What do you want in new equipment?

[mosc]

From another topic, Dr. Justice said:

[elektro80]

Firewire does support networking, and so does MLAN to a certain extent. MLAN can also be used over say GIGIABIT ethernet, but Yamaha has not published any on this yet.

Firewire/MLAN is good enough for most purposes and far better than any alternatves right now. Steinberg did have some DAW sync ( VST Link ) tech over ethernet though, and I reckon Yamaha will study that closely.

[mosc]

Well, it would seem if you want to network musical equipment, you should use IP so you can stand on all the hubs and routers that are already out there and working so well. Firewire is not really suitable for this. It was designed to connect audio and video devices to each other and computers.

If I'm going to have the gear in my studio networked together, then why wouldn't I want to tunnel out over the internet and link to your studio? We aren't going to do that with Firewire.

I want to network and play with people all over the world simultaniously.

[elektro80]

Actually, IP over Firewire is working great. Apple is into this bigtime.

[DrJustice]

Hey elektro80, you really do believe in Firewire, dont'you Not to worry, thats' cool. I'm sure MLAN works very nicely within an MLAN environment. And Firewire is an OK local interface. But whichever way you look at it, it is not and will not become anything like ethernet in many important aspects.

Some points about of Ethernet are:
- open and free to use, as in no secrets an no licences
- well developed and stable, will 'never' go away or change much
- supports all the internet potocols out of the box
- supports OSC out of the box
- well known, cheap and easy to implement
- global plug&play infrastructure already in place
- local latency less than a milli second
- bandwidth of 1GB/sec
- cheap and cheerful equipment and global feeds

The fact that ethernet is a collision detection protocol by design, was an issue once, but now everything is switched. Appropriate synchronization is defined in OSC. I think ethernet is more than fit for the fight.

As for transporting IP over Firewire; back in the olden days I actually made an IP over MIDI transport that was being used for internet access - Which goes to prove nothing really...

DJ
--

[mosc]

[jksuperstar]

I would prefer ethernet. I find local 100-T just fine for moving/syncing MIDI with other laptops. I can't comment on audio latencies, though (ie- what total latencies are there, realistically, from one PC to another and back again? faster or slower than typical FireWire audio interface cards? (10ms))

I've been experimenting with midi-yoke and netmidid/c from "midi-utilities-for-windows". Not much to report yet, but once it's running, it's pretty nice. I use a switch on my network, which is VERY much worth the $25 they cost nowadays.

It seems, for once, that Linux is ahead of Windows in the Midi over Lan department in terms of tools available and the quality of those tools. I've had some difficulties getting my MIDI interface to work on Linux, so I started playing with NetMIDI protocols instead as a quick solution. (Need to get my arms around OSC still, since I haven't found a simple MIDI-over-OSC-over-LAN server that runs on windows)

[DrJustice]

[zynthetix]

It would be interesting to see OSC become a standardized similar to MIDI. I have seen OSC incorporated similarily to MIDI in Reaktor or Max msp, and a few very neat OSC controllers. The thing I'd like to see most in future gear though is new features AND retaining of backwards compatibility (such as MIDI and CV) so old gear can still be utilized in new ways.

[mosc]

Yes, and nobody has mentioned wireless. There are many very good low cost wireless network components. 50 Mbs may not be enough for audio (it isn't) but it's pretty fine for MIDI.

[DrJustice]

zynthetix:
Yes, it's essential to keep the traditional interfaces, like you mention. A future with just the one digital wire is a bleak one indeed.

And mosc:
Of course! Wireless! Hadn't really thought of that. What's the price of a bluethooth chip now? - Probably a couple of dollars or something. You could be online MIDI jamming through your cellphone with no wires attached

We (who are we...? They?) ought to work on this I guess.

So when we're finally well connected, what's next - Tought control? Or are knobs and wooden end panels eternal? GUIs, knobs, blinking lights and the feel of a an instrument truly merged? Do we want racks, keyboards, tabletops or? Do we want new synthesis methods or instrument types? Will anyone ever bother to read an LCD again? Hmm.... the questions are many.

DJ
--

[mosc]

Telepathic links to equipment would be very nice, but if we could get thought links to the equipment, we could interconnect directly with the audience.

Here I think we'd be in a funny place. You'd have to be a musician that can think music. I personally enjoy interacting with the instrument to make sounds I can't think of. If I could think music that was as interesting as I can play on a good synth, then I'd probably give up music and just spend my time thinking.

[zynthetix]

[DrJustice]

Just some quick numbers:

Take wireless ethernet at 50Mbs. Lets say that we settle for half of that bandwidth to guarantee a continous stream, that we weant 24 bit audio at 48kHz and that the software works:

50Mbit / 2 / 24bit / 48kHz
= 21 channels

That's not too bad. And theres faster wireless kit available.

DJ
--

[Kassen]

I´m greatly in favour of OSC, MIDI is way too limited. What´s particularly bothering me is that in MIDI you only get velocity and polyphonic aftertouch as ways to adress individual notes. Of those velovity can´t be changed once the note is sounding. Worse yet; nobody seems to realise that this is a tremendous problem and limitation and polyphonic aftertouch is hardly ever implemented. This leads to endless amounts of hoop-jumping.

I also desperatly need ways of communicating between voices of a single instrument.

And while the world is at it, I´d like it to establish a devide between the interface to sounds and the way of changing those sounds.

Tunings should be a part of the song´s settings, defined in the sequencer, not a part of the instrument´s settings.

In general we are stuck with a paradigm based on the piano, that´s fine if you want to play the piano or use other sounds with a piano interface (to the point where people are making velocity sensitive harpsichord presets(!!!!)) but it´s very limited by modern standards. We desperatly need to get rid of this as soon as possible.

I read on the Ableton board that Live is not going to get OSC within this year, I´m kinda sad and disapointed about this.

[Kassen]

Oh, and granular synthesis and re-synthesis in a hardware box, why is this taking so long?

[jksuperstar]

[mosc]

[Kassen]

Yeah and it seems to need another 200 for the doughterboard for the interface. It does look like it can also do video, that´d be rather exciting, I think. 4 in 6 out is also pritty nice. Programing it all up would be rather time consiming though. There´s the risk that by the time you´re done there´ll be a commercial implementation.

[jksuperstar]

Those kits come with libraries that set up audio DMAs, video DMAs, etc. It's an incredibly efficient DSP. So, really, I just need to port some code over from a few open-sourced libraries, then embellish. I've already taken the RS-232 port, and added a MIDI translator. Linux has been ported to those boards, but I'd rather avoid the whole OS thing altogether, and just make it a fast stand-alone device.

Anyway, once it's working, I'd consider making my own board. Maybe it could be sold as a pre-stuffed PCB kit. (The 600MHz Blackfin is ~$5 in 1000 count lots). It's the logistics of getting all the other random parts that really takes time.

Hmmmm....

[DrJustice]

That's good going jksuperstar!!!

Designing the future, not just talking about it Glad to hear of yet another project brewing, commercial or not. I look forward to hearing more in the developers corner. Got some AD DSP kits here too, btw...

And Kassen,
What do you mean by "communicating between voices of a single instrument"? I can think of some uses, but please explain what you want from it.

DJ
--

[Kassen]

well, I´m particularly interested in synthetic versions of sympathetic resonance for organic and intuitive modulations but I´ve ran into this matter in more fields. It would be realy nice for making your own mutated arpegiator and I´m also interested in issues of timing the various sections of one note in relation to other notes. I think we can do better then just having one lfo for all voices or a seperate independant lfo per voice, we should be able to have those compliment eachother in various defined ways.

Lately I keep runing into this matter as in the various systems I use voices can´t "see eachother" without lots and lots of trick or ensembleery. It´s simply a brick wall that need not be there. After all; in a orchestra the different musicians are hopefully listening to eachother.

[trash]

i would like to see..
a frickin program that writes the tracks for me
to my own exacting specifications!

[paul e.]

i thought that was Garageband hehehe

[jksuperstar]