Conference napalm::commusic_v1

    
    	Japanese audio companies have always *loved* the "DIN" connector.
    It's like they're thing - it goes back to the 60's. I 'spose they
    could have chosen a BNC, like our thinwire Ethernet.
    
    							Joe Jas

    I don't have all the answers to .0, but I can speculate.  As .1
    implied, the standard was heavily influenced by the Japanese, who like
    the DIN connector.  The data rate had to be high, to get enough bits
    across, so RS232, with its limit of 9600 bps, was out.  I don't know
    why they didn't use RS423, with its good noise immunity, high speed and
    long lines.  Possibly they just weren't aware of it. They needed high
    speed and no errors, since they didn't want the expense of an error
    correcting protocol like HDLC, so they picked a conservative circuit
    technology (current loop, shielded twisted pair) and made strict rules
    on fanout (one receiver per driver).
    
    Since they needed such high speed (the biggest complaint is that the
    speed is still too slow) they probably couldn't have connected to
    standard serial interfaces anyway, so I don't think they lost much by
    using other non-standard parts.
        John Sauter

    ....a bunch of analog engineers designing a digital communication
    bus...ecch! You can see how well they understood the spec they wrote
    by buying on old roland synth and finding out it is not "fully MIDI
    conpatable"...but of course since we blew it we'll gladly SELL you
    the fix for OUR errors....
    
    By being non-standard they created a whole new industry that sells
    us junk to fix the oversights/inadeqauacies in the original spec.
    
    They should have gone with Rs-423 or IEEE 488, this would have allowed
    anyone who owned a PC of any manufacture to convert to MIDI  easily,
    instead we have a few machines that are supported with hardware
    and software. The use of a standard bus/interface would not have
    necessarily raised the price of MIDI, contrary to arguments I've
    seen used elsewhere.

    dave
    

    I think it was done out of ignorance rather than medacity.  I think
    they probably thought they were being cost effective, but they're
    not network guys and they're just barely digital guys.  Yeah it
    could have been better, but Monday morning quarterbacking isn't
    going to do much good, and it's going to be a while before MIDI
    is far enough behind the technological possibilities to warrant
    a new standard, especially considering the investment everyone
    (yours truly included) has made.

    len.
    

    Hey Guys I don't like to be too controversial but as a someone who
    invested not to large an amount of money in something which
    revolutionised my apporoach to composition and performance, and which
    uses connectors you can buy in any hi-fi shop. And who is also a
    musician rather than a technician "It may not be too satisfying
    as a technical solution but it works". AND WELL ! AND ALL THE TIME.
    
    er how does it go "FLAME OFF ?"
    
    				PAUL.

    I agree with Paul. Except for a few "glitches" (like when my water
    pump comes on while I'm sequencing), I've found MIDI to be pretty
    fool-proof.
    
    Does that "low" baud rate REALLY bother any of you?
    
    Edd

    As a former IEEE-488 hack I agree that would have been the way to
    go... however, the packetizing software would've hadda been burned
    into a ROM... and nobody wanted to A)fund a startup to do it, or
    B)pay (license) from somebody like Omnibyte, who really knows...
    
    Dreamtime... there are several QBus IEEE-488 interfaces on the
    market... imagine having a 31mb hard drive in your very own u11/23...
    
    KM2
    

    
    I should add that I too have substantial investment in MIDI as
    it stands. It has changed the way I write and perform music, for
    the better (I hope/think). An instrument interconnect was long overdue
    and this one does work.
    
    Even though I am not a technician, I still wonder at why the design
    was so limited. It offends my sense or software/hardware architecture,
    but that does not stop it from working for me and helping my meagre
    efforts.
    
    I did not pose my questions to overly criticise MIDI or light a
    few flames. I'd really like to see if anyone knows why it happened
    the way it did - perhaps there's a moral in the history?
    
    Robert
    

    Don't get me wrong, I've got some MIDI stuff and eventually when
    I decide what package to buy I'll drive them from my C64. I just
    think that a PROFESSIONAL would have done a PROFESSIONAL job when
    researching and designing a bus for musical instruments, instead
    we got a hack job that allows an entire industry to develop, one
    that delivers devices to overcome the limitations and shortcomings
    of the original design. If they had done it right in the first place
    (I assume they will in the second attempt) we would not have to
    buy MIDI mergers and MIDI bridges and MIDI mixers, the bus would
    have supported it. So in the long run if one was going to push the
    limit of MIDI as it now stands you will pay MORE for the devices
    to allow you to do that in real time with no errors than it would
    to do the same thing if it had been designed right in the first place.
    ie: you interface may have cost more to start with but you would
    have been spared the cost of buying all this other shit to make
    the bus function right in the first place.
    
    MIDI is here to stay for a while AS IT CURRENTLY STANDS. I'll use
    it, but that doesn't stop me from recognizing that it could have
    been cheaper for everyone if they had done the job right.
    
    I aslo resent having to buy the fix for my synth, when Roland sold
    me the machine as "Fully MIDI compatable" when it was not, due to
    their lack of understanding of the bus they helped to spec.
    
    Flame off....Ill not spout about MIDI again.
    
    dave
    

    re.-1 Does that mean that your real gripe is with the manafacturer
    or the SPEC. As far as I am concerned I now have 5 pieces of MIDI
    Kit linked by one 50 dollar 2 in 8 out MIDI connecter and they all
    work all the time and I'me delirious. What more could a simple guy
    like me ask for.
    	
    			PAUL.
    
    
    				

You've basically got it right. The people who designed it didn't realize
they were stepping into the world of computers and digital communication. It
was a radical step forward for people who were used to thinking in terms of
control voltages. It was a stroke of genius to realize that you could
time-multiplex control voltages by going to a serial as opposed to a
parallel bus.

As to the speed issue: as far as I can tell this is a red herring from the
Synclavier/Fairlight crew. In the first place, if you're just one musician
using all the bandwidth available to you through keyboard/fretboard
controllers with a breath controller and two foot controller and jiggling a
pitch wheel with your nose you aren't even close to saturating the available
channel. Where the speed gets to be an issue is when you're using a computer
to control many sound generators at once, along the idea of simulating a
symphony orchestra. In a recent article in Computer Music Journal Gareth Loy
calculated that the delays incurred due to saturating a MIDI network is on
the same order as the delay of sound propagating from a source 10 meters
away. Ten meters is not as long as the distance from the back row of violins to
the back row of contraviols in a large symphony orchestra (the Boston
Symphony for example) so we're really looking at an effect here that is all
but zero.

Don.

    re -1 MIDI delay can give a nice human, nonrobotic feel. Also there's
    an article in the latest Electronic Musician about how to use MIDI
    delay and a second synth to give some nice ambience when recording.
    
    KM

    re:.10
    
    Two seperate gripes one with the spec (hopelessly obsolete before it
    was completed) one with the manufacturer of my synth.
    
    Yeah it's simple, that doesn't mean that a better spec'd bus would
    not have been just as simple. Using IEE488 or RS423 would have opened
    up the bus to all PC's/computers not just those few that seem to
    be popular for muscians.
    
    RE: MIDI delay.....yeah maybe it's no big deal but I've read of
    people who have had to reconfigure their systems to overcome the
    problem as it was not a negligable delay, it was substantial.
                                            
    It seems hopelessly stupid to limit yourself the way the spec does.
    It also seems foolish to carry on about the way it could have been.
    
    dave
    

    People with serious delay problems can solve them by using more
    than one bus.  I too do not believe delay is a problem, especially
    since from what I understand most synths take more time to respond
    to a keyboard or MIDI event than MIDI latencies will ever amount
    to.
    
    len.
    

    
    re: .9, et. al.  
    
    An interesting bit of history... I believe that Roland DID get the
    spec correct.  Unfortunately, Yamaha did not with their DX7 but
    it became such a popular synth that the spec was changed to conform
    to Yamaha's interpretation.  Roland was a victim of "the floating
    standard".
    
    					Monty
    
    

    Roland bungled it in the beginning.
    
    I'm still waiting for my JX3-P FCO.
    
    (As a matter of fact, they did come close. It DOES power up in OMNI):^(

Am I wrong in assuming that MIDI delay is a result of the optoisolators used 
for noise immunity?