STUDIO

aFIG 5Appearancehead finding audio fromdifferent locations on diskand filling the RAMRAMpreloaded audio(unlike water it staysas discrete parts)audio output at a constant rateIn reality (again, none of the original audio has been altered)FIG 62nd verseskipTape (mono recording)A dio2nd chorusSilencePic'-f2 116 minsDisk Analogy (mono recording)4 minsFIG 81114 mins48 Studio Sound, June 19891A3j3space free forfurther recordingFIG 7Appearance15 beatist verse 1 1st chorus' 2nd verse 1In realitytst versewads n beat.,2nd chorus I ist "'elAvailable recording time = 16 minsActual audio = 6 minsTotal tape used = 16 minsWaste of space = 10 minsAvailable recording time = 16 minsActual audio = 6 minsTotal disk used = 6 minsWaste of space = 0 minsearly or late the whole process would have to berepeated. With disk, the system simply needs tobe told to play the 1st verse with a delay ofwhatever amount is required, as shown in Fig 7.Random allocationIt follows that if audio can be played in anysequence from any location on disk there is noreason why the audio should be recorded in alinear fashion in the first place. This concept is indirect contrast to tape recording. Since tape doesnot have random access, audio must be recordedin the physical order and time (directly proportionalto tape length) in which it is to bereplayed. For example, if there is a minute'sworth of silence between two sections of audio,there must be a minute's worth of tape betweenthem. Disk based systems can be used to record inthis linear way yet still take advantage of randomaccess, or they can flout the convention altogether.Consider the comparison shown in Fig 8.The disk system need not record the silences inorder to space the sections of audio the correctdistances apart. External sync information can bestored along with the audio and can be used toautomatically place the audio correctly on playback.As with Fig 7, the system will simply waitthe required time before playing the next sectionof audio.However, not all disk based systems allow totalrandom allocation. This is obviously a point toinvestigate when assessing a system and isdiscussed further in Part 2 of this article.ArchivingWhat to do with the audio once the disks are fulland new material is to be recorded is a problem.Hard disks are not removable so the audio mustbe offloaded onto another medium for achiving.This is where tape re- enters the scene as this isuniversally used for archiving audio from`tapeless systems' (a few systems claim to useoptical WORM drives for back -up- discussedfurther in Part 2). The tape used however, is notusually multitrack tape, but rather high speedtape, video 8 or some other suitable for recordingdigital data, yet realistic in size and cost. Thetime it takes to offload from disk to tape or viceversa, varies from system to system. Some arefaster than real time, some slower, but themajority take around real time.ConclusionIt may seem at first that it is defeating the objectif tape must eventually be used for storagepurposes, and indeed, if a disk -based system werebeing used purely in the same way as a stereo ormultitrack tape machine, it would. Hopefullyhowever, this article has shown some features ofdisk -based audio recording and editing which areeither impossible, or would take a great deal oftime and hassle to achieve using a tape basedsystem. This does not mean that hard diskrecorders are suitable for everyone's needs andthat tape no longer has a role in audio production.But disk based systems do have theiradvantages, particularly when it comes to sonicfidelity, editing and allowing the user to quicklytry out different arrangements of the audiowithout being destructive.In Part 2 of this series we shall cover the mainfeatures of disk -based systems, explaining thejargon and what to look for should you beinterested in reviewing such a system.