 |
|
You are not logged in! F.A.Q Log in Register |
 |
|
 |
 Now online (2)recycle big ...and 147 guests Last 5 registered Oplandisks nothingstar N_loop yipe foxtrotromeo Browse members... |
 |
Members 8025Messages 2614231 Today 13 Topics 127550 |
|
|||||||||
| |||||||||
|
|
ok so i'm shit at maths but i was just thinking, if a 3 minute pop song is encoded on cd in 16bit samples at 44.1Khz, then there are theoretically: (44100 * (2**16))*(3*60) different possible 3 minute cd tracks. thats 260112384000 possible songs/sounds/whatever. Seems wrong. seems too few. Even if my maths is wrong it's still Quite mind blowing to think about the fact there is a finite number of cd encodable sounds. |
||||||||
|
|
|||||||||
| |||||||||
|
|
Yeah, I don't know if your calculation is right, but that's an interesting point. I've never thought about it before. |
||||||||
|
|
|||||||||
| |||||||||
|
|
but there's also songs that are 3:01 or 2:59 |
||||||||
|
|
|||||||||
| |||||||||
|
|
i've checked it and i think its right now |
||||||||
|
|
|||||||||
| |||||||||
|
|
all theory, no action ;) |
||||||||
|
|
|||||||||
| |||||||||
|
|
actually i got that wrong :) its ((2**16) ** 44100)=1445068800 possible sounds in one second so its (1445068800**(3*60)) for one minute and thats gonna be a fucking massive number. still finite though ;) |
||||||||
|
|
|||||||||
| |||||||||
|
|
substract to that number all the posible 1 minute files that are only noise |
||||||||
|
|
|||||||||
| |||||||||
|
|
now ya givin me reason to try to expand on that by makin a zillion 3 minute tracks |
||||||||
|
|
|||||||||
| |||||||||
|
|
oh yeah....have you considered the # of ways those sounds could be rearranged/EQd/FXd ? |
||||||||
|
|
|||||||||
| |||||||||
|
|
I don't think you need to consider the permutations of the samples, which would be (3*60*44100)! - that exclamation mark denotes factorial - since you are already considering the fact that each sample could take any of the 2^16 possible values. That would mean the number of possible 3-minute long sounds (most of which would be noise) is 3*60*44100*65536 = 520224768000. Feel free to say I'm wrong. |
||||||||
|
|
|||||||||
| |||||||||
|
|
this reminds me of the idea of discovering a poem's greatness by plotting it's attributes in the xy-plane. |
||||||||
|
|
|||||||||
| |||||||||
|
|
how exactly? |
||||||||
|
|
|||||||||
| |||||||||
|
|
yeah :| |
||||||||
|
|
|||||||||
| |||||||||
|
|
would you have wanted to rreally? |
||||||||
|
|
|||||||||
| |||||||||
|
|
Ok, maths geek here: You were closer in your first calculation than in the later one. Consider it this way. In each sampling period on a CD, the sample can have one of 2^16 values. In the next sampling period it can also have any one of those 2^16 values. So for two consecutive sampling periods you can have (2^16)^2 different possibilities [=65536^2 or approximately 4.29 x 10^9] In 3 minutes there are 7938000 sampling periods [44100*3*60] so the total number of 3 minute tracks would be 65536^7938000. [which is a whole heck of a lot!, more than 520224768000] The question I find more interesting is how many of these tracks would even be distinguishable from one another, let alone listenable. |
||||||||
|
|
|||||||||
| |||||||||
|
|
"The question I find more interesting is how many of these tracks would even be distinguishable from one another, let alone listenable." Yes, think how many of those would share the same bassline/break but with different vocals etc. *hopes* some f***ers like Sony don't make a supercomputer to generate all possible tracks under 6 mins and copyright them thereby preventing anyone from getting royalties on new music ;-( |
||||||||
|
|
|||||||||
| |||||||||
|
|
"Yes, think how many of those would share the same bassline/break but with different vocals etc." I'm talking even more indistinguishable than that...e.g. among those gazillions of tracks would be two that are exactly alike except that the first sample in one would be 45000 and the first sample in the other would be 45001, for instance. The human ear couldn't tell the difference... "*hopes* some f***ers like Sony don't make a supercomputer to generate all possible tracks under 6 mins and copyright them thereby preventing anyone from getting royalties on new music ;-(" Not likely to happen, even supercomputers aren't fast enough to generate them all in a reasonable amount of time. And I don't think that they could copyright an algorithm to do so, or else I could go and claim copyright on all written works in English under 1 000 000 characters in length. After all I can describe an algorithm to generate them... |
||||||||
|
|
|||||||||
| |||||||||
|
|
Sorry, I should of emphasised the "AND". I meant to say, as well as all the ones that were indistinguishable due to only have the odd bit here and there different, you would also get a lot of remix type ones. And my supercomputer thing was meant to be a bit of a joke, the law tends to favour published material in copyright cases... |
||||||||
|
|
|||||||||
| |||||||||
|
|
"maths geek here" neverr werre 3 trruerr worrds typed. |
||||||||
|
|
|||||||||
| |||||||||
|
|
Yes, and your point is? :) Ceri, yeah it's a good thing the law favors actual material over potential material in the realm of copyright. |
||||||||
|
|
|||||||||
| |||||||||
|
|
that neverr werre 3 trruerr worrds typed |
||||||||
|
|
|||||||||
|
Messageboard index
|
|||||||||
