Do you REALLY hate tarnce/tracne?

[Frantz]

Take Cosmic Baby for instance.

I actually liked the sequencing in Cosmic Baby. I would enjoy it more with some leads or vocals on top but I guess it worked for me as is.

It seems like Trance removes the foreground elements (top line melodies or vocals) of dance music and just features what I think of as background elements. It's not really my thing but I guess I could find Trance I like if I dig deep enough.

[ariston]

I know Zweiraumwohnung very well... it's what many Germans call "Cappuccino music"... to me, it's the musical equivalent of cottage cheese. The kind of music that people who don't care about music love to buy, something to enhance their lifestyle. Not my cup of tea at all, sorry. I fondly remember the Humpe girls in their heyday (DÖF & Ideal), but Zweiraumwohnung and Ich + Ich... ugh.

I'm not a big fan of them but I like Zweiraumwohnung because they combine well-crafted pop music with some electronic elements as Tommi Eckart made trance music before this project. For me it's a progression because German Schlager & Diskofox music are the worst things I've ever heard in my life (besides "Volksdümmlicher Musik"). And Zweiraumwohnung is a bit different!

Discofox! Ach du Scheisse, das weckt schlimme Erinnerungen!

Another dumb question: would you consider Underworld to be Trance? They played the Southside festival once (right before Radiohead), and it was bloody awesome. They kept on pushing them to finish their set and leave the stage, but the crowd was going wild and wouldn't let them. Danced my arse off that night.

[Tricky-Loops]

I know Zweiraumwohnung very well... it's what many Germans call "Cappuccino music"... to me, it's the musical equivalent of cottage cheese. The kind of music that people who don't care about music love to buy, something to enhance their lifestyle. Not my cup of tea at all, sorry. I fondly remember the Humpe girls in their heyday (DÖF & Ideal), but Zweiraumwohnung and Ich + Ich... ugh.

I'm not a big fan of them but I like Zweiraumwohnung because they combine well-crafted pop music with some electronic elements as Tommi Eckart made trance music before this project. For me it's a progression because German Schlager & Diskofox music are the worst things I've ever heard in my life (besides "Volksdümmlicher Musik"). And Zweiraumwohnung is a bit different!

Discofox! Ach du Scheisse, das weckt schlimme Erinnerungen!

Another dumb question: would you consider Underworld to be Trance? They played the Southside festival once (right before Radiohead), and it was bloody awesome. They kept on pushing them to finish their set and leave the stage, but the crowd was going wild and wouldn't let them. Danced my arse off that night.

Never heard of Underworld but a few years ago I listened to Diskofox for some hours and the only song that I couldn't forget was a song with some samples of a barking dog... ...all other songs sounded the same...

[ghettosynth]

It's something that cannot be accomplished by the best of musicians

How do you know?

Because human timing is, wait for it, "analog." In the past I've posted links to research work that highlights our understanding of music perception, action, and neuroscience. They seldom elicit comment, however, and so I'm not going to spend my time giving you links today. The bottom line is that no matter how good you are, your variation is driven by the distribution of neuronal timing and the number of systems involved in creating the action. These limitations are also driven by the nature of feedback and correction which limit the ability to maintain an error bound, especially over the long term. No human is a machine over the long term. It's not unlike how analog oscillators will drift.

Machines do not have such limitations and it matters. If we could not perceive the difference, we wouldn't be having this discussion.

[ghettosynth]

To preempt any comments regarding the so called intelligence of this music here are some excerpts from an email exchange between a friend of mine and one of the guys in Hypnocoustics:

... while listening to some of your songs, it ‘feels’ the same way in the brain as listening to some of Mozart’s violin sonatas.

Anyone can relate to what he's saying, even if not able to directly apply it to the music I posted.

I do have some classical training! Played the violin all my life and lots of orchestras/ string quartets and stuff.

Make of that what you will, but it seems pretty unlikely to be pure chance ...

Peace,
Andy.

I disagree, not that it isn't pure chance, but that, based on those two quotes, that one can reasonably apply a probability to the outcome. Not that I think that it matters in any case.

[ghettosynth]

Nah i can see that, but to be fair you could also write an algorithm that produces simplistic rock music.

The difference is that rock music still needs someone playing guitars, someone singing lyrics, etc. My feeling is that trance could be entirely machine generated.

But your argument is nothing more than a statement regarding the limitations of technology to reproduce someone singing or playing guitars, certainly both are possible today. So just as you would reject rock music created that way on the basis of the quality of that reproduction, the same can be said for your reproduction below. It isn't good trance, it simply attempts to poorly mimic the compositional component of a genre.

For example, here is one guy's attempt at it. Every time you click this link on anticulture.net, it generates a new MIDI file: Random Trance Music Generator

It will output random progressive melodic trance music that is similar to high energy eurodance and techno music from the late 90's.

Admittedly it's not that great but it illustrates the point.

No, it's not great, it's not even good. It does illustrate a point, but not the point that you think that you're making. It illustrates that pop music has predictable structure and that you can easily simulate that structure probabilistically.

We actually understand this much better than the crappy example that you posted demonstrates.

http://igm.rit.edu/~jabics/GenJam94/Paper.html

The interesting fact is, however, that we are far more able to create "art" music and quasi-art music than pop music. The reason for this is largely twofold: 1) There is simply much more interest and ability in the music research community on art music, and 2) Timbre generally plays a more constrained role in such music. I'm sure that the clergy among us will take issue with this statement, but, I only mean that it is sufficient to state that a particular line in a jazz or classical piece should be played by a particular orchestral instrument to convey almost complete information regarding timbre to a fellow academic or critic. This is not the case with synthetic instruments.

http://www.aaai.org/Papers/AAAI/2007/AAAI07-133.pdf

There is structure in all music that interests any statistically interesting proportion of the population. That is, I'm sure that you can find some art pieces that lack any easily perceptible structure, I'm equally sure that they're relatively unappreciated. Since music has structure, it can be predicted, therefore, any music style can be created algorithmically, whether the creating is appreciated by humans is a different question.

I remind you again of my favorite quote attributed to George Box, "All models are wrong, some models are useful."

[robogone]

It's something that cannot be accomplished by the best of musicians

How do you know?

Because human timing is, wait for it, "analog." In the past I've posted links to research work that highlights our understanding of music perception, action, and neuroscience. They seldom elicit comment, however, and so I'm not going to spend my time giving you links today. The bottom line is that no matter how good you are, your variation is driven by the distribution of neuronal timing and the number of systems involved in creating the action. These limitations are also driven by the nature of feedback and correction which limit the ability to maintain an error bound, especially over the long term. No human is a machine over the long term. It's not unlike how analog oscillators will drift.

Machines do not have such limitations and it matters. If we could not perceive the difference, we wouldn't be having this discussion.

That's it? That's your argument? Perhaps you misunderstand the vast array of techniques used for making electronic music, many of which don't involve machines or computers playing the notes.

...and BTW - analog equipment with all its inherent timing and tuning issues is used to create electronic music.

[3ee]

[ghettosynth]

It's something that cannot be accomplished by the best of musicians

How do you know?

Because human timing is, wait for it, "analog." In the past I've posted links to research work that highlights our understanding of music perception, action, and neuroscience. They seldom elicit comment, however, and so I'm not going to spend my time giving you links today. The bottom line is that no matter how good you are, your variation is driven by the distribution of neuronal timing and the number of systems involved in creating the action. These limitations are also driven by the nature of feedback and correction which limit the ability to maintain an error bound, especially over the long term. No human is a machine over the long term. It's not unlike how analog oscillators will drift.

Machines do not have such limitations and it matters. If we could not perceive the difference, we wouldn't be having this discussion.

That's it? That's your argument?

You asked a question, I answered, you haven't refuted my answer. The point is simple, humans are incapable of creating the same degree of precision as a machine with respect to timing in a musical context.

Perhaps you misunderstand the vast array of techniques used for making electronic music, many of which don't involve machines or computers playing the notes.

In which case, my premise doesn't apply. So, why would you bring it up and, further, why would you assert that I don't understand the "vast array of techniques?" You statement is a non-sequiter, it has absolutely nothing to do with my argument. In fact, if you had read my initial statement you see that I point out that when people try to "play" , e.g., house, that it comes off sounding like funk. I'm well aware of how EDM is created and how different techniques are combined. You are being disingenuous here, however, as I'm sure that you're well aware of how little long term variation in timing there is virtually all commercial EDM. If you've ever DJed you know this implicitly. The few tracks that move around are a pain to mix, like most classic funk/disco. I play house, I've mixed disco, have you?

To be clear, however, my statement is not an assertion that ALL of any track is sequenced, or, even sequenced in the same way. So, because tracks often have lead lines that are "played" over the top of the beat doesn't refute the point. Of course, that line will have inherent human variation. That variation, however, will be over a beat that is sequenced or looped which will not share the same distribution of timing.

Also, my response was given in justification for understanding a particular kind of EDM which is, in fact, almost universally sequenced and precisely intended to elicit the kind of response that I'm describing.

...and BTW - analog equipment with all its inherent timing and tuning issues is used to create electronic music.

Wow, really, thank you Captain Obvious. I think that I wrote just a few days ago about playing out with my 909 as a live techno act. I think that, maybe, despite my pure amateur status, that I might know a little something about how analogue gear is used in EDM.

Second, the problem with your simple interpretation is that you are trying to prove a point with analogy whereas I was using analogy to help in understanding. The use of an analog synth to create, e.g., a sequenced bassline, does not introduce the kind of timing variation that is a consequence of neuronal timing.

Finally, we're talking about trance/tarnce here, which is quite constrained, we're not talking about "electronic music" in general. Please tell me which tarnce hits use little or no sequencing?

[Frantz]

We actually understand this much better than the crappy example that you posted demonstrates.

http://igm.rit.edu/~jabics/GenJam94/Paper.html

You not only disagree with me but you have PDFs, papers, flowcharts, and PowerPoint slides to prove your case. I'm not willing to invest so much time in this debate. I'd rather be making music.

I think I'll pack up my things and switch on the DAW instead. See ya.

[V0RT3X]

when i was 18 which was 11 years ago.. i was really enjoying the sound of Juno Reactor which is sort of a trance genre i think. Now I'm not really into as much, but i still enjoy a decent trance song when i'm driving or something.

[ghettosynth]

We actually understand this much better than the crappy example that you posted demonstrates.

http://igm.rit.edu/~jabics/GenJam94/Paper.html

You not only disagree with me but you have PDFs, papers, flowcharts, and PowerPoint slides to prove your case. I'm not willing to invest so much time in this debate. I'd rather be making music.

Yes, indeed, my "case" has support. I think about these things because music cognition is an interesting subject to me and I'm paid to think about statistical models and how they help our understanding of the real world. If it's not interesting to you, perhaps you should spend your time making music instead of making assertions for which you have no support?

[nasenmann]

We actually understand this much better than the crappy example that you posted demonstrates.

http://igm.rit.edu/~jabics/GenJam94/Paper.html

You not only disagree with me but you have PDFs, papers, flowcharts, and PowerPoint slides to prove your case. I'm not willing to invest so much time in this debate. I'd rather be making music.

Yes, indeed, my "case" has support. I think about these things because music cognition is an interesting subject to me and I'm paid to think about statistical models and how they help our understanding of the real world. If it's not interesting to you, perhaps you should spend your time making music instead of making assertions for which you have no support?

Eh, misinformed assertions can still spawn good discussions. He basically said "you win" in a roundabout way, didn't he

On the matter of art music lending itself well to being algo generated, there's a very obvious chicken and egg reason:
Algo music itself is often understood as a niche of art music.

[hibidy]

Enjoyed the F# lydian here!

Nice tune, well mixed!

[hibidy]

I think I'll pack up my things and switch on the DAW instead. See ya.

Another bitwig convert!

(hehe, only kidding dude! )