Hey Christian and whoever else sees this–I just came across this while surfing the net today: http://news.harvard.edu/gazette/story/2012/07/when-the-beat-goes-off/
I’ve always wondered why adding random quantize just doesn’t seem to add much, and now I know. If you could pass this tidbit on to the developers, perhaps they can get the jump on every other DAW out there and introduce an improved algorithm based on this research in Cubase 7!
I found this utterly fascinating, and it makes perfect sense!
so why did you even bother reading it then . it
s obviouse machines will never be able to play like a bloby,get a grip .
I always find that human-Q helps quite a bit, whether it’s staccato violins in an orchestra peice, or kick and snare in a rock tune. Defitely adds a human ‘element’ that was not there prior.
if the music is worth humanising then it is worth the extra dollar for some real humans , otherwise it`s just a cheap gimmick .
You obviously have absolutely no idea how much money it costs to hire an orchestra; if you did, you would not be saying that.
you play me a piece of music deserving an orchestra and i`ll pay for it take more notice of the lol
Interesting read. Thanks for sharing!
The point of the article is that the random quantize added in software like Cubase is NOT human-like random but computer random. Very interesting read indeed, and hopefully this research can lead to better randomizing!
The article insists on calling “not quite with the metronome” an “error”. Then goes on to explain that it isn’t!
A scientist finds it hard to understand that mathematically perfect rhythm isn’t “right”. Any practical musician knows that straight away. But it’s good to have it confirmed again that “human feel” does NOT equal “random wrongness”.
I like it when they state that the fluctuation follows fractal-like patterns. In the visual arts, fractal algorithms are used to interpolate data when blowing up images, and this with a lot of success. So they can indeed be used.
Yes, I think that if I were a developer/coder, I’d jump on it!
Thanks for sharing!
Does cubase just use a pure random? I always thought these things were cleverer than that. What about Battery 3’s humanise and Jamstix humanise?
Nothing is cleverer than the humans who program it. The thought process is. “I have bad timing. I’m also a programmer. So! I’ll design a program that corrects my bad timing!”
Anyone see the flaw there?
No two humans have the same timing. No human has the same timing from day to day. No two humans in the same band have the same timing from day to day and room to room, as random reflections have an effect on that too. And two or more human musicians will react to each other differently depending on what the other does and what the room does to his sound and from day to day.
Now give me an unlimited number of monkeys and an eternity and I’ll program a drum machine for you.
They might get close to it if they ever decide that any given drum machine or quantisation device ALSO affects any accompanying instrument parts/tracks as well as JUST itself.
Like the sympathetic strings in a piano.
Human timing and quantisation is not a solo instrument. It is an esemble. Ensembles push each others’ timing around.
That’s why it’s not a pure random. Because it’s all done on ONE element only.
We’re gonna need bigger computers.
Yes indeed! Until then though we could be able to download human performances like reverb impulse responses and apply them via groove quantise. So you could get ones from your favourite performers.
Thanks for the link, very interesting.
Interesting article. It would more likely to be noticed by the developers if it were posted in the C7 features and requests sub-forum instead of the C6 forum.
If only C7 had been out when this topic started…
opps, well it was brand new to me.