Conference napalm::commusic_v1

    Medialink article in a recent EM, as well..
    
    karl

    The similarities between MediaLink and Ethernet are kinda interesting.
    
    The device Dave refers to is very analogous to a DEMPR, only it appears
    not to need any device analogous to an H400n. (Appears to be built in.)
    These devices can be connected via a fiber-optic link analogous to the
    orange EtherHose. There's even a topological diagram that looks just
    like your basic Ethernet LAN.
    
    My initial reaction is a reserved yawn. I don't need it. Yet. And
    especially at the reported $2500 a pop price tag. The uses I see for it
    AT THIS TIME (I ain't committing to nuthin!) wouldn't even save me a
    cable.
    
    Any interest I show is purely professional, not artistic.
    
    Yet. ;^)
    
    Edd (Network Engineer)

    I read this, too, but I've gotta wonder how useful this thing is to
    most people. 

    Sure, the studio heavies will love it - and the bozos with big bux
    (like David Bryan of Bon Jovi fame) will probably use them to get even
    *more* beef out of those difficult two note chords ... but I wax
    cynical. 

    How many home doodlers really *need* this capability?  How many people
    who gig regularly *need* it?  I have yet to run out of channels,
    although I've come close.  On the other hand, I do lots of layering and
    counterpoint.  I've gotta wonder how many folks really *need* what this
    offers. 

    Want is another issue, though ...

-b

    re: .2 and .3
    
    .3> How many home doodlers *need* this capability?
    
    Anyone who uses anything like a DMC MX-8.
    
    It's true that if your stuff only stays at home, the attractive
    value of simplified cabling isn't very compelling.
    
    However, MIDI is a single-talker, multi-listener system.  There's
    lots of serious limitations in that even for the home doodler.
    
    If your only input device is ONE keyboard, your fine.
    
    Some of us however like to do our inputs from several different kinds
    of controllers (multi-keyboard setups, drum pads, sequencers, tape
    syncs, etc.).
    
    I think the single-talker limitation was an unfortunate limitation
    even if it can be got around by spending a couple hundred extra
    bucks on merges/MX-8s/etc and dealing with extra power, MIDI cables
    and additional space requirements.
    
    .3> How many people who gig regularly *need* it?
    
    I don't know how many folks "need" it, but I think nearly EVERY
    gigging person could "benefit" from it in several SIGNIFICANT ways.
    
      o	Cabling
    
    	It GREATLY simplifies cabling of any equipment rig with much
    	more than 2 devices (not even just MIDI devices)
    
    	ONE connection for EACH device gives you audio and MIDI.
    
    	Currently you have to run between one and two MIDI cables (4
    	connections) and one or more audio cables (2 connections).
    	
    	Note that:
    
    	   o You have to make sure that the cables are corrected to
    	     the right places and in the right order
    	   o You have to make sure you've got enough cables
    	   o You have to make sure that each cable is long enough
    	     to reach
    	   o You don't want to have any cable be needlessly long because
    	     unbalanced long cables increase noise
    	   o You may have to figure out WHICH cable is bad, or which
    	     is producing the noise, etc.
    
    	If Medialink reachs its full potential you only need one cable.
    	You make exactly ONE connection and there is NO WAY to make
    	an incorrect connection.  Only length requirement is that it
    	be able to reach all your instruments.  You also remove the
    	problem of noise picked up through the cables.
    
        The point is that cabling is INCREDIBLY simplified with this.
    
    	That translates to:
    
    		o Less setup time
    		o Less breakdown time
    		o Less mistakes (almost NONE) you can make setting things
    		  up
    		o Less components to fail and less modes of equipment failure
    
      o	Flexibility
    
    	It's hard to imagine that the single talker limitation of
    	MIDI isn't a limitation for a typical keyboard rig if your
    	making almost any use of MIDI in your live playing.  If
    	your keyboards aren't connected via MIDI it does you no good.
    
        But there are many typical (I believe) situations where the
    	single talker limitation is a limitation:
    
    	o Two keyboards and rack-mounted gear where you want access to
    	  the rack stuff from EITHER keyboard
    
    	o Combining sequenced stuff with live playing
    
    	o Using a system controller to control setups for each song
    
        Also note that with LANscapes, you get the equivalent of these
    	automated signal routing devices PLUS it's all done in the digital
    	domain.
    
    I mean, if you're not making much use of MIDI it's easy to yawn at this
    stuff.  I don't know about you guys, but I'm always creating layers
    between synths, using the sequencer on my SQ-80 to automatically set
    up each of the two keyboards in my rig to be connected to the right
    sounds, using a sequencer to provide various kinds of things I don't
    have enough hands to play (brass accents, background pads, latin perc,
    etc.)
    
    Are you guys using MIDI at all in your live work?  Tell me how, and
    maybe I can tell you how this would benefit you.
    
    I can't justify this for the current price tag.  In fact, I can't
    justify it until instruments have Medialink ports instead of or in
    addition to MIDI ports.
    
    What I'm saying here is that THIS is definitely a better way to do
    things than MIDI (if they can implement all the possibilities and
    do it at an affordable price).  I think that's just a matter of time.
    
    	db

Figuring this would get some discussion today, I brought in the rag.

Dave is correct when he says it gives you the versatility of multiple
controllers. From there on tho, I believe there are some significant
mistakes being made.

        1. I've yet to see any mention of audio capabilty. The open
           architecture of MidiaLink would allow it, but it's not
           yet being done. To do so would require either...

                 A. An external A/D converter and cables to convert
                    the audio out to something that MediaLink can handle.
                    Plus cables. Plus a D/A on the other end. Plus cables.
                    
                               or

                 B. Integral digital outputs. Not widely (if at all)
                    available on synths.
   	
    
You'll still need 2 MIDI cables for each unit you plan to send and receive
from, unless the manufacturers put direct MediaLink (F/O) I/Os on synths.

Each MidiTap supports 4 MIDI INs and 4 MIDI outs. These get cabled to your
gear just as you'd anticipate. 

There may be lots of powerfull ways to exploit this, but many of the items
Dave outlined as plusses just don't seem to flush... 

Edd

Re: Edd

I think Dave mentioned earlier that a lot of the stuff he wants to do isn't
implemented yet, only thought of as a possibility in the architecture.

re: all

I think the technical problems will be overcome -- I foresee the main
obstacle to acceptance of this scheme to be its "proprietaryiness".
MIDI was successful because none "owns" it in a patented or "royalty" drawing
way, so manufacturers could get on board "for free".  I don't see anywhere in
the near or not so near future of gear having "medialink" connectors as
standard -- hence you'll need to spend biggo bucks on the basics plus interfaces
and they'll still have all the cables etc.  Or "medialink" will disappear.

Now, is this is available to manufacturers for free, we may see the "second"
MIDI revolution.

Chad

    Apology in order...
    
    The article does indeed mention audio, as well as HDTV, SMPTE, computer
    data, etc.
    
    But, the future still isn't here. Yet. MediaLink won't eliminate
    all those cables. Yet.
    
    Edd

Is there a spec for this bus, or just a product?   /Mitch

    Edd,
    
    As Chad pointed out, and I've tried to mention in my replies is that
    what excites me is the potential, not the current reality.
    
    At the moment, Media link doesn't eliminate ANY cables.  In fact,
    it ADDS to the complexity.  The current value in it is for big studios
    and media labs that have lots of resources, but no way to connect
    them flexibly.
    
    Another reason why I'm excited is that I *HATE* cables and the
    complexity and problems they introduce.
    
    In fact, LANS aren't really the ideal either.  The ideal is that
    nodes communicate over the airwaves without any physical connection.
    
    Combine that with a Tesla coil for power transmission and the world
    becomes wireless!
    
    	db

    I've read your replies, Dave, and understand the potential of
    MediaLink.
    
    I see nothing in existance at this time to get excited about. Lone
    Wolf has developed a $2500 box that right now offers nothing more
    than a programmable switching/merging device. But, there are no synths
    that support the product directly (without encapsulting a standard MIDI
    message into the Medialink protocol) and digital audio I/O virtually doesn't
    exist.
    
    It's vaporware now. The technology exists to make it happen, but it's
    not available NOW. Not banking on futures and potential can hardly be
    considered "not looking at the future".
    
    When MediaLink can offer a solution to a problem, be it next week or
    3 years from now, I'll embrace it. Until then, it's an expensive
    solution to a problem that's already been cheaply solved.
    
    Edd

	Actually, since it's supposed to provide addressable use of an
	END NODE (for lack of a better term), where the END NODE can
	be using the same MIDI channels that another END NODE is using,
	I see it as quite a solution. In my case, I have an MX-8 that
	has a multitude of patches used to re-direct my set up. Most
	of my gear gets re-patched depending on what the MIDI controller
	is (I use 4 different MIDI controllers), what gets loaded with
	SYSEX stuff from my MIDI DISK drive (Thats 2 different sequencers),
	and wheather or not I'm recording or playing live. 

	MIDI is stuck at 16 channels, and I have a problem with that, now
	that I have more than one controller that shares MIDI SGU's.
	I have a need for a LAN, not a serial device that gets clogged
	with messages for too many SGU's. I've felt that MIDI was a poorly
	defined LAN for quite a while, but it sure is cost effective in it's
	current state.

	I have to agree with Dave, Unique addressability is a need of the
	near future, and this box is a step in the right direction. It's
	just too expensive right now.

								Jens

    
    	The 16 channel limitation is just easily overcome at a much
    	lower cost. Not a good argument. What is it you really NEED
    	to do, again? 4 midi controllers in? and =< 8 out? tell my
    	why a MX-8 wont work...
    
    	Im gonna take devils advocate position with Edd: This is
    	really no solution to any PROBLEM, but is a cute box possible
    	cause someone thought there might be a market, and of course
    	the technology to impliment same is no big deal.
    
    	Other than eliminate wires (and believe me, after giggin almost
    	every weekend and hooking midi AND audio AND footpedal controls,
    	I am sensitive to 'wires'), I see no immediate advantage. 
    
	not sold....but hey, for $250, um............thats different.
    
    	So maybe the debate here is what it's 'worth'???
    
    	ron
    we're talking about why it costs so much?
        
    	

    >  What is it you really NEED to do, again? 4 midi controllers in? and 
    > =< 8 out? tell my why a MX-8 wont work...
    
    Suppose I want to do 20 senders sending to 60 listeners in totally
    arbitrary configurations?  That's NOT just ALL 20 sending to ALL
    60, it's also these senders sending to these listeners,
    and those senders sending to those listeners.  And also, note
    that "these" and "those" don't have to be mutually exclusive.
    
    You can't do that sensibly with ANY amount of MX-8s.  You can do that
    TODAY with what Lone Wolf is offering (although you better hope
    they give volume discounts).
    
    It *IS* too expensive (today), and given what it does (today) there
    aren't too many applications (today) that justify that expense.
    
    But if it got cheaper, and if MIDI manufacturers picked up on it,
    you'd all want it, believe me.
    
    16 channels will probably do us home hackers on limited budgets fine
    for quite awhile, but even my humble system can easily listen on about 48
    channels.  I already have all my channels allocated out, and if the
    current allocation didn't suit a particular tune, I'd have to
    reconfigure (either via programming or recabling).
    
    In fact, if I had limitless channels, I would almost certainly have
    EACH drum on a separate channel so that I could do the entire drum
    mix dynamically via MIDI volume. 
    
    	db

    > If I had limitless channels...
    
    With the lone wolf set-up you'll not only have limitless channels,
    they can all be the same one. You simply address the port on the tap...
    
    Edd

    
    I understand THAT, Dave, but your key phrase is 'arbitrary
    configurations', NOT the number of senders and recievers.
    What is accomplished. You can matrix MX-8's and do 20 xmit and
    60 rcv. But you dont say what real-world application NEEDS
    'arbitrary configurations'.
    
    I maintain that TODAY you have a fairly statically defined
    configuration, er, network, and it doesnt change that often. 
    Nor does the number of senders and recievers, in fact. So WHAT
    exactly is it that you think you'll be able to DO with midi-LAN
    either now, or in the future?????. PERSONALLY. Musically.
    
    Ok, If you're a university and have x instructors tied to y music
    labs...ok, you MIGHT want a net (UM, BUT FOR WHAT, THO? GANG teaching?)
    
    In this situation, you personally would not have control of the 
    entire net. Dont expect to play the "girl-next-lab-over's-gear" with
    some love song from YOUR 'sender' 'sequencer-system'. She aint gonna
    like you much if she's in the middle of her composition final project,
    even if you can BREAK the protection code scheme that will be
    necessary, ( and not even alluded to) in a midi-lan with multiple
    users....See? you're stuck again with a limited configuration. 

    What is the useful real-world (even experimental) situation that
    warrants the cost of this hi-bandwidth SINGLE cable system?
    
    lucifer
        
    
    The situation is clearly this: 
    
    	A
    
    	Point is, of course, that some net-management application is
    	in control
    (altho Lone implimentation implies listeners-only ( ex: any SGU )
     would have to impliment transmit functionality in order to
    alert senders of transmission or other ERROR situations)

    oops..........er, the situation is clearly this:
    
    	{the following didnt get editted/deleted during session
    	 as it should have}
    
    har har harhrhrhr....ok lets talk Network level error detection......
    
    rr
    

One thing you will have is an unlimited amount of virtual MIDI cables and 
therefore channels (in some unspecified MIDI LAN spec).  That means, even with
old normal MIDI gear hooked into a tap, you could set up a virtual MIDI cable
with 16 channels directed at that machine, with another vrtual cable at another,
etc.  We could still use MIDI perfectly well as the protocol (the software part
that is, not the hardware spec) implemented on the LAN protocol.   Machines
that had the direct link would have then some aribitray n MIDI ins and OUTS
defined by the manufacturer implemented as virtual ins or outs on the LAN.
All of our existing software etc. would be compatible.

Chad
thinking_while_typing

    > What would you do with MIDI-LAN now or in the future.
    Ron,
    
    First: SET CONTEXT=FUTURE
    
    I'm talking about an reasonably priced LAN where each device plugs
    directly into the cable and handles audio, video, MIDI, control, etc.
    
    Ok with that context established:
    
    With Medialink you can not only have any network arrangement but you
    can define those networks easily and dynamically switch between
    configurations easily.  
    
    You can't do that with umpteen MX-8's.  Not easily, and not
    practically.
    
    As far as how I would use it, I would use it to get rid of essentially
    ALL of my audio, MIDI and control cables for starters and never have
    to recable anything for any reason.
    
    I'd also use it to get around the 16 channel limitation.  Yes, you
    can do that with an MX-8, but I find that to be a pain: more equipment,
    more cables, more setup, more breakdown, more programming, more
    modes of failure.
    
    With a true MIDI-LAN you have something that isn't significantly more
    complicated than MIDI-but-with-lots-more-channels.
    
    	db