively poor medium to transfer to video. This is due largely to the static resolution of film, the 3:2 pulldown needed to correct frame rate disparities between film and television and the grain structure of film. Film transferred at 24 frames per second (fps) typically will not challenge the ability of most professional video recorders to reproduce the image.

Film grain is a noise pattern that does a lot to mask problems associated with video tape since most of the undesirable affects of tape involve noise of some sort. A viewer would be hard pressed to tell the difference between D1, D2, Betacam SP, Digital Betacam or 1 inch film transfers on a standard NTSC monitor. The 3:2 pulldown cadence of the film transfer holds each film frame static for two or three video fields which allows our eyes to fixate on and include the grain structure of the film in the image. If the film were run faster, such as 30 fps instead of 24, the subjective resolution of film improves dramatically simply because our eyes will average the film grain over more frames. This is the basic principle of noise reduction. If you have ever seen 16 mm film shot at 30 fps, you can relate to the sensation of the image quality approaching that of 35 mm film. The static resolution of each frame is the same, only the noise is subjectively reduced.

The advantage of digital recording lies in the ability of the format to exactly reproduce the same image quality with exactly the same colorimetry no matter which machine or digital editing system the tape is played back in. An analog recording and editing system has the opportunity to alter the absolute values of color and luminance even if carefully adjusted. Digital recorders are also immune to the minor dropouts of analog recordings that are accepted as normal. A digital recorder generally either puts out a perfect picture or a disastrous picture in a failure mode with little room for a gray area in between. Film transfers to any digital format is a choice that provides a margin of insurance against analog deficiencies contaminating the product.

What are detractors of Digital Betacam likely to claim against it? The video portion of digital Betacam uses a scheme called Bit Rate Reduction (BRR) or Coefficient Recording to pack very large amounts of data on a small amount of tape. This is a form of compression that detractors will point to as immediately destructive to the picture. The fact is that the compression is so mild that it would be a major challenge to notice the difference in the playback signal from the original, even after many generations through the compression scheme. Eventually, the compression scheme will show defects, but our experience shows that this will occur after 150 or so generations. Film transfer or original acquisition, long form editing and color correction to any tape format typically only requires three or four generations. Even if revisions increase that generation count to ten generations, you are still well within the limits of Digital Betacam to stand up to an alternative such as D1 editing. Pre-read, stunt modes, tape stock costs and other advantages of Digital Betacam will far outweigh anything editing to D1 can claim as an advantage. I would claim, and can demonstrate, that the 10 bit Digital Betacam picture will be degraded if it were recorded onto an 8 bit D1 tape.

How should a room be booked with Digital Betacam? For mastering to Digital Betacam, the optimum editing suite must be digital component. An analog component editing suite may be used but the advantage of transparent use of pre-read will be lost. The majority of the edit sources should be either D1, Digital Betacam, Betacam SP or some other component source. As noted earlier, if the bulk of the source material is composite, the client would be better served if a composite edit were done on D2 and then duped to a Digital Betacam tape through a high quality decoder if the delivery requirements specified a Digital Betacam end product.

If Digital Betacam is to be used as an edit suite source machine, all functions expected from a Betacam SP machine such as a BVW-65 or BVW-75 will be retained. The added benefit of easily mixing Betacam SP and Digital Betacam sources in the same source machine is an advantage.

Digital Betacam may be freely used to transport graphics from the digital component graphics environment into any edit suite equipped with a machine that will play it. When playing back a Digital Betacam tape in a digital component edit suite, picture quality and video levels may be absolutely locked down for easy layering without fear of video level anomalies. Playing back material in a composite environment most likely requires that the analog composite output of the Digital Betacam be used. Since the composite output is analog, it can not maintain absolute level matching that a pure composite digital signal path would allow. Care must be taken to match picture quality between Digital Betacam machines which makes them less than optimum as graphics sources in a composite edit suite.

Steve H. Wiedemann,
Sr. V.P., Director of Technology,
Henninger Media Services


Also See: Film Formats and HDTV: A Case for the Future Proof Standard

Page Top | Home | Facilities | Gallery | Services | Talent
Contacts | Press Releases | New Stuff | Search | Library