where next with software instruments?

[aciddose]

i was trying to point out hypocrisies, btw. bones is one to constantly rant on about how wonderful software and DAWs have made his life as a "musician" while a lot of the rest of us are thinking of him as a total loser for not being happy to deal with the analog instruments of the past. no patches, no projects being saved, no unlimited amounts of multi-tracking capability. lots of noise, lots of cables, great expense in terms of time and money. constant aggravation and turmoil, unending tortuous effort required to produce even the most simple of compositions.

this is exactly the same thing as bones crying out against "auto-magic" composition tools, or their results.

WE think that the results of software and the software work flow are equally as lame.

[eduardo_b]

There's already more than enough technology. There aren't enough customers who know how to use it well.

"already more than enough technology" for what?

For making music. All it requires is talent, skill and ability. Technology doesn't make great art by itself anyway. It is the user who takes the tools and does what others haven't, can't or won't. If the problem is a lack of technology, then all those people using traditional musical instruments should have long ago given up trying to create quality music. They are somehow muddling along with rather ancient technology. Really, more music technology is not the answer. It just adds to the problem of unrealistic expectations -- that machines should be the artist while humans take credit. It will be as mediocre as most of what those using current technology create, produce or invent.

[aaastronomer]

Interesting dicussion
I think it would be really interesting to do something based on statistical analysis of a large database of presets - ideally with some form of attribute tagging. I think the approach could be a great way to generate new sounds. Hopefully I'll find some time to spend on this in the future. I think it would be a cool thing to do as research MSc or PhD these (if by some chance someone reading this is interested doing something along these lines, please get in touch as I'd be happy to help.)

an alternative (non-parametric) approach, if you have a big database of patches, is cluster analysis. i happen across papers in this area all of the time. you can also start from a small database of presets and generate presets using genetic algorithms. the generated presets would be compared to presets in various categories that represent 'the best' sounds produced by a particular sound engine. however, since you don't want to just create copies of what you already have, what criteria do you use to stop generation short of the known best (or some ideal statistical model)? how different can a good preset be from a known best? in fact, what sort of 'preset metrics' would be used?

[gnu23]

Speed? RTP-MIDI.

Semitone-based scale? MTS.

Horribly limited? HD-MIDI (not on the standards track yet, but the MMA seems to like it)

so MIDI, as it currently exists, does need replaced then. thanks.

No, you completely missed it. Sorry. The protocol can be extended, e.g. improved upon, not replaced.

MIDI hasn't been replaced because there isn't a pressing need to replace it -

and thanks for agreeing that its not because there's nothing better.

Again, your assertion is incorrect. Nowhere did I say any alternative was better. Nor did I say any alternative was worse. What I did say is taken at least partially out of context in your quote, but my reply is not that far back in the thread so it is easy to recapture that context without re-quoting here.

The point is that MIDI is an open standard with 25 years of good service. Protocols like MIDI can be extended or enhanced to reflect improvements in the physical devices used to transport them and in future hardware that will support those improvements. MIDI has 8-bit resolutions in some areas now only because computers and synthesizers had 8-bit resolutions in 1983. We are in a 64-bit multi-core world now, so there's no reason why MIDI can't be extended to support at least 32-bit values while retaining support for legacy applications that retain 8-bit word lengths.

No I can't think of anything better than MIDI. I didn't say, however, that MIDI should remain a static standard.

Ah. So as long as its retrofitted on top of the existing old serial MIDI with all its limitations, then its fine, and its still MIDI so it doesnt need replaced, but god forbid that the need for those 'enhancements' should indicate that MIDI itself needs rethought. That makes sense.

There is a significant difference between "retrofit" and "extension". And serial transport is a physical layer convention. Yamaha's mLAN is a physical layer improvement over serial, and it supports MIDI. HD-MIDI is a modernization of MIDI which also adds support for new channels and message types in addition to physical layer improvements. mLAN and HD-MIDI are improvements, not replacements.

OSC is a replacement. While it may be that OSC is as good as it is billed, the fact that it is a wholesale replacement of an existing universally-adopted-and-implemented standard makes it very difficult to for OSC to gain traction. Again, I raise the point of retrofitting legacy apps to OSC. While it certainly could be done, I'd bet good money that it won't be done simply because MIDI is extensible, but more importantly deeply entrenched.

Closing out my posts on this topic, I don't have a dog in this hunt. I am not a member of the MMA nor am I in the music software industry. MIDI has served me well for the 23 years I have used it. I have yet to be restricted by it and don't anticipate problems with it for the forseeable future. But I have been in my particular industry niche for 30 years and have learned that standards with durability are the ones that allow room for the future. MIDI is such a standard. Extend it? Sure. Replace it? No.

I understand your position on this issue but I do not agree with it. Obviously, you don't agree with me. But I wasn't trying to start a debate with anyone. I was simply posting an opinion. Your impersonal point of view would be just as welcome to me. However, you have taken part of what I have posted out of context to bolster your own argument and to employ your well-known swipes at people you don't like. I'm not the first person you've pulled this stunt on and I won't be the last. Personally I could not care less. My ego is not nearly that fragile. It is your neck that resembles a foreskin here, not mine. So hey, good luck with that. Better you than me.

[aciddose]

while you certainly could "extend" midi, it would cease to be midi at that point. the existing strategy for controllers and the different sorts of commands are absolutely obsolete. doubling for example system-commands like "allnotesoff" in the space of controllers is rediculous, absolutely rediculous. the limitation on the number of controllers and the use of high/low words for controllers is likewise rediculous.

if you take apart the standard to "extend" it in every one of it's parts leaving really nothing in common with the original standard, that's "replacement", not "extension", wouldn't you agree?

no part of the midi standard as it stands would function acceptably in a modern "extended" version. you might call it "midi 2.0", but it would really be something entirely different.

you mention backward compatibility, yet translating messages internal to a sequencer into the midi format is no more or less difficult than translation from any one standard to any other, such as OSC to MIDI. the problem with the OSC standard is that OSC is ONLY a transport and protocol standard, it is NOT a message format standard. OSC does not define controllers, events or anything else of that nature, it merely provides a transport upon which to base the definition of these things.

for example, midi could be transported across OSC.

not to mention, the OSC namespaces transport definition is terrible in my opinion. i've also read about how great OSC is and how it should "change the world". i have no idea what the authors of these articles are smoking, but it's pretty good (or bad) stuff.

[aaastronomer]

...not being happy to deal with the analog instruments of the past. no patches, no projects being saved, no unlimited amounts of multi-tracking capability. lots of noise, lots of cables, great expense in terms of time and money. constant aggravation and turmoil, unending tortuous effort required to produce even the most simple of compositions.

not for me, ever again. i've bagged my cables and racked my h/w and put my kbs on a stand. it's all nice to look at, but my studio's on a laptop (with the exception of some outboard mix enhancement h/w) and it'll stay that way. i must admit, the prophet 08 and the g-synth are tempting, but all their sounds would just end up digitized.

[eduardo_b]

as nothing is perfect, everything needs improvement! technology merely raises the bar for quality. there was lots of mediocre 19th century piano music (which you find out if you undertake a traditional course of piano study). going from candle light to gas light probably didn't change that situation from the 18th century. i would think that even the most accomplished users of a given technology expect (and can point out) what improvements need to be made of a technology. a tool is just a tool. however, it is obvious a tool can increase productivity and precision in the hands of a skilled user of the tool.

All well and good, but the emphasis is on the tools, not the processes of creation, of artistic growth, of expression. Just listen to the uneven, fairly good to utterly horrible music being produced by this technology in the hands of anyone and everyone. Some even produce an "album" of the stuff, one or two tracks of some artistic interest surrounded by rapidly descending awfulness.

Listen to the same technology in the hands of the truly gifted and it's a completely new experience. No amount of technology ultimately can change that critical component. Another poster noted the progress in animation technology and tools, but without compelling story lines and characters it's all meaningless. Shiny but boring...or worse.

[aciddose]

aaastronomer; and such a choice is entirely respectable, it's the way of the future. when bones rants about this choice it makes me feel like an old man yelling:

"you just can't handle it sonny. why back in my day i had 439 cables measuring in total some 2/3rds of a kilometer hanging off four walls of modules. the room had no door and we had to use a rope and pully system to get in and out through a hole in the ceiling that was made when one of the 400watt diesel generators blew up. ...and get off my lawn!"

the fact remains that some of the best existing compositions were written at a pianoforte, harpsichord or something similar hundreds of years ago on paper with a quill and ink.

bones just doesn't seem to have any clue he's one more step, step #492 between bach and some 12 year old kid pressing one button and writing a hit to make millions some time in the future.

[BONES]

I think strong storytelling and characterization will always be the most important but the better tools we have now make it possible for us to make more fully fleshed out and dynamic characters and convincing sets. Of course, this is partly due to the amount of raw power we now have available for rendering but also owes a lot to the much more powerful tools we have available.

But most of that technology has been around for a long time, it just never got put in until the hardware could make it usable. e.g. When I first started using 3D Studio Max 1.2, there was no way I could ever have justified the render time involved in using raytraced shadows, so I always rendered shadow-maps. It wasn't until I got onto a quad-core SGI box in 2000 that raytraced shadows and reflection became viable. Now we can do them real-time in the viewport, so they are pretty much the default option. Radiosity and global illumination were also around a long time before they became standardised in the major 3D packages but if you wanted to use them, the technology was out there in 3rd party renderers. You just had to be willing to make the time sacrifices inevitably involved.
Ultimately though, the stuff we are using now has evolved from simple beginnings. An excellent example is teh UVW unwrap stuff in Max. It was put into v2.0 with a limited toolset and no-one really understood the power it gave artists. It stayed there, largely ignored, until v6 or so, when someone worked out that with a few extra tools it could be all-conquering. Now it is pretty much the default texture-mapping tool in Max, which has put a lot of 3rd party plugins out to pasture.

music is and always has been a collection of styles... to not recognize this you must has missed elementary education somewhere!

Possibly but what sort of loser woudl want to bind themselves to any of them in such an arbitrary way? I mean, we all do it inside our heads, which is part of the spark of inspiration, but to reduce it to a mechanical process is frankly appalling, kind of like those $50 DJ applications you see in computer stores.

of course people want machines to do ever more work for them. this is what is called 'progress' in western culture. (i assume you'd prefer to be sawing on some cat gut wood box somewhere.)

That would actually involve a lot less work that using my current tools.

what is 'process' today will not be 'process' in, say, ten years. for example, it use to be 'process' to write out orchestral score by hand and then have it copied out to individual parts by copiest. now, of course, you just use notation software. it is also now possible to send the parts to the individual players on stage by wireless intranet.

Which are all matters of refinement - you are still creating the work and distributing it to others in the same format - and ultimately completely unrelated to the creative process.

of course, the line between 'composer' and compositional software does get blurred, but who's afraid of a little ambiguity of attribution?

It all comes down to the value of the work. Do you treat your cover versions the same as your original songs? It seems like the same kind of thing to me - if its written by a machine, I can't see how I would have the same emotional investment in it just because I pressed the start button. A good analogy would be a car with a manual gearbox or an automatic transmission. Its actually a really good one because we are starting to see automated manual gearboxes, which are capable of changing gear faster than any race driver and deliver greater fuel economy, but most driving enthusiasts will still tell you that the whole point is the experience and skill of changing gears yourself. i.e. Ultimately it is worth losing some performance to retain the essential enjoyment of the process of driving. So the people who end up embracing this technology are the bogans and boy-racers, who just want to have the latest toys or be the fastest at any cost, while the rest of us are much happier striving for that perfect gear-change and accepting that we might not be quite as fast.
Democratising music is simply to devalue it. Listening to the radio, it might seem that it has very little value now but I have no doubt it can sink a lot further, or at the very least be diluted a lot more.

"already more than enough technology" for what?

To create any piece of music that you can possibly imagine. You may not be able to do it in exactly the way that you want to but that is your limitation, not the technology's. Every sound you have ever heard can be reproduced faithfully and put together in any way that anyone would like to. On top of that, we can make sounds that had never been heard before synthesisers and if you look closely, you'll realise that it has not really improved much in the last 20 years, its just become more accesssible and more powerful.

[pdxindy]

OT

Look at what's been done in computer animated films in the last ten years. There's no question that giving people more powerful and expressive tools has made their art more expressive too.

i would suggest that the big improvement there, and its the industry I work in, is rendering. It is now possible to make stuff look a whole lot better than it did 10 years ago but I would argue that Toy Story was every bit as good a film as anything recent. 10 years ago it might take 2 hours to render each frame of a feature [x 24 frames per second x however many seconds the film runs], so things like global illumination, fur, etc had to be compromised in order to get the job finished on time. These days all that stuff can be included and rendering times will still be around one-tenth of what they were 10 years ago. i.e. It is the hardware that has facilitated improvements, rather than anything gee-whizz in the software. Most of us are still working in much the same way we did back then.

The software is vastly better today than 10 years ago. For example, with modeling, there was nothing at all like Mudbox or Zbrush back then. That is some gee whiz software and has allowed modelers to truly sculpt in 3d.

[aciddose]

i wouldn't even say the most impressive advancement in cganimation is modeling or rendering - i'd say the motion capture and animation systems of today are extremely, extremely improved against those used when animations like "toy story" were built. today it's possible to connect sensors to a real actor and have all physical properties of their bodies and clothing accurately modeled by the system while oversampled motion capture is done providing for accurate motion-blur and other effects.

the modern quality is like the difference between a bumpmap textured phong shader and a flat shader.

[aaastronomer]

There's already more than enough technology. There aren't enough customers who know how to use it well.

"already more than enough technology" for what?

For making music. All it requires is talent, skill and ability. Technology doesn't make great art by itself anyway. It is the user who takes the tools and does what others haven't, can't or won't. If the problem is a lack of technology, then all those people using traditional musical instruments should have long ago given up trying to create quality music. They are somehow muddling along with rather ancient technology. Really, more music technology is not the answer. It just adds to the problem of unrealistic expectations -- that machines should be the artist while humans take credit. It will be as mediocre as most of what those using current technology create, produce or invent.

well...yes, making music requires "talent, skill and ability", but what we are talking about here (i think) is the palette with which you express your talent, skill and ability. i'd hate to have to listen to just clarinet all the time, even if all the clarinetist were virtuosos! the synthesizer and the sampler push the cochlear to the boundaries of its capabilities, opening up vastly greater perceptive horizons than were available with traditional instruments.

traditional musicians trying to create quality music (in their own musical sphere) has nothing to do with the tech-race going on the electronic music sphere. of course, new technology gives traditional musicians new strings, reeds, mouthpieces, etc. which have more durability and ease of use than the older ones. don't think quality suffers for that.

the tech-race in the electronic music sphere is all about the brain and ultimately going around the cochlear, directly to the neural pathways that mediate audition to produce more sounds and new musics! (even, if each piece isn't of the highest skill level.)

[BONES]

aaastronomer; and such a choice is entirely respectable, it's the way of the future. when bones rants about this choice it makes me feel like an old man yelling:

"you just can't handle it sonny. why back in my day i had 439 cables measuring in total some 2/3rds of a kilometer hanging off four walls of modules. the room had no door and we had to use a rope and pully system to get in and out through a hole in the ceiling that was made when one of the 400watt diesel generators blew up. ...and get off my lawn!"

the fact remains that some of the best existing compositions were written at a pianoforte, harpsichord or something similar hundreds of years ago on paper with a quill and ink.
bones just doesn't seem to have any clue he's one more step, step #492 between bach and some 12 year old kid pressing one button and writing a hit to make millions some time in the future.

Are you autistic in some measure? I am the opposite of that guy - I'm the bloke who has embraced the technology and pushed it forward to a point I could never have imagined at the turn of the century. That 12-year-old has been around since I was born, making hits by numbers. From the Monkees [the earliest I can remember but the people responsible for them had been around for years at that time], through Stock, Aitken and Waterman right up to the endless formulaic pop that infects the airwaves today. Those guys will always be there, working the formula, its completely irrelevant to anything that interests me even slightly.
What matters to me is being able to express my creativity. If all I wanted to do was make music or money, I'd be a DJ. Technology has transformed the way I do it and allowed me to follow it to its logical conclusion. I released two solo albums in around 15 years of using hardware, because the limitation of that technology was the size of my wallet. So the main limitation that has been removed is that I can now afford to do all of the things I want/need to do to see my creative urges realised. Consequently, since going all software we have been able to complete 3 albums in half that time, all to a significantly higher standard.
But cost is only one aspect, I have also been able to refine my workflow to the point where it is so close to perfect that it just doesn't matter. Almost every tool I use has an interface designed by me to facilitate my process. I have even been able to create instruments and effects that work exactly the way I need them to, something I could never have envisaged when I started down this path.
People who think they need something more are simply not approaching things from the right perspective - there are still uncountable depths of ability in the tools we have now, its simply a matter of looking at things with a fresh eye to see that most of us have barely scratched teh surface.

[pdxindy]

i wouldn't even say the most impressive advancement in cganimation is modeling or rendering - i'd say the motion capture and animation systems of today are extremely, extremely improved against those used when animations like "toy story" were built. today it's possible to connect sensors to a real actor and have all physical properties of their bodies and clothing accurately modeled by the system while oversampled motion capture is done providing for accurate motion-blur and other effects.

the modern quality is like the difference between a bumpmap textured phong shader and a flat shader.

I was using modeling as an example because the thread is talking about improving the interface for music software so it is more intuitive and less in the way. Mudbox and Zbrush have done that for modeling. It is like sculpting and you can just focus on the shape whereas before, you needed a lot of technical skill to not end up in a dead end corner.

[aaastronomer]

...of course robots and other forms of mechanization are not needed in industry or agriculture either, and the fact we have so much food and goods available that everybody is becoming obese whilst listening to their i-pods is a terrible, terrible thing. let's do everything by hand again, sure everybody will be poor except a privileged few, but we'll be those few won't we?

when you cry about ejay or related terrible, terrible technologies or tools you're only crying for fear of your worth disappearing. you'll be replaced, eventually. get used to it.

you must be joking ! the level of population on this planet makes it impossible to return to doing everything by hand again! (just think of doing medicine by hand again!!!!!! how bout with unwashed hands again? no thanks!) further, not "everybody is becoming obese whilst listening to their i-pods". what about those people in the third and second worlds?

hey, i attempt to play the didgeridoo and middle eastern drums, but i also use all of the electronics i can buy in composing and performing.

as for my worth, i started out the child of slaves and technology has allowed me to be free to travel the world and live reasonably well (and in health), purely off my intellect. you'll have to pry my technology from my dead hands!