LAYERING SAMPLES...LAYER ISSUES...

Here's the deal: I have a sample on the pads...and I want to filter the sample and add it to the 2nd layer of the original sample's pad.

at first i tried to place the new sample on the 2nd layer of the sample so they'll play at the same time, but i couldn't filter the 2nd layer without messing up the original.

i also tried copying the sample, filtering the copy, then placing it in the B section of the program, but it won't play at the same time as the samples on A. tried changing to poly, did nothing.

any suggestions?

wat mpc u on?

edit: nevermind stupid question, didnt realize what forum it was in

Here's the deal: I have a sample on the pads...and I want to filter the sample and add it to the 2nd layer of the original sample's pad.

folks have done this for years but there's no point - all you're doing is increasing the frequency of the original sample with those that the filter do not cut out in the second copy: basically, this exact same action could be done more surgically with an EQ - that's how this task is meant to be achieved.

when you're doing this by layering a sample with itself on the mpc, it's not even sample accurate, so you'll end up with phasing.

but as far as playing 2 pads together (one with a different sample on it) you need to use the SIMULT function in jjos, otherwise you need to resample, trim, and then layer

Nym wrote:

Here's the deal: I have a sample on the pads...and I want to filter the sample and add it to the 2nd layer of the original sample's pad.

folks have done this for years but there's no point - all you're doing is increasing the frequency of the original sample with those that the filter do not cut out in the second copy: basically, this exact same action could be done more surgically with an EQ - that's how this task is meant to be achieved.

when you're doing this by layering a sample with itself on the mpc, it's not even sample accurate, so you'll end up with phasing.

but as far as playing 2 pads together (one with a different sample on it) you need to use the SIMULT function in jjos, otherwise you need to resample, trim, and then layer

adding a filtered sample to the sample gives it a little more bottom than just eq'ing it imo.

thanx.

asr boy wrote:adding a filtered sample to the sample gives it a little more bottom than just eq'ing it imo. thanx.

think about what you're actually doing with the frequencies. here's a diagram.
ones equal bass frequencies, twos equal lower midrange, threes equal higher midrange, fours equal treble frequencies.

original sample:

1111222233334444

sample filtered with a lowpass:

111122223.

stack them:

1111222233334444
111122223

all you're doing is increasing the level of the existing low end by doubling it. this is the exact same thing as eqing, just more convoluted, less precise, and with less options. in an eq, giving a bump to those frequencies on the original sample would do the same thing.

the idea of layering is not to stack duplicate sounds but to add what's missing. ie if you have a bass multisample that is just plain weak in the lower bass range, you can add a sine wave and tune it down until it's adding what you wanted.

the idea of stacking duplicates wouldn't be a bad move in the MPC if it weren't for the fact that it's not sample accurate. since it's not sample accurate, your stacked layers do not always play at the exact same split second time. as such, there will be a phase sound as the two identical frequencies play slightly off time with each other. you'll hear it. it's not desirable.

Nym wrote:

asr boy wrote:adding a filtered sample to the sample gives it a little more bottom than just eq'ing it imo. thanx.

think about what you're actually doing with the frequencies. here's a diagram.
ones equal bass frequencies, twos equal lower midrange, threes equal higher midrange, fours equal treble frequencies.

original sample:

1111222233334444

sample filtered with a lowpass:

111122223.

stack them:

1111222233334444
111122223

all you're doing is increasing the level of the existing low end by doubling it. this is the exact same thing as eqing, just more convoluted, less precise, and with less options. in an eq, giving a bump to those frequencies on the original sample would do the same thing.

the idea of layering is not to stack duplicate sounds but to add what's missing. ie if you have a bass multisample that is just plain weak in the lower bass range, you can add a sine wave and tune it down until it's adding what you wanted.

the idea of stacking duplicates wouldn't be a bad move in the MPC if it weren't for the fact that it's not sample accurate. since it's not sample accurate, your stacked layers do not always play at the exact same split second time. as such, there will be a phase sound as the two identical frequencies play slightly off time with each other. you'll hear it. it's not desirable.

I totally understand what you're saying...but, this is MUSIC, not mathematics.

it's an asr-10 trick that i use for 90% of the beats i make. on the asr, i filter the sample to nothing but the low end and layer it on the sample. only THEN do i eq it.

also, i think you think that i simply lay the filter on the sample without manipulating it at all. untrue. obviously, that would sound nasty. the filter is the low, and the original sample becomes the high, THEN eq and compress. so, what im doing is recreating the natural bass (and high) of the sample rather than playing (bass) over. basically, it's more surgical than eqing.

listen to any 90's era hip hop..

Just try what he is saying, it just might be what you are asking.

MUSIC, not mathematics...listen to any 90's era hip hop..

music is 100% mathematics, especially when we're talking about the frequency spectrum
regarding 90s hip hop - welcome to hard drive, i think you'll love it here

from your original post, it sounded like you were just trying to make the bass louder by stacking it with the original. due to the phasing, that isn't gonna work, and furthermore, if you aren't going to snip the original sample's low end off before applying the filtered sample, the process is still nothing other than blunt force EQ.

in your last post you've begun to describe something different: the equivalent to running a sound through a band splitter and manipulating however many different frequency bands you're working with (in your case, 2 - "low" and "high").

you say that the original sample BECOMES the high - it's only going to become the high if you in turn filter out the low end that you're going to be replacing with your filtered/eqed sample. ie, both samples must be filtered, otherwise you're needlessly overlapping (and as such, might as well use a single sample with EQ to bump up the values where you're overlapping).

so ideally, 1111222233334444 becomes

11112222 and 33334444, and you can then manipulate them individually all you want

when played together, no phasing or issue would occur since there's no overlap. this would also allow you to manipulate each band separately.

however, this is still no different from coyping the sample and equalizing it differently. filters/eq are the same thing idea, eqs just have far more bands and aren't restricted to 3 types/1 cutoff point (as they are in the mpc)

eq by nature is going to be more precise than an mpc lowpass filter. appliying a 20hz rolloff, a slight bump at 5khz, or finding the sweet harmonic spots are all things you can't do with a blunt filter (no, not that kind of blunt filter)

from what you described in your last post i think you could get away with either filtering or eqing, but you are going to have to have 2 copies of the sample - either recorded w/ different eq (in effect separating it into bands) or simply resample the original twice, one with a high pass and the other with a low pass (also separating it into bands)

then you can chop each sample identically and create different programs for them, and then sequence them separately

wow that was nuts, i had no idea i wrote that much
made myself late haha i'm so lame...

music is 100% mathematics, especially when we're talking about the frequency spectrum

no. this ain't a god damn classroom. music is a feeling, not a theory. you really thing the plantation-era bluesmen gave a rats ass about frequency spectrum??
The use of a turntable violates ALL of the established rules of music, homie. this the deconstruction of music laws. now, go listen to some early wu-tang records RIGHT NOW.

from your original post, it sounded like you were just trying to make the bass louder by stacking it with the original. due to the phasing, that isn't gonna work, and furthermore, if you aren't going to snip the original sample's low end off before applying the filtered sample, the process is still nothing other than blunt force EQ.

yea. no $hit. obviously there is some eqing.... JUST LAYERING them (to make the bass louder) would sound like $hit. and, the og is manipulated so the two compliment each other with the drums and whatever else.

you say that the original sample BECOMES the high - it's only going to become the high if you in turn filter out the low end that you're going to be replacing with your filtered/eqed sample. ie, both samples must be filtered, otherwise you're needlessly overlapping (and as such, might as well use a single sample with EQ to bump up the values where you're overlapping).

exactly! the orig is filtered (or eq'd, or compressed, or some effect added to it) to remove the low end. the filtered sample (the copy) then becomes the low end (bass).

however, this is still no different from coyping the sample and equalizing it differently. filters/eq are the same thing idea, eqs just have far more bands and aren't restricted to 3 types/1 cutoff point (as they are in the mpc)

sure. but, i like my way.

as a matter of fact, i went through the process last night. copied the sample, filtered it in a new program, adjusted the og (and the drums), then laid them nicely in a bed together...then ran it through the asr compressor...added some delay and it sounds ILL.

no, it's not a classroom - this is the real world. every single sound you make falls within the frequency spectrum and follows those rules. this isn't about cultural innovation like "gol' dang, he's spinnin a record backwards" this is about the science of what makes sound sound. there is nothing metaphysical about it - it's governed by rules and laws of mathematics. you can't pump 20 ccs of spirituality into a spectrum analyzer, there's simply no place for it to hide.

you really thing the plantation-era bluesmen gave a rats ass about frequency spectrum??

nope, but if they opened their mouths and made a peep, they played by math's rules. and as a hip hop producer, not caring about the frequency spectrum is going to result in bad mixes, disagreeable layers, and overall bad music. you shouldn't get so worked up, after all your post is ABOUT caring for the frequency spectrum, so i have no idea why you're determined to take this so aggressively.

no, it's not a classroom - this is the real world. every single sound you make falls within the frequency spectrum and follows those rules. this isn't about cultural innovation like "gol' dang, he's spinnin a record backwards" this is about the science of what makes sound sound. there is nothing metaphysical about it - it's governed by rules and laws of mathematics. you can't pump 20 ccs of spirituality into a spectrum analyzer, there's simply no place for it to hide.

I'm still unimpressed with your pedestrian, musical-meets-algebra jargon. Music is an aural artform that fails once married to to such constructions......sure, there are certain absolutes, but the CONSTRUCTION of music is not a damn midterm. people make what sounds good and worry about the constructs later.

....am i saying that the science of sound and math are purely human constructions? yup. and they matter roughly 0.000001% to music in my opinion.

nope, but if they opened their mouths and made a peep, they played by math's rules.

sure....when you open your rectum to take a dump, you play by math's rule also. does that "rule" influence the music? was it a compelling force for leadbelly belting out "blood done signed my name?" probably not.

and as a hip hop producer, not caring about the frequency spectrum is going to result in bad mixes, disagreeable layers, and overall bad music. you shouldn't get so worked up, after all your post is ABOUT caring for the frequency spectrum, so i have no idea why you're determined to take this so aggressively.

true. but if it sounds good, it sounds good.

listen to Live Convention 81. classic hip hop tape. then tell me about bad mixes.

i'm at a loss for words here. i'm just real glad i'm not trying to impress you.

the terms of this conversation have changed several times since i first posted a helpful response and they continue to do so.

well...thanks.

Nym wrote:

MUSIC, not mathematics...listen to any 90's era hip hop..

music is 100% mathematics, especially when we're talking about the frequency spectrum

i disagree. music is 100% expression