where next with software instruments?

[SadPuppyBlues]

if you were all tool-makers you'd understand there are in fact a lot of tools and features of tools left out today - and most of them because guys who don't make tools don't take advantage of those sorts of features.

That kind of assumes people's perception of an advancement is an entirely new feature. Most likely, it's going to be redesigns of existing features that makes things less cumbersome and unintuitive. Also, just like in the past, the generation who went without them won't be able to fully absorb many advancements.

[BONES]

... you have almost anything (not everything certainly!) available to you and you actually use about 15% or so a majority of the time. nobody likes to experiment, or produce something that really stinks it seems.

You need to put a reasonable amount of features into an instrument so that it has a basic level of appeal across the board. Some people will use A, B & C, where others will use A, C, D & F [maybe no-one will use E, but E is stupid anyway]. And I'm sure all we tool-makers exercise some quality control [although you may not always think so], so that most of the experiments that don't turn out are shelved before anyone gets to see/hear them. It would be pretty arrogant to assume that every little thing you did was worth releasing.

Somebody with some guts and some first rate programming chops is going to have to spend a year or two hacking on something that's going to blow us all away and make us wonder why things got so stuck for so long.

I admire your optimism but it hasn't happened in the last 20 years, since the birth of the workstation synth, so I can't understand why you think it might be somewhere around the corner now.

This is exactly the problem. Musicians are now expected to wear three or four hats at once and wear them well. Hardly anybody is actually capable of this. Let's let the computer take over some of that.

Anyone who has been in a band will tell you its not the same thing. I've let the computer do everything that a band might do for me for 25 years, for various reasons, but there have to be limits. And I'd challenge the notion that we require any more skill than Mozart did, its just different skills that are needed today. That's where people are slow to adapt, they still think they need to learn to read music and play an instrument. Its crazy. I put the effort in where it pays dividends.

... if you want fresh and new sounds without too much randomization happening and without becoming a synth geek...

This is what I mean - being a "synth geek" is a far more relevant thing today than being a musician, and that's why I work hard at that and not the other.

[soulata]

That's where people are slow to adapt, they still think they need to learn to read music and play an instrument. Its crazy. I put the effort in where it pays dividends.

It's still nothing to be ashamed of ...

... if you want fresh and new sounds without too much randomization happening and without becoming a synth geek...

This is what I mean - being a "synth geek" is a far more relevant thing today than being a musician, and that's why I work hard at that and not the other.

I'd say as relevant maybe, surely not far more relevant. But I guess it depends on many things such as lifestyle, circles you move in, where you live ...


k

[kuniklo]

I admire your optimism but it hasn't happened in the last 20 years, since the birth of the workstation synth, so I can't understand why you think it might be somewhere around the corner now.

For exactly the reasons you state. We've spent the last twenty years collapsing a conventional hardware based-studio into the computer, complete with virtual mixing consoles and photorealistic controls. Now that there's not much left to do with that approach, as you yourself correctly diagnose, it's time for the next stage.

[kuniklo]

That's where people are slow to adapt, they still think they need to learn to read music and play an instrument. Its crazy. I put the effort in where it pays dividends.

I don't see these skills as obsolete at all. You still need to understand music theory, and that's easier on a staff than a piano roll. You also need to be able to play things out unless you're just banging beats into a grid. If you can give me a convincing rendering of something along the lines of Eric Dolphy's "Out to Lunch" or a Mozart string quartet by step-editing it in Orion I'll retract both claims.

[BONES]

Why? I don't. I used to when I was playing trombone in the high school band but those skills atrophied long ago, through disuse. I can read a piano roll far more easily than sheet music. I still have to count off "every good boy deserves fruit" to work out notes, where a PR has them written on the bars for me, and even shows how they correlate to the black and white keys on my controller.
As I said before, when I used to have to play two parts and sing at the same time, I needed to be a competent player and spent a lot of time practising. Now I devote that time to the minutae of the production process, because I can programme a performance to be perfect in no time at all. I have been able to free up a lot of time to devote to devote to areas that 25 years ago were beyond my means entirely.

For exactly the reasons you state. We've spent the last twenty years collapsing a conventional hardware based-studio into the computer, complete with virtual mixing consoles and photorealistic controls. Now that there's not much left to do with that approach, as you yourself correctly diagnose, it's time for the next stage.

Why? What's been going on has been a refining of a process to a point that it is incredibly flexible and infinitely usable. Its not like a KORG M1, full of limitations that provided the impetus to continue refining the paradigm. Quite the opposite, in fact - we've pretty much achieved nirvana. What you are saying seems to be like spending a week in heaven and then wondering what comes next.
At some point we have to get to the stage where we don't need anything more or anything new. I think that happened a few years ago and we are now free to concentrate of using the damned stuff, instead of waiting for this or that limitation to be addressed so that we can get on with the job. As someone suggested, many developers are simply adding bells and whistles to make their products look more attractive, just as car makers offer more and more kit as standard, when once it was the preserve of luxury models. In both cases, the new stuff isn't different but it is quantifably better, and surely better is far more important than different, isn't it?

[Terrafractyl]

That's where people are slow to adapt, they still think they need to learn to read music and play an instrument. Its crazy. I put the effort in where it pays dividends.

... if you want fresh and new sounds without too much randomization happening and without becoming a synth geek...

This is what I mean - being a "synth geek" is a far more relevant thing today than being a musician, and that's why I work hard at that and not the other.

man I hate this attitude...
And people wonder why you have to search through piles of crap, uninspired music to find anything decent these days...
Synthesizers, Music theory, DSP theory, Equalization, orchestration....(ad infinitum)
They're all just tools, If you REally want to write good music, why ignore any of them???

edit: missed your post above BONES, but I'd still like to hear why you think no-one should care about music theory...
I for one would absoloutely detest a world with no counterpoint, harmonic structure and correct voice leading... these are the things that make my music work, and that I appreciate the most in the music that I like.

These are things that do not necessarily just come to people as they point and click. You have to work at some of this stuff.

[kuniklo]

Why? What's been going on has been a refining of a process to a point that it is incredibly flexible and infinitely usable. Its not like a KORG M1, full of limitations that provided the impetus to continue refining the paradigm. Quite the opposite, in fact - we've pretty much achieved nirvana.

You don't seem to be able to get your head around the idea that just because your tools and process are well suited to making *your* music by *your* method that they might not be ideal for everybody else. Style matters. EDM is easy.

[ouroboros]

I think the future will be
vocal to sythesis/vocal to midi.

On this I would put money. When people can hum in a mike and [easily] the computer accurately translates the hum/whistle into something that sounds like a virus, it will be and end game of sorts.

I realize there are fledgling attempts already, but as comp resources continue to increase and the tracking becomes precise, "intelligent" and easy...well I might buy one. Just imagine feeding Strong Bad's "Meedly-meedly-meedly-mee" into a mice and having Jimi come out the speakers.

I'll be able to hum my brilliant ideas and instantly create top-40 gold..

[kuniklo]

On this I would put money. When people can hum in a mike and [easily] the computer accurately translates the hum/whistle into something that sounds like a virus, it will be and end game of sorts.

I agree that's an interesting avenue to explore. The human voice is still our single most expressive and flexible instrument by far.

[aciddose]

take the mic out of that picture, a lame, physical interface for a physical artist. it requires the artist to hum in tune, in time, and even know what they're humming.

think in terms of simply sitting in a chair, having absolutely nothing but the chair in a room and commanding every aspect of an orchestra of instruments, synth or not, with absolutely no requirement other than to dream of the music.

[Shy]

We've achieved no "nirvana". If you listen to "electronic music" too much you might actually forget how different real instruments sound, behave and make you feel.

A good piece of music played on a chello by a good cellist is in a different universe than anything we can do with synths. Same goes for trumpet for example (and don't give me an argument about physical modeling. I'm a fan of it but I know what's available and I know what it can and can't do), and all kinds of drums.

When it comes to synthesis, in terms of sound, control, expression and feel, we're in a different universe. Not necessarily a bad thing. But when you try to imitate that other universe, even with extreme efforts, the results you'll get are basically like what a toy is compared to a real thing. There are awesome things that can be done with synths, and that's why they're my favorite instruments and I can't survive well without listening to some for long, but I know what can and can't be done. The thing is, I don't care. If I really want to express myself in a way that I can't with synths, I'll use acoustic instruments. I'll learn to play those few notes I need, and I'll learn to tap those few beats I need. I won't try to imitate playing on that maple Ludwig drum with a synth, nor with a huge set of samples, because the result will be completely different.

I think that achieving some greater sounds with synths is always a good thing. But as far as expression goes, you can't expect to control an instrument in a non physical way (or in a very limited physical way) and for it to be as intuitive and expressive as some acoustic instruments and to sound as "real" and "alive". You produce sounds by interacting with and directing an instrument, and its sound is a result of its characteristics as well as how you do that. With some synths you can achieve fairly complex sounds by manually editing or recording many parameters and you can get useful and nice results, but you can't expect a few parameters (and even dozens is a few) to result in a sound that's really from the same universe as that of real instruments.

Instead of things like very complex sets of parameters that are supposed to imitate effects of real instruments, I would be much more interested in parameters that shape the sound in far different ways than we're used to. What do we have today? Not a very innovative selection of amplitude, frequency, resonance, delay, spectral, distortion based shapers and modulators, and a few mostly similar forms of sound generation. The innovation I'd like to see is first of all in quality and variety of what's already common, and then I'd like to see some good truly new generators and effectors.

As for control, you have 5 fingers on each hand. You could use a fader controller to control 10 parameters in real time. Or maybe someone will design a special hand/arm held instrument that controls a special virtual instrument in a fun and unique way and the instrument would make unique sounds and life will be peachy. And then maybe you'll be able to control more unique virtual instruments with it, until one day there won't be any more unique ways to use it and you'll figure that you should get another instrument that controls a virtual instrument that makes fun and unique sounds in a fun and unique way. You know what, maybe great control is physical control. And maybe not. Maybe it's your preferences and aims that count, and maybe musicians express themselves in many different ways and what's next for software instruments doesn't necessarly have to be one thing or a bunch of things in order for the -entire- virtual instrument world to be considered progressed. Instruments are individual first of all. What's available reflects what's been attempted and what's been demanded.

[zerocrossing]

Ah, one of my favorite soapboxes. AFTERTOUCH. Why did we go backwards?

Money. People want it cheap and mass produced.

The good news is that at least Novation is making keyboards that have good feeling action and smooth aftertouch. It's not poly (damn them) but my Remote 61 and 25 sl feel great.

But I agree. We've gone backwards in that sense, but, the hardware is significantly cheaper than what was available in the 80s.

I'd sure like to have a new controller with Poly Aftertouch and a breath controller input and great key action as well as a bunch of endless knobs, foot pedals and well built.

Controllers today are mostly shit. If you want poly AT you gotta buy an old synth. Lots of the controllers today with Channel Aftertouch the subtlety is crap. Many of them are more like an on/off switch than a smooth controllable range.

We've gone backwards on that front

[zerocrossing]

No need to wait for anything. In the 80s I took a cheap dynamic mic, ran it through some stomp boxes and then directly into a mixer and in my best falsetto I wailed the most amazing guitar solo you'll ever hear.

Advances don't have to be in the instrument itself. Look at a simple electric guitar. In the hands of Dick Dale, you get a nice surfer sound, but nothing one would call advanced or pushing the limit. Great style though. Put that same instrument in the hands of Jimi Hendrix... and it's a totally different deal. Again, same instrument in Eddie Van Halen's hands is something new yet again.

Same with synths really. I think we need to forget the dialog about the future of instruments and talk about the future of playing. I imagine the next advances in synth technology will be biotech. Imagine using a parrot brain as your CPU. That baby would scream. Yeah, I'll be the first in line to buy the BaGOOK2000.

I think the future will be
vocal to sythesis/vocal to midi.

On this I would put money. When people can hum in a mike and [easily] the computer accurately translates the hum/whistle into something that sounds like a virus, it will be and end game of sorts.

I realize there are fledgling attempts already, but as comp resources continue to increase and the tracking becomes precise, "intelligent" and easy...well I might buy one. Just imagine feeding Strong Bad's "Meedly-meedly-meedly-mee" into a mice and having Jimi come out the speakers.

I'll be able to hum my brilliant ideas and instantly create top-40 gold..

[Lunatique]

At some point we have to get to the stage where we don't need anything more or anything new. I think that happened a few years ago and we are now free to concentrate of using the damned stuff, instead of waiting for this or that limitation to be addressed so that we can get on with the job. As someone suggested, many developers are simply adding bells and whistles to make their products look more attractive, just as car makers offer more and more kit as standard, when once it was the preserve of luxury models. In both cases, the new stuff isn't different but it is quantifably better, and surely better is far more important than different, isn't it?

Nirvana has not happened for many of us yet. Sure, we are in a far better position today than we were just a decade ago, but I think humans have a tendency to improve upon things that needs improving, and lay off on things that are already plenty good enough. For example, you probably won't see any significant advances in the design of wall clocks, because we don't really need it to be more than what it already is. But music production tools? There is so much room for improvement left. This is especially true for people who deal with high-end sample libraries. We are just now seeing some interesting things happen on the front of sampling/resampling/physical modeling with products like Synful Orchestra or Sample Modeling's The Trumpet. It'll be a while before we can truly do away with having to fiddle around piecing together articulations to form a realistic and expressive performance. Of course one can learn to play the real thing, but the ideal is to not have to spend tens of thousands of dollars on quality physical instruments and then invest years of your life to learn to play them at a high level--you can just use your MIDI controllers to play virtually any instrument for a fraction of the cost and time.

As for control, you have 5 fingers on each hand. You could use a fader controller to control 10 parameters in real time.

I can tell you right now that most today's computer musicians can barely control three parameters at the same time--most can't even play an instrument that requires any kind of separate limb/finger independence. the keyboard is one of the easiest instruments to play yet some can't even go beyond "one-finger'plinking" on it. Ten parameters at the same time? Hell no.