no zoning is killing me

stephen wrote:I'm pretty sure the numbers in trim mode are the sample numbers. So if you wanted to start your sample at 1 second, and end at 2 seconds, you'd set start to 44100 and end to 88200.

or am I still missing something?

Ohhhhhh I see. Thanks!

a rough way to do 'manual' zoning is to take the number of samples and divide by 16, say, for a 2-bar loop (8 for a 1-bar loop and so on). so now you know the length of each slice, and in the trim mode punch in the approprate numbers for the start and end points...you need a calculator, but it's pretty straightforward...i used to do it this way all the time on my old roland s-550.

(if this doesn't make sense and someone wants to know, i can try to explain in more detail)

sans soleil wrote:a rough way to do 'manual' zoning is to take the number of samples and divide by 16, say, for a 2-bar loop (8 for a 1-bar loop and so on). so now you know the length of each slice, and in the trim mode punch in the approprate numbers for the start and end points...you need a calculator, but it's pretty straightforward...i used to do it this way all the time on my old roland s-550.

(if this doesn't make sense and someone wants to know, i can try to explain in more detail)

please do...because me not having a 2000 before, i dunno what zoning is actually...I think its that it divides a loop into equal parts....right or wrong?

ok...so first let's assume that you've trimmed your sample to a 1-bar loop that loops cleanly, discarded the bits that aren't necessary...

say this 1-bar loop is 16000 samples long..normally the hits fall fairly evenly in 8th or 16th note increments (for a 1-bar loop, let's say it's ...

16000 divided by 8 is 2000...so if you divide this loop into 2000-sample chunks, each chunk should contain one distinct hit from the loop:

start point end point

hit 1 0 2000
hit 2 2001 4000
hit 3 4001 6000
hit 4 6001 8000

...and so on...

...hopefully this makes sense. i think that this is what the mpc2000 did, rather than search for peaks (like recycle). obviously, most samples aren't as easily divisible in length, but if you have a calculator, it's pretty straightforward.

sorry...last post got a bit malarkied...i didn't put that smiley thing in - it's supposed to be an 8 - and the numbers next to where to says 'hit 1', 'hit 2' are the start points and end points, respectively.

sans soleil wrote:ok...so first let's assume that you've trimmed your sample to a 1-bar loop that loops cleanly, discarded the bits that aren't necessary...

say this 1-bar loop is 16000 samples long..normally the hits fall fairly evenly in 8th or 16th note increments (for a 1-bar loop, let's say it's ...

16000 divided by 8 is 2000...so if you divide this loop into 2000-sample chunks, each chunk should contain one distinct hit from the loop:

start point end point

hit 1 0 2000
hit 2 2001 4000
hit 3 4001 6000
hit 4 6001 8000

...and so on...

...hopefully this makes sense. i think that this is what the mpc2000 did, rather than search for peaks (like recycle). obviously, most samples aren't as easily divisible in length, but if you have a calculator, it's pretty straightforward.

So what happens if I were to just record a sample and go straight to trim mode without setting it to a 2-bar/1-bar loop...will this technique still apply with this raw sample?

You'll need to do this:

Loop length / number of sections = Length of sections

ex:
The sound loops every 8000 samples and you want 4 sections:

8000 / 4 = 2000

0 - 1999
2000 - 3999
4000 - 5999
6000 - 7999

stephen wrote:You'll need to do this:

Loop length / number of sections = Length of sections

ex:
The sound loops every 8000 samples and you want 4 sections:

8000 / 4 = 2000

0 - 1999
2000 - 3999
4000 - 5999
6000 - 7999

So when I extract 4 samples, I start with *0-1999*(0:start pt; 1999:end pt), then so on.

You can do it all by knowing two things. The number of samples in the loop, and the number of sections you want. Then just do this:



length of section = length of loop in samples / number of sections

start point: (n-1)(length of section)
end point: (n)(length of section) - 1

where n = the section number (1, 2, 3, ...)


(the values posted in other peoples' posts are slightly off - this should be correct)

Nygh wrote:^^^Most typos in one post award!!!

Anyway, punch drunk, I use your method. I need to invest in a card reader tho like you said because its hard as **** to keep moving my MPC back and forth to the computer....

lol,

well i'll try this when i get home stephen...thanks.

stephen wrote:I'm pretty sure the numbers in trim mode are the sample numbers. So if you wanted to start your sample at 1 second, and end at 2 seconds, you'd set start to 44100 and end to 88200.

or am I still missing something?

Ive sampled something and the st pt # was 2055 or something and then end was 440999 HELP somebeody!!!

stephen wrote:You can do it all by knowing two things. The number of samples in the loop, and the number of sections you want. Then just do this:



length of section = length of loop in samples / number of sections

start point: (n-1)(length of section)
end point: (n)(length of section) - 1

where n = the section number (1, 2, 3, ...)


(the values posted in other peoples' posts are slightly off - this should be correct)

Thanks stephen you helped me alot, this formula worked perfect. algebra/trig formula...i love math...lol.

good thread. I need to do some manual zoning.

wasent it confirmed that the 1000 is getting zoning and timestrech in the next (may?) os update