[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference napalm::commusic_v1

Title:* * Computer Music, MIDI, and Related Topics * *
Notice:Conference has been write-locked. Use new version.
Moderator:DYPSS1::SCHAFER
Created:Thu Feb 20 1986
Last Modified:Mon Aug 29 1994
Last Successful Update:Fri Jun 06 1997
Number of topics:2852
Total number of notes:33157

1479.0. "THOR Read/Write Optical Disk to Replace DAT?" by MDVAX1::EDLUND () Tue Jun 21 1988 16:55

    GET READY FOR NEW TOYS!!!!!!!!!!!!!!!!!!!!!
    
    TANDY CORP.  ANNOUNCED THIS PAST WEEK THAT THEY HAVE PERFECTED
    CD READ/WRITE TECHNOLOGY, AND WILL BE RELEASING IT THIS FALL
    FOR SALE TO THE GENERAL PUBLIC.  STORAGE ON THE DEVICE WILL
    BE 550 MEGABYTES, AND WILL BE AVAILABLE IN A CD DRIVE FOR
    COMPUTERS, AS WELL AS A CD DRIVE FOR AUDIO.  BOTH WILL RETAIL 
    FOR ABOUT $500.00.
    
    UNLIKE SOME OF THE CURRENT EMERGING CD TECHNOLOGIES THIS WILL
    OFFER DIGITAL READ/WRITE NOT JUST ONCE OR TWICE, BUT CAN BE 
    ERASED AND RERECORDED AS MANY TIMES AS YOU WANT TO.
    
    LET THESE FOLLOWING TWO SENERIO'S ROLL AROUND IN YOUR HEAD
    FOR A WHILE AND SEE IF YOU START TO DROOL LIKE I DO:
    
    1.  GOODBYE 3.5 INCH FLOPPY FOR YOUR SEQUENCER, HELLO 55O MEGABYTE
        ONLINE SEQUENCE AND PATCH STORAGE.  HELLO 550 MEGABYTES OF
        RECORDING SPACE FOR SAMPLING KEYBOARDS (TALK ABOUT CUSTOM
        SOUNDS?).  ETC...  ETC...   ETC...
    
    2.  GOODBYE SPENDING $5000.00 ON A GOOD 16 TRACK STUDIO QUALITY
        TAPE RECORDER, THAT MAKES GOOD TAPES FOR STEREO MIX DOWN.
        HELLO DIGITAL RECORDING STUDIO (NO MORE HISS, POP, ETC...)
        PUT ALL YOUR SOUNDS SOMEWHERE ON THE DISK.  WE DON'T CARE
        WHERE.  PING PONG THEM ALL TOGETHER, AND YOUVE GOT A STUDIO
        QUALITY DIGITAL RECORDING.
    
    JUST AS AN ADDED PLUS AUDIO RECORDINGS MADE ON THESE MACHINES 
    WILL CONFORM TO CURRENT CD STANDARDS, AND CAN BE PLAYED ON AN
    ORDINARY CD PLAYER.  SO KNOW YOU CAN MAKE YOUR OWN STUDIO 
    QUALITY RECORD (IF YOU CAN PLAY STUDIO QUALITY MUSIC) AND SEND
    IT TO UNCLE BUD.
    
    CAN HARDLY WAIT TO
    DROP MY $500.00
    
    JEFF
    
T.RTitleUserPersonal
Name
DateLines
1479.1GIBSON::DICKENSSurfing with my BuickTue Jun 21 1988 17:122
    Sounds great, but please don't YELL.  It's tiring to the eyes.
    
1479.2Case sensitiveDREGS::BLICKSTEINYo!Tue Jun 21 1988 17:205
    In case you're new to noting, .1 refers to typing in all upper-case.
    
    Please use mixed case for writing notes.
    
    	db
1479.3Tandy -> real?DEALIN::NELSONTue Jun 21 1988 17:2421
    Ditto sentiment of -.1  --  as ancient FORTRAN kinda guy, uppercase
    gives me such a pain, right between my eyes, just above the lower
    medulla, kitty-corner from...
    
    As long-time watcher of microcomputer scene (my main vice until
    a recent, more tuneful mania took over) I've grown far too cynical
    to register such excitement (even in lowercase). I'll believe it
    NOT when it's been actually sold in my local mall's store, but six
    months later when there are real applications that make it useful.
    
    On the other hand, Tandy has been exceptional historically for NOT
    making premature announcements. In fact, it's out of character for
    them to discuss the product before it's actually available in the
    stores. So maybe it's real...
    
    Does anyone care to start a little development project? Between
    the computer and musical expertise represented in this COMMUSIC
    community, we should be able to spec out and implement something
    seriously big-time. (Or is this contrary to contractual arrangements?)
    
    ?^) Hoyt
1479.4Two Month Old News?DRUMS::FEHSKENSTue Jun 21 1988 17:3826
    This is old hat by now, and as noted, it remains to be seen if this
    will make it to the stores as a real product.  I suspect it will,
    but not at $500 retail.  Also, with only two tracks, it will not
    be useful for "pingpong" multitracking (the best you'd be able
    to do is a high quality *mono* mix - to do stereo, you need two
    distinct open tracks available all the time to mix to; once you've
    got anything you want to use later on one of the tracks, you can't
    use it anymore!).  Also, the virtue of multitracking is the opportunity
    to reconsider mix decisions; the fewer tracks you have (three is
    the minimum for stereo work), the more irrevocable your decisions
    are.
          
    However, it *should* be a viable two track master/mixdown medium,
    and the data storage possibilities are intriguing.  Some relevant
    questions:
    
     - what will blank media cost?
    
     - how many erase/record cycles will it support?
    
     - regardless of the *medium's* inherent quality possibilities,
       will the supporting electronics yield a 95 db s/n (etc.)
       *record/play* cycle?
    
    len.
    
1479.5Still breathing.DYO780::SCHAFERBrad - DTN 433-2408Tue Jun 21 1988 17:4212
    I'll be interested in this technology when:

	a) someone BUYS one for $500
	b) someone invents a multi-track one (16 track CD!)
	c) someone buys b) for $500

    Until then, I'll stick to dbx.  But yes, it is intriguing.  Of course,
    some bozo will come along and legislate copy protection mechanisms ... 

    {sigh}

-b
1479.6SorryMDVAX1::EDLUNDTue Jun 21 1988 18:202
    Sorry about the uppercase, brand new to noting as of today.
    
1479.7More STuffMDVAX1::EDLUNDTue Jun 21 1988 18:2817
    Sorry about the uppercase only.  As far as the ping ponging goes,
    i can tell that your highly knowledgable with analogue.  But in
    this instance, (digital) we don't care how many actual tracks we
    have, all we care about are are combining data groups from various
    locations on the disk to a new area.  By the way as i understand
    this technology there will also be selective erase.  i.e.  removale
    of a single data group, instead of complete erase of the entire
    program.
    
    Tandy also claims that there will be no limit to the amount of times
    that the media can be erased and rewritten.
    
    
    
    Still Drooling
    
    jeff
1479.8Still Not Possible Without Additional TechnologyDRUMS::FEHSKENSTue Jun 21 1988 19:3423
    You can't read from multiple places on the disk to combine them
    together unless you have multiple read heads (a possibility) or
    sufficient storage to buffer one of the "data groups" (also possible,
    but far less practical).  I have yet to hear of anybody providing
    multiple read heads on a CD and I highly doubt that Tandy's consumer
    version will do so.  Perhaps a special "pro" or "semipro" version
    will be made available that does so, but even then it's unlikely
    that more than 4 head assemblies could be provided (given the physical
    dimensions of the media and problems of physical access), and there are
    then important restrictions about spatial layout on the disk of data
    groups that could be read at the same time given the possibility
    of the heads physically interfering with one another.  Perhaps a
    combination of two heads and modest buffering could be "interleaved"
    to provide "scatter/gather" read/write capabilities?  Things are more
    complicated if the read and write heads have to be physically distinct.
    All these problems could be solved, but only for a limited appeal
    semipro market, i.e., low volume and hence rather higher retail price.
                    
    Maybe I'm missing something, but I know of no way to mix two
    tracks without reading them more or less at the same time.
         
    len.
    
1479.9Buffered audio?PAULJ::HARRIMANHell's only command: 'SET'Tue Jun 21 1988 20:017
    
    re: .-1 Len
    
      Unless you *like* to listen to lots of starts and stops in your
    audio program....
    
    /pjh ;^)
1479.10Oh, it ain't no big deal...NAC::PICKETTDo the voices in my head bother you?Tue Jun 21 1988 20:2516
    re .8
    
    Len:
    
    To Mix in the digital domain, just XOR the bits. That should work,
    right??
    
    dp 
    
    p.s. Ya know, that would be interesting to try, just to see what
         you get. I think I'll write a program to XOR sine waves, and
         see whaty I get. Better yet, I'll make a hardware mod to my
         ESQ-1, and report back ;^)
    
        
    
1479.11the amiga could probably do this tooCOUGAR::JANZENTom 2965421 LMO2/O23Tue Jun 21 1988 21:003
    that's like saying, to add two numbers just XOR the words, but that's
    not right.
    Tom
1479.12I'll get in the pool...CTHULU::YERAZUNISMadness in the method...Tue Jun 21 1988 22:3433
    To mix tracks digitally:
    	
    	OUT = IN_1 + IN_2 + .... 	(if there's no need for fading/EQ)
    
    		or
    
    	OUT = F_1 * IN_1  + F_2 * IN_2 + ...	(allows fading, 0<F_N<1 )
                                                                         
    There's no need to have more than one read/write head.  Provided you
    can describe the mix (either it's static, or it's slowly varying ONLY)
    and have some small amount of memory (2 sec worth is fine) just:
    
    	1) read the first second's worth of the first track.
    	2) use the above equation(s) to add in the second, third, etc.
    	   tracks (read from disk, add to memory, repeat).
    	3) write out the first second's worth to the output track.
        4) update seconds counter.
    	5) go to 1.                                     
                                  
    There's no need for anything esoteric.  You don't get to hear the
    realtime mix until you've finished the above (well, you could probably
    set it up so that you would hear "snippets" as each second was
    completed) but that's just a minor annoyance.
                                 
    -----
    
    Even if _all_ the first-generation machines do is let me make CDs of
    audio (no editing) I'm highly interested.  Being an old PCM-F1 user,
    I'm used to "one chance, realtime mixing".  Where do I put down the
    $500? 
    
    	-Bill
             
1479.13This Is a Joke, Right?DRUMS::FEHSKENSWed Jun 22 1988 14:1357
    
    Yes, you've described in more detail exactly what I meant when I
    said a combination of a modest buffer and interleaved access.  However,
    it's not real time, and that's not what most people mean by
    "pingponging" or mixdown.  Being unable to hear the mix until after
    it's done makes adjusting the mix a little difficult.  Do you know
    of any engineer who tweaks his levels without listening to what's
    going on?  Set the levels, run the mix, see if I got it right?
    Imagine being forced to do this with analog technology!  Would anybody
    buy it?  I think this is more than just a "minor" annoyance.
    
    Look, I know how to operate on digital data (sheesh, I've only been
    a computer professional for 24 years).  The qualifications you put
    on this "mixing" technique are unbelievable!  "Provided you can describe
    the mix" indeed.  Beforehand, in full detail, algorithmically!
    I can hear "snippets" as each second is completed!  How long does
    it take to mix a second's worth of material?  Suppose I'm "mixing"
    n tracks.  Then I have n seconds' worth of reading, 1 second's worth
    of writing and (n+1) seeks to get to the tracks to read or write.
    If the drive is suitably capable (I can just imagine a CD player
    doing random access seeks all over the disk!), maybe the seek time
    will be less than a second.  C'mon, how long does it take a CD player
    to get to a randomly selected track?  What's the worst case?
    
    Let's suppose it's a second. Then it takes 2*(n+1) seconds to create
    1 second's worth of output.  Let n = 4, the usual minimum number
    of tracks considered legitimately called "multitracking".  I get
    1 second's worth of output every 10 seconds!  This is a minor
    annoyance?  Do you expect anybody to actually listen to this and
    make useful judgements about it?  Let me be generous and assume 3 source
    tracks "pingponged" (mighty leaden pingpong balls these are) to
    a single destination track (more like the 4 track situation); but it's
    still 8 seconds of silence, then 1 second of output, then 8 more
    seconds of silence...   I'm sorry, I have to consider this no more
    than a cruel joke.
    
    Finally, this scheme makes overdubbing a real challenge.  It's not
    possible to hear the tracks you're overdubbing against!  This is
    synch city, strictly by the book.  I suppose I could overdub
    "snippets", right?  Hear 1 second of existing tracks, then play
    my one second overdub, then listen to another 1 second, then ...
    Just a minor annoyance, right?  Is this your vision of hell for
    studio musicians and engineers?
        
    Also, you've ignored the question of any processing at all besides
    amplitude scaling.  "No need for EQ" indeed.  Tell that the mixer
    makers, they could save a lot of money.  Somewhere along they line
    they got the foolish notion that EQ and effects sends/returns were
    useful features to make available for mixdown. 
                                     
    This whole discussion reminds me of an old joke about the "excellent"
    tailor who never had to make any alterations; he just got his customers
    to scrunch their bodies up until they fit the clothes. 
                                                
    len.
           
    
1479.14NYMPH::ZACHWIEJAFlushing my buffers...Wed Jun 22 1988 14:4711
    
    Why is it that y'all seem to be discounting the notion of all the
    tracks being read into a buffer on a computer,  allowing  you  to
    mix n tracks all at once and hear it all as it goes into  another
    buffer and eventually back out onto the disk ?
    
    Granted we are talking about a ton of data,  but with virtual mem-
    ory and independent IO off  of  the  disk,  it  shouldn't  be   a
    problem.
    
    _sjz.
1479.15Let's stir this pot a littleCCYLON::ANDERSONWed Jun 22 1988 15:108
    Let's take this a bit further... With the advancing state of memory
    technology our CDs will soon be replaced with EEPROM wafers. And
    then the date comes out 16 bits as a time. Now thats either in
    alternating word per channel as is currently done or 16 simultanious
    tracks serialized you decide...
    
    Jim
    
1479.16Why Make Unnecessarily Outrageous Claims?DRUMS::FEHSKENSWed Jun 22 1988 15:1131
    You get 5.94 seconds of CD quality (16 bit, 44.1KHz) stereo data
    per Megabyte, discounting any overhead (and there's a *lot* of overhead
    on a CD track, at least a factor of 2's worth).  So you wanna mix
    a 3 minute pop tune from 16 tracks (that was the original assertion,
    no more 16 tracks needed to do CD quality mixes) that's 24 minutes
    worth of source material.  Say 10 MB per minute, that's 240 MB of
    "direct to disk" data.
    
    So, we sell the CD player for $500, and then a 300 MB hard drive
    and processor as a $10000 accessory?  No problem.  In fact, you
    can buy one today from New England Digital.  It's called Synclavier
    Direct to Disk, and it will only cost you $100K.
    
    I must really be missing something. 
                                                    
    Look, a writeable CD is pretty neat even if it *doesn't* replace
    multitrack recorders.  It works just great as a stereo mixdown
    medium.  Why ask it do more, especially more that it's really quite
    ill suited for?
    
    The right way to do digital multitrack is to sync multiple DAT decks
    together.  Then you get extensibility (you buy tracks two at a time)
    and real time and you don't need a microVAX and a huge disk.  Once
    DAT decks are commodity items the price will fall incredibly: remember
    when you couldn't touch a CD player for less than $1K?  Why will
    DAT be any different?  Or maybe it'll be possible to synch multiple
    writeable CDs together to the same end.  DAT's here today and it
    works;  this writeable CD for $500 is still an assertion.
    
    len.
    
1479.17Oh, Yes, One Other Small MatterDRUMS::FEHSKENSWed Jun 22 1988 15:2413
    One last thing - if you think the RIAA is unhappy about DAT (which
    *can't* record at a 44.1 KHz sampling rate), how do you think they're
    going to react to a medium that's not only directly data compatible
    (the CD standard is 44.1 KHz; if the machine can't record at that
    sampling rate, then its CDs can't be played on an "ordinary" CD
    player) but is also *media* compatible with existing CD players.
    Now those pirate teenyboppers won't even have to buy a DAT deck
    to play their illegally copied CDs!  You think the RIAA, having
    taken on the entire Japanese audio industry with respect to DAT,
    is going to let Tandy get away with this?
    
    len.
      
1479.18Is There Something in The Air?DRUMS::FEHSKENSWed Jun 22 1988 15:3210
    re .15 - What do you mean by soon?
    
    A 60 minute CD stores about 600MB.  When we get *close* to that
    kind of storage capacity (the next generation of chips are 1Mb,
    so a CD is the equivalent of a "handful" of 4800 chips; the generation
    after that of 4 Mb parts will mean only 1200 chips are necessary)
    I'll start thinking "soon".
     
    len.
    
1479.19Food For Thought?MDVAX1::EDLUNDWed Jun 22 1988 16:3325
    Wow I never thought I would open such a can of worms!!!!!!
    Well anyway to drop a few more names and words Tandy calls this
    new technology THOR (and yes it is in all caps).  Tandy is being
    so secretive about the technology they won't say what THOR stands
    for.  
    
    They will admit that instead of using laser heat to cause a bump
    on the expansion layer of the disk, and conversely laser heat on
    the retention layer to flatten out the bump: they will employ
    special dye-polymer reccording techniques to use a disk pitting
    format.  Just like nonerasable CD.
    
    Finally Lynn Haley, spokesperson for Tandy says that they are already
    experiencing resistance from the record industry, but she also adds
    that just like the tape recorder. "You can't stop a speeding train
    thats already going down the track".
    
    (How about another thought for multi tracking.  Instead of multiple
    lasers, what about a rotating mirror to split/scatter the beam.
    Then receive at multiple photodetector stations?)
    
    Just a thought.
    
    jeff
    
1479.20Like wow, THOR...DARTS::COTELook!! Eeet eees BASSOON!Wed Jun 22 1988 16:523
    Totally Harmonic Optical Recording
    
    Edd
1479.21Think like an inventor!CTHULU::YERAZUNISYou're walking along the beach and you find a tortise...Wed Jun 22 1988 18:1470
    Re: no need for eq:  You can put EQing into the equation; I didn't
    want to muddy the waters.  Even have non-causal filters, if you
    really want them.  
    
    Sure, mixdown will take time, but you can barely get a decent 4-track
    cassette these days for $500.
    
    Overdubbing: no, you can overdub continuously with a small additional
    hardware feature.  Works like this:
    	
    	1) Make MONO mixdown.
    
    	2) Transcribe mono mixdown to consumer cassette, with mix on
    	   one channel and sync track on the other.
    	
    	3) PLAY mono mix, THOR device listens to sync track to get
    	   timing information, THOR records new track continuously.
    	   (FFwd and Reverse are options).  [sync-in is the additional
    	   hardware feature needed]
    
    	4) Mix/punch-in/punch-out new track into mix description.
    
    If your THOR can seek at a reasonable speed, you can omit the 
    cassette interstep (which is only there for the live performer to
    listen to, it isn't re-recorded.)
    
    -----
    
    Re: mixdown: How do we know that a THOR box will only be able to
    seek as quickly as a CD player?  My RX50 floppies seek about 100x
    faster than my CD player; we have no real information about seek
    time on THOR.  There's a continuum of seek mechanisms (from worm-gear
    to stepper, to voice-coil); all the CD players I know of are the
    lowest performance system (worm-gear).
    
    It would only take an order of magnitude improvement in seek time
    (or less, if we could spin the disk a lot faster) to do mixdowns
    in real time (the controls would have a bit of lag, but the sound
    output would be continuous).  Likewise real-time dubbing.
                                          
    -----
    
    Tandy is the right size to take on RIAA.  Also, they have the right
    location for their headquarters- Texas, not Tokyo.  Let's hear it
    for chauvanism!
    
    -----
    
    I wouldn't be surprised if the _first_ system Tandy came out with
    was a 4-track mixdown system; it's much easier to explain to a judge
    that your system isn't designed to make illegal copies if it has
    a bunch of features that are clearly designed to help the original
    musician, and are useless fluff for copying purposes.
    
    -----
    
    This entire discussion may be moot if one important number is too
    high:  
                                   
    
    
    			How much do blank disks cost?
    
    
                        
    
    
    	-Bill (lost-in-the-oberheim-matrix)
    
    
1479.22I'm thinking like a mad inventorANGORA::JANZENTom 2965421 LMO2/O23Wed Jun 22 1988 18:152
    How about a separate audio channel on each track?
    Tom
1479.23my guessSUBSYS::ORINAMIGA te amoWed Jun 22 1988 19:008
Tandy Hifidelity Optical Recorder?

Sell your decks. Hypermedia is happening now! CD ROM, CD RAM, SMPTE/MIDI,
MIDI workstations, megasampling, multitimbral, fiber optics...

The next 5 yrs are going to be incredible for technology.

dave
1479.24My Dollar Says OuchMDVAX1::EDLUNDWed Jun 22 1988 19:0810
    You bet its incredible!  I'm sure it is bound to have a profound
    negative effect on my bank balance as well.  Speaking for myself,
    I can never get enough of high tech toys.
    
    
    
    Nearing deficit spending
    
    jeff
    
1479.25not to mention KoreaANGORA::JANZENTom 2965421 LMO2/O23Wed Jun 22 1988 19:118
    >MIDI workstations, megasampling, multitimbral, fiber optics...
>
>The next 5 yrs are going to be incredible for technology.
>
>dave
>
and for the Japanese econony, as well!
    Tom
1479.26GIBSON::DICKENSSurfing with my BuickWed Jun 22 1988 19:1619
    re .6: Welcome
    
    re others
    
    Even the mighty Synclavier Direct-to-Disk doesn't have real-time
    mixdown.  I just think that would be a colossal step backwards to
    have to wait for the processor to crunch out your mix.  Ugly.
    
    I like the multiple-dat deck theory too, but that could get expensive.
    I want something like the Akai digital 12 track only scaled down
    to the $1-2K range.  I believe it's possible in the near future.
    I'd pay more if it had the digital-domain mixer (real time like
    the dmp7 !!!) built in.  
    
    Oh great techno-gods, we beseech thee, hear us...
    
    							-Jeff
    
    
1479.27Not the $6 a disk I wanted to see... But...CCYLON::ANDERSONThu Jun 23 1988 15:097
    Media will be in the $10 - $15 range. They are also mumbling about
    different media for digital vs audio app's although I find that
    hard to believe is anything more than marketing BS. Especially when
    one deck east all.
    
    Jim
    
1479.28Right, But How Many "Inventions" Actually Get Manufactured?DRUMS::FEHSKENSThu Jun 23 1988 18:5359
    re .21 - Everything you say is true, but I'm even more reminded
    of the excellent tailor.
    
    So your *consumer* oriented CD player/recorder now has an external
    sync input and must be capable of recording on one track (of the
    stereo pair) at a time.  In addition you *expect* Tandy to bring
    out a 4 track multitracking version so as to avoid the wrath of
    the RIAA.  Oh yes, it's also going to have a high rate high durability
    head actuator.  All this at no additional cost?
    
    Sorry, the only way Tandy could possibly make this device for $500
    retail is to take considerable advantage of existing CD player parts.
    Each one of the features you propose as modest extensions to the
    basic product has significant impact on this.  Why have no 4 track
    cassette recorders (never mind semi-pro 8 track units like my Tascam
    38, which costs a mere $2400) bothered to add the "modest" extension
    of chase lock drive motors so multiple decks could be synched?  Uhm,
    'cause it's not that modest an extension.  And it's no easier for
    a CD player.  You think that a high rate head actuator is going
    to be no extra cost?  Think again.  And a 4 channel format means
    a new CD data format, and the corresponding additions/changes to
    the supporting logic.  I.e., no off the shelf VLSI CD parts need
    apply.
    
    And how much market is there for such a device?  Not much in the
    consumer market.  You're providing features the consumer market doesn't
    want (a new format disk, rendering it incompatible with existing
    players) or can't use (chase lock), and requiring them to pay more
    for them.  Not likely to go over too well with the sales/marketing
    types.  Tandy is a business.  They'll go where the money is, not
    where we wish they would.
    
    Finally, your "dub to tape" overdub scheme would certainly work,
    but again, I doubt most home recordists would find it less than
    a major annoyance.  Every overdub requires a partial mixdown?
    Most home recordists would say "screw it, digital quality isn't
    worth this much trouble".  Besides, somebody for whom time is no
    object and expense is the overriding concern probably doesn't need
    (or couldn't exploit) digital quality.  All their other components
    have 95 db S/N, right?  And 16 bit samples, right?  And 44.1 KHz
    sample rates, right? 
    
    I appreciate your desire to get multitrack digital recording capability
    at minimal cost, but the lengths to which you're willing to go
    bespeak an extraordinary desperation.  But more relevant, it's
    extremely unlikely that Tandy (or any other consumer audio component
    manufacturer) would give these notions more than a moment's thought.
    Perhaps some speciality manufacturer will bring out a "semipro"
    version somewhere down the line, but don't expect to see this
    technology (chase lock 4 track high seek rate CD player/recorder)
    for $500.  And the mixdown/overdub techniques you've proposed would
    be the subject of scathing reviews in all the magazines.
    
    Really, I don't want to be a wet blanket, but the bulk of this
    discussion is just plain utterly impractical in the real world I
    know.
                                                                  
    len.
    
1479.29Mother of InventionMDVAX1::EDLUNDThu Jun 23 1988 19:3018
    re.28
    
    I understand your position on the supositions made as to possible
    features which could be implemented on such a device.  As far as
    your pessimistic fews on the birth and growth of this device are
    most likely true; I honestly feel that as a people and a nation
    we would be nowhere without the dreamers, speculators, and risk takers,
    that shape the adolesence of the products we see on the shelf today.
    
    I contend that in our minds and in this type of forum, we have the
    right as well as responsibility to conjure up any visions of this
    device that we please.
    
    
    still dreaming
    
    jeff
    
1479.30Engineering vs. DreamingDRUMS::FEHSKENSThu Jun 23 1988 20:0630
    I encourage that kind of thought, but I'm also practical enough
    to want to see things like this actually happen (e.g., see my proposed
    design of the "ultimate drum machine" in another note, a proposal
    which pushes available and probable technologies to the limit;
    also the discussion where I proposed a synthesizer based on
    Gigaflops computing capacity).  Visions that are unattainable may
    be fun, but I think there are enough radical visions that *are*
    attainable that fantasizing the impractical or impossible seems
    to be no more, at least to me, than fantasizing.
    
    For example, I think it's worth pursuing the notion of multiple
    heads on a CD player.  I believe the track allocation problems could
    be worked out to allow synchronized recording of 8 tracks on a Tandy
    style disk with no need for chase lock or a different disk format.
    You'd only be able to get maybe 14 or 15 minutes worth of material
    on a single disk, but that's ok for a lot of applications.  The
    same head actuators as are currently available could be used.
    Adding three or four more head actuators and some track allocation
    logic to an otherwise unchanged CD player/recorder *is* something
    a semipro vendor might seriously consider.
    
    I'd rather you saw my positions as practical and realistic rather
    than "pessimistic".  I have not desire whatsoever to discourage
    invention or innovation, but there's a difference between that and
    woolgathering.  As somebody once said, "Engineering is the art of
    the possible".  Anybody can come up with pipe dreams - the challenge
    is coming up with "pipe dreams" that can be made real.
    
    len.
    
1479.31Laser-One?MIDEVL::YERAZUNISWhy are so few of us left healthy, active, and without personaliThu Jun 23 1988 21:0731
    Three actuators, everything else the same, would get you the equivalent
    of a "porta-one".  You can write up to two channels, or read two
    stereo pairs (four channels) in realtime, mixing down to a stereo
    pair.
    
    (note- you gotta do some gymnastics and can't do:
    	
    	write 1&3
    	write 2&4
    	write 1&2
    
    because you run out of heads.  However, you can't do this on a
    Porta-One either- but for a different reason.
    
    With 4 heads, you can do everything a Porta-one can do and a bit
    more, like recording on all 4 channels simultaneously, shuffling
    1<-->4, etc.
    
    You also can use more than 4 tracks, as long as you do NOT try
    to access more than 2 stereo pairs at a time.
    
    Question: Does the same head read as well as write? 
    
    -----
    
    Len, you are correct.  Tandy _will_ go where the money is.  Nothing
    else is important.  Lawyers can be bought, if there are enough millions
    at stake.
    
    	-Bill
    
1479.3278 rpmDFLAT::DICKSONNetwork Design toolsFri Jun 24 1988 13:4417
This multiple head business is forgetting a crucial fact about CDs: their
rotational velocity is not constant.  They spin faster when the head is
on an inner track than when it is on an outer track, so as to give a constant
linear velocity past the head.

Thus you can not have multiple heads at different radii while preserving
the standard encoding format.

It could be done with a different format, allowing for say four bands with
different bit spacing, but you would lose capacity.  But if you are going
to change formats, you are already incompatible so you might as well go
to a bit-interleaved method, thus saving on mechanical complexity.  Mechanical
complexity is always to be avoided, even if it means you have to design
a different VLSI controller for this market.

Each track of 16 bit 44.1kHz samples needs 705k bits per second by the way,
not counting ECC.  CDs carry a LOT of ECC.
1479.33Aughhh. The CAV's got me!CTHULU::YERAZUNISVAXstation Repo ManFri Jun 24 1988 18:119
    AUUGGHH !
    
    I forgot!
                                                                     
    Sounds like you gotta have big buffers or do some really funny stuff
    with formatting.
    	
    	-Bill
    
1479.34Devil's Advocate meets THOR.TALLIS::HERDEGMark Herdeg, LTN1-2/B17 226-6520Fri Jun 24 1988 19:3021
    If I correctly recall the announcement of THOR by Tandy, which was
    several months ago, they said that the under-$500 audio recorder would
    be available in the fall of 1989, not this fall. It sounds pretty
    tentative to me, both the price and ship date.
    
    Also, the cost of a blank CD for the machine may be quite high. I
    remember them saying that they hoped it wouldn't be more than $25
    initially and should then go lower.
    
    They weren't sure how many write/erase cycles the blanks would tolerate.
    I think they had only tested them for a few dozen so far. Also, I think
    you can only bulk erase the entire disc, meaning it wouldn't make a very
    good random-access read/write device.
    
    Finally, the audio recorder is not useable for computer data purposes
    since the error correction, which is plenty to for good sound, won't
    assure data integrity. This would have to be another more expensive
    device.
    
    -Mark
1479.35DFLAT::DICKSONNetwork Design toolsFri Jun 24 1988 20:005
A recent article in USA TODAY about optical computer disks had the per-disk
price for these things running around $200, not $25.  They were not talking
only about Tandy, but had a table of several.  The guy in the picture was
holding up a thing that looked more like a Bournoulli removable hard-disk than
a CD. 
1479.36MIZZOU::SHERMANBaron of GraymatterSat Jul 02 1988 19:277
    re:-2 or so
    
    Yeah, I remember the announcements about THOR coming out around
    Fall of next year.  Has there been some sort of update on the release
    date?
    
    Steve_who_just_got_back_from_vacation
1479.37NYMPH::ZACHWIEJASubstitution Mass ConfusionTue Jul 05 1988 14:2727
    
    [ Radio-Electronics, August 1988, p.6 ]
    
    The surprise announcement by Tandy Corporation (owner of the Radio
    Shack chain and the Memorex video and audio tape business),   that
    it had developed a low-cost optical disc recording  system,  could
    have great impact on both the audio and video insdustries. The sys-
    tem,  which Tandy calls THOR-CD (for Tandy High Intensity  Optical
    Recording) is claimed to permit  repeated  digital  recording  and
    erasure of discs which are playable on standard CD players.  Tandy
    says the recorder will cost less than $500 and  be  ready  in  two
    years.
    
    Although the system is capable of making digital audio  and  compu-
    ter discs,  Tandy noted that video discs are among  its  possibili-
    ties as well.  The discs,  however,  won't be compatible with  cur-
    rent laserdiscs,   according to Tandy officials,  because THOR  is
    a digital record medium.  The THOR video  disc  could  become  the
    first medioum for consumer digital-video recording.
    
    The THOR system is based in part on a process developed by Optical
    Data Inc.,  using a disc coated with a dye polymer coating,  which
    forms bump-like deformations when exposed to  a  laser  beam.  The
    mark or bump,  which doesn't deteriorate with repeated use, can be
    erased and rewritten when exposed to a laser beam of another  wave-
    length.  The system is inexpensive and can be manufactured in high
    volume with existing magnetic-media techniques.
1479.38NYMPH::ZACHWIEJASubstitution Mass ConfusionTue Jul 05 1988 14:308
    
    Re .37
    
    Which means sometime in 1990.  More than enough time to fall behind
    schedule and more than enough time for the  recording  industry  to
    file some sort of a legal case and push this into the 21st  century.
    
    _sjz.
1479.39More Latest NewsDRUMS::FEHSKENSTue Jul 05 1988 14:4118
    The latest MIX included two interesting developments.
    
    First, an ad for the Sony professional DAT recorder.  No price
    mentioned.  It *does* record at 44.1KHz.  And I'll bet it has a
    digital input.
    
    Second, a brief announcement about a DAT defense fund.  The RIAA
    has apparently threatened to sue anyone who imports and sells a
    DAT recorder.  So somebody organized a defense fund that the
    manufacturers and importers have been kicking money into.  So far
    there's only a coupla hundred thou in the fund but it's looking
    like the DAT contingent is going to tell the RIAA to stick it where
    the sun don't shine.  I think (hope?) this is all just going to
    turn into a replay of the Sony Betamax case, i.e., the RIAA will
    lose.  But it may take years.
    
    len.
    
1479.40MIZZOU::SHERMANincompetence knows no boundsTue Jul 05 1988 14:413
    Hmmm.  Looks like I'm stuck with my cassette masters ...
    
    Steve_who's_too_cheap_to_go_VHS_Hi-Fi
1479.41McLuhan to me!SHIRE::PETRAITISZen windows look inwardWed Dec 28 1988 14:4717
    DAT's selling now in Europe. I've seen a CASIO DAT for SFr 1650
    (about $1100). Can't Imagine that RIAA will stop a tidal wave...maybe
    just slow it down enough to build up massive force/pressure.
    
    DAT's and writeable opticals are coming to America, too. It's a
    question of how can we value the contributios of our artists while
    freeing up the usage of media. Just ruminating on how lawyers and
    short sighted interest group politics can possibly stop technology.
    I don't believe they can. I think there may be some other way to
    value the unique contributions of artists. Maybe by paying them
    to perform. BTW - some artists have contributed freely to things
    like the COMMUSIC tapes. Freeware and shareware music?
    
    The media is the message still true I guess.
                                                          
    David
    
1479.42DNEAST::BOTTOM_DAVIDEveryday I got the bluesWed Dec 28 1988 15:1119
    tehre was an intersting editorial in an English recording studio
    mag I had given to me a few weeks ago on this topic. Essentially
    the editor felt that the R/W optical disks were going to blow the
    market open on DAT as the RIAA and it's circus of record companies
    could not afford to litigate with the world computer manuifacturers
    and any legislation that banned thier use could not be selective
    enough to ban only audio to digital (etc.) applications, essentially
    our employer would fight for the right to record digitally.
    
    he also mentioned how while the record companies dread DAT and writable
    optical disks, they can't wait to sell pre-recorded DAT tapes...and
    that DAT was a better medium for auto use than CD, and they (record
    companies) even admit this.
    
    DAT is available here in the US, you just have to be willing to
    pay the price.
    
    dbii
    admit it
1479.43DAT ain't the way I did itSALSA::MOELLERThis space intentionally Left Bank.Wed Dec 28 1988 15:2613
    I couldn't wait for DAT.. why pay premium for a deck whose cassettes
    are hard to find and fragile ?  So I'm happy as can be with my VHS
    HiFi **PCM** Toshiba deck.. thanks to Dave Blickstein, who first
    told me about it in this very conference.  BTW PCM recordings can
    be 'backed up' and song order changed by just hooking the video
    out to another VHS recorder, and it never leaves the digital domain.
    
    So it's true I can't play a VHS PCM videocassette in my car.. the
    background noise in my car makes a standard cassette sound good...
    did I say that right ?  That is, at speed, I don't even care if
    the cassette playing has Dolby (tm) on or not.. why do I need DAT?
    
    karl
1479.44MIZZOU::SHERMANLove is a decision ...Wed Dec 28 1988 17:298
    I suspect/hope that when the artificial constraints come off the
    price of DAT will come down.  Meanwhile, my metal_tape/Dolby_B combo,
    sad as it is, proves to be sufficient for my needs right now.  It's
    pretty easy to clean up the cassette master in the studio if you
    know what you're doing (watch levels, compress, demagnetize, clean
    heads and so forth).
    
    Steve
1479.45Confused of ManchesterWARDER::KENTThu Jan 05 1989 10:409
    
    
    re -2                                         
    
    Are all VHS-Hifi decks digital by definition ?  I.E. is the prerecorded
    soundtrack on a VHS-Film which plays back in stereo on a suitable
    equiped recorder recorded digitally onto the media?
    
    					Paul.
1479.46NopeAQUA::ROSTMarshall rules but Fender controlsThu Jan 05 1989 11:416
    
    Absolutely not.  None of the hi-fi VCR modes are digital.  The Toshiba
    is set up to work either as a VCR with (analog) VHS hi-fi or as
    a PCM digital audio *only* deck.
    
    
1479.47No, it implies just the oppositeDREGS::BLICKSTEINYo!Thu Jan 05 1989 13:1815
    > Are all VHS HiFi decks digital by definition?
    
    Nope.  All VHS Hi-Fi decks are ANALOG by definition (both audio
    and video).   VHS Hi-Fi is a trade-marked standard for recording
    audio channels on video tape.  It records an analog signal onto
    helical video track.  The extra bandwidth for the audio comes from
    using a technique called "depth multiplexing".
    
    "Digital", as a adjective in front of VCR, implies that the VCR
    has certain VIDEO features like PIP (picture-in-a-picture or "inset",
    digital freeze frame, polarization, etc.).  Basically "digital
    effects involve digitizing the video signal manipulated the picture
    in the digital domain.
    
    	db
1479.4820 KHz = .02 MHzDRUMS::FEHSKENSFri Jan 06 1989 20:0710
    Right, it's actually analog.  What's done is the audio bandwidth
    is frequency shifted up into the video realm where there's far more
    bandwidth available (e.g., between the chroma and luminance bands)
    than an audio track will ever need.  Then on playback it's shifted
    back down.  Beta sticks the audio in the chroma/luminance gap; VHS
    "depth multiplexes" it over the video (both chroma and luminance).
    Both techniques are pretty clever, and work quite well.
    
    len.
    
1479.49More More More...WARDER::KENTMon Jan 09 1989 10:249
    
    
    re- 1,2,3.
    
    In that case could Karl give us a bit more detail as to what he
    is using.
                                     
    				Paul.
    					
1479.50stereo digital mastering, VHS styleSALSA::MOELLERFrom AZ to OZ...Mon Jan 09 1989 15:4618
    okay, paul..
    
    The Toshiba deck I have does have normal, analog VHS Hi-Fi available.

    It also has a separate stereo circuit that utilizes PCM (Pulse Code
    Modulation) sampling/playback.  There are separate RCA-type line
    level audio ins/outs for PCM.  
    
    The D-A and A-D converters work at 44.1KHz sampling rate, 14-bit
    resolution, in stereo!.  This DIGITAL data is recorded in the video 
    portion of the video tape.  So all I do is enable PCM record/playback,
    hook up the line level ins/outs, set to no auto-leveling, set my
    input levels, and start recording.  CD quality.  2 hours of music
    on 1 videocassette at SP, standard speed.  Apparently also records
    digitally just fine at EP, slow speed, which supposedly gives a
    full 6 hours of music.
    
    karl
1479.51still awaitin' ...MIZZOU::SHERMANECADSR::SHERMAN 235-8176, 223-3326Mon Jul 16 1990 18:583
    Hmmm.  Musta' meant fall of '90?  Methinks I smell vaporware ...
    
    Steve