You are here

Where from here?

For performing musicians and engineers: stagecraft, engineering and gear.

Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 12:05 pm

Hi,

I am looking for input from the hive mind, so I asked Hugh if it was ok to post, and he gave the go-ahead.

For the last 12 years we have been developing "Architectureless Audio", which consists of "connections" which are used to join "functions" in any order to make it possible for one piece of hardware to perform any existing design, but more importantly to make it possible to "roll your own".

A conventional solution to a problem uses a subset of the same functions in a specific order to be, say, a mixing desk or a synthesiser. The way the functions are connected could be described as the architecture - hence "Architectureless Audio"

Functions are easily recogniseable, like Bi-Quad filters, multipliers, summers etc and their parameters can be connected in any order. As an example, a voltage controlled filter in a synthesizer controls the frequency - here you can also control the Q and gain with either controls, other functions or with audio. Similarly, in a delay the tap positions can be controlled either by user input or some other function or audio.

To be clear, there are only either inputs or outputs and outputs can only be connected to inputs - it doesn't care which input or which output, as far as it is concerned the output of a function is the same as the output of a control or the input from an audio source.

The whole thing happens "Live" - there is no compilation time - allowing one to hear the changes as they are made.

The number of connections / functions is mostly limited to the amount of hardware as the whole thing is modular. To give some sort of idea a very modest system would have around 4000 connections and a medium sized (before the hardware gets large) would have in the order of 100,000 connections.

So, for example, one can build an 88 note keyboard type instrument where each note is fed into one channel of an 88 channel mixer that has surround sound output on each channel so each note's spatial position is governed by - anything the musician wants.

But we don't yet have a user interface that any sane person would want to operate. At the moment we type in co-ordinates to make the connections which is about as user unfriendly as it gets :crazy:

So I would like to ask anyone who has any ideas as to thoughts on where to go from here?

One idea which we like a lot is to make two user interfaces, one to do the Architecturing with and one to present the user controllable parameters - and to make both of them operate from a web interface - to make it super accessible. It's likely that lots of people would want to work on it at the same time. Of course, user controls can also be in the form of physical things, swipe, poke, twang, flick, turn, slide etc.

But just because we like it doesn't mean it's right.

Whatever the answer is, it will be complicated and take a while to do so it makes sense to get as many ideas as possible before that commitment is made.


All suggestions gratefully received
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby The Elf » Tue Jun 16, 2020 12:36 pm

I read all the words, but I don't feel much wiser! :lol:

What are you hoping will come out of this? A piece of hardware? A piece of software?
User avatar
The Elf
Jedi Poster
Posts: 14553
Joined: Mon Aug 13, 2001 11:00 pm
Location: Sheffield, UK
An Eagle for an Emperor, A Kestrel for a Knave.

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 12:55 pm

Good question Elf,

I was hoping it would be like the "Leap of Faith" moment for Indiana Jones, faced with an abyss some sand might get thrown out to reveal a camouflaged way forwards.

I've imagined quite a few scenarios myself, like music and dance becoming part of the same thing where the movement of a dancer changes the music. That's a kind of user interface.

But here there seems to be two distinct things to work on, how stuff is connected together and how it gets controlled. The dancers would be doing the controlling whereas the "Architects" would be changing how the structure of what the dancers can do changes the music.

You are quite right, there are a lot of words!
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby CS70 » Tue Jun 16, 2020 1:40 pm

Is it a commercial venture? Or the basis for one?

I do not understand the whole thing, but from what I get you want to be able to mesh together pieces of hardware (and software perhaps?) in any order and get some sound processing out of it? I suppose you must still have some classes - like sound generators, sound emitters etc.

If hardware is involved, do you see a "sandbox" environment where you have software emulations of the hardware and you get test specific designs?

You could look at user-oriented game building software where you can both build the game and play it or let others play it.

But yes two specific applications seem like the way to go. A combined one which acts both as a development environment and runtime sounds very Smalltalk and nowadays is not a great design imho.
User avatar
CS70
Jedi Poster
Posts: 5567
Joined: Mon Nov 26, 2012 12:00 am
Location: Oslo, Norway
Silver Spoon - Check out our latest video and the FB page

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 2:11 pm

More good questions CS70

Ultimately it will be commercial, but at the moment it is something we fund ourselves "because no-one else is mad enough to try". If people knew how long something was going to take before they set off, they would never set off!

The audio part is HDL (hardware description language) rather than software, because no matter how many DSPs you throw at the problem there just isn't enough affordable processing power. So the connection and functionality happens inside lumps of silicon.

When I said it was modular, it is, but the idea isn't necessarily to have lots of separate units connected together - one unit can have 100,000 "connections" of its own was what I should have put; and then if you need to you can connect it to another unit.

The sandbox you mentioned might now make sense in that it is possible to just add your new bit of architecture, send it to your headphones, then mix it into wherever you wanted it to go when happy with it; all inside the same lump of silicon.

Your idea of using gaming software sounds really interesting. The only thing that would be provided are the "functions" which have inputs and outputs, so the user can make their own higher level functions from those. It would certainly feel like a game!

It's interesting that you also like the two applications approach. After I posted it I wondered if a dancer might want to change the architecture - the ultimate in jamming..
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby BJG145 » Tue Jun 16, 2020 3:40 pm

where_theres_a_will_theres_a_way wrote:After I posted it I wondered if a dancer might want to change the architecture - the ultimate in jamming.

I think the kind of (inside-out/outside-in) motion tracking being developed for AR/VR is the way forward for dance. The only other tech promoted specifically for dance control of MIDI in the past has been under-specced and overpriced (eg Soundbeam and Mi.Mu.)

(The idea of mixing music and control messages reminds me of Belcanto. The Eigenharp developers imagined being able to play or reprogram the instrument using the same controls. It's all open-source now; interesting, but complicated.)
User avatar
BJG145
Jedi Poster
Posts: 4312
Joined: Fri Aug 05, 2005 11:00 pm

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 4:16 pm

Interesting, the Eigenharp

There's also the Reactable which is of a similar ilk except all the functions (pucks) are pre programmed and there's a real-estate issue with the table if you had a few thousand pucks. User input comes in the form of moving the pucks, but so does (in part) the designing. It's a really neat piece of engineering.

One of the things that happens once you have large complexity is you have to be able to "fold" functionality inside something more manageable (like a puck). Being able to "open up" a folded function and modify it is going to be a must.

A virtual reality user interface would be... complicated... but does give total access, including to dancers, of both the control and design. If only the audience could see what they could see!
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby BJG145 » Tue Jun 16, 2020 4:34 pm

where_theres_a_will_theres_a_way wrote:If only the audience could see what they could see!

Yes...I've been thinking it would be cool to get hold of a Hololens 2 and summon up a virtual piano, but from an audience perspective it's not so exciting. To give Imogen Heap her due, she's created some great performances with Mi.Mu,

As an out-there dance/music performance instrument, I also like Karlax...not sure if they're still in production.
User avatar
BJG145
Jedi Poster
Posts: 4312
Joined: Fri Aug 05, 2005 11:00 pm

Re: Where from here?

Postby Eddy Deegan » Tue Jun 16, 2020 4:36 pm

It sounds an interesting idea, but having read through the whole thread I still have almost no idea what it is.

Obvious questions to me are:

- How is it different to a modular synth and the Eurorack scene?
- What can it do that existing solutions can't?
- What problem is it seeking to solve?
- Is it a kind of 'hardware audio lego'?
- If it's hardware what purpose does a user interface serve, and how does that interoperate with the hardware in a practical sense?
- Will it be something end users can actually use, or more a 'construction kit' for hardware manufacturers?

You mention a large number of connections which infers that there is a difference between it and Modular/Eurorack systems, as the number of connections in those is much lower but I'm still very, very confused as to exactly what you're talking about.

I would be interested to get some clarity but at the moment it's very abstract, there are too many words and not enough explanation :)
User avatar
Eddy Deegan
Moderator
Posts: 4354
Joined: Tue Aug 31, 2004 11:00 pm
Location: Brighton & Hove, UK
Some of my works.
Please consider supporting the SOS Forum Album project.
 

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 4:40 pm

I was looking at "if only the audience" and thought - why not?

Why couldn't it be that the audience wear the VR as well, and be able to see the performance from any perspective?

That would give "Live Music" a whole new meaning - no two performances would be the same, ever!
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby BJG145 » Tue Jun 16, 2020 4:43 pm

where_theres_a_will_theres_a_way wrote:I was looking at "if only the audience" and thought - why not?

Why couldn't it be that the audience wear the VR as well, and be able to see the performance from any perspective?

That would give "Live Music" a whole new meaning - no two performances would be the same, ever!

That's a nice idea; I wonder if it's actually been done yet...a live performance delivered by an artist and watched by the audience in VR. It's definitely got a long way to go.
User avatar
BJG145
Jedi Poster
Posts: 4312
Joined: Fri Aug 05, 2005 11:00 pm

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 4:58 pm

Great questions, thank you

- How is it different to a modular synth and the Eurorack scene?
It isn't, just in its scale and that the connections aren't cables - there are also functions available you don't get in a synth

- What can it do that existing solutions can't?
There is no solution out there that has no architecture - so it can do all of the things that are out there and put the design in the hands of the owner.

- What problem is it seeking to solve?
The main problem is that existing solutions are "optimised" so when some new fandango comes out you have to change the whole thing. The new fandango is simply some functions joined together with connections in a new way.

- Is it a kind of 'hardware audio lego'?
The main problem in describing it is that there isn't anything to compare against. The physical hardware is lumps of silicon (like CPUs in a computer but not CPUs because they are too slow) but in a practical sense it is like having a huge modular synth.

- If it's hardware what purpose does a user interface serve, and how does that interoperate with the hardware in a practical sense?
The user interface (whatever it ends up being) will be similar to how one effectively plugs in and out cables in a modular synth - just that at this scale there isn't room for a physical represention. Imagine the modular Moog being 1mm square, this is more like 6 Metres square at the same scale. No-one has arms long enough!

- Will it be something end users can actually use, or more a 'construction kit' for hardware manufacturers?
Definitely for end users.

Hope that helps..
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby where_theres_a_will_theres_a_way » Tue Jun 16, 2020 5:03 pm

BJG145 wrote:
where_theres_a_will_theres_a_way wrote:I was looking at "if only the audience" and thought - why not?

Why couldn't it be that the audience wear the VR as well, and be able to see the performance from any perspective?

That would give "Live Music" a whole new meaning - no two performances would be the same, ever!

That's a nice idea; I wonder if it's actually been done yet...a live performance delivered by an artist and watched by the audience in VR. It's definitely got a long way to go.

You could also have "Scene designers" working on the landscape of the performance that the architecture resides in. Who thought machines would take away all the jobs!
where_theres_a_will_theres_a_way
Regular
Posts: 53
Joined: Mon Feb 19, 2007 12:00 am

Re: Where from here?

Postby RichardT » Tue Jun 16, 2020 7:50 pm

What’s the advantage of this over, say, Max or Reaktor Blocks?
RichardT
Poster
Posts: 41
Joined: Thu Aug 12, 2004 11:00 pm
Location: London UK
My last album sold over a million copies - if you don't believe me, come and look in my garage.

Re: Where from here?

Postby James Perrett » Tue Jun 16, 2020 8:22 pm

where_theres_a_will_theres_a_way wrote:The main problem in describing it is that there isn't anything to compare against. The physical hardware is lumps of silicon (like CPUs in a computer but not CPUs because they are too slow) but in a practical sense it is like having a huge modular synth.

So basically a bunch of DSP chips running in parallel with standard software modules already written and a way of hooking all the software modules together?
User avatar
James Perrett
Moderator
Posts: 9363
Joined: Sun Sep 09, 2001 11:00 pm
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Next