Forum to discuss all matters relating to the MPC1000 and MPC2500 operating systems created by 'JJ' (all versions).
User avatar
By Antonym Thu Nov 22, 2007 4:32 pm
this is looking to be a long series of posts from me. please please please add and comment on my ideas.

velocity range was up until the announcement of os2 one of the most important (if not THE most important) parts of my workflow. when it was announced that velocity range of layer was going to be removed in os2, i immediately started thinking about how to get around this.

here i am just flexing my mind trying to come up with different ideas. there are pros and cons to every idea. please add them if you see them.

THE SITUATION: previously, multiple samples could be triggered independently on the same pad via QLINK-layer or velocity. i used this to create very realistic drum patterns via arranging the drum samples from softest to hardest, depending on the strength of the hit. this will no longer be possible in OS2.

MY OBJECTIVE: to figure out a good way of emulating this. using the same pad for every snare/etc does not provide the subtle variance necessary to create realistic, multitimbral drum hits that sound significantly different when struck at low velocities compared to high velocities.

THINGS TO AVOID: overusing polyphony. some of the techniques i've been considering have the potential of eating 3 or more voices with every hit of the drum. using more than 1 pad bank. having to switch pad banks while playing drums is a pain. in drum programs, multiple pad banks can be used either for simulted layers OR a different drum set. however, for ease of use, it's best to be able to access all of one's current drum kit from a single bank.
Last edited by Antonym on Thu Nov 22, 2007 6:42 pm, edited 1 time in total.
User avatar

By Antonym Thu Nov 22, 2007 4:40 pm
idea one: relocation, old style.

instead of layering 4 samples velocitatively across 1 pad, put those 4 samples on 4 consecutive pads, vertical or horizontal.

PROs:
-instant access to all 4 "takes" of the same drum sound.
-the program's midi note assignment can be edited so that all pads are "grouped" together on the midi grid, so they don't appear all over the place.
-additive layers can be created individually (example, each kick is layered with a SINE wave) or "grouped" (example, each kick is simulted to a single SINE pad in B bank)

CONs:
-harder to play. hands are all over the place. though this in the end may be a good thing for practice and muscle memory.
-overuse of a pad bank. what previously took up four or five pads now takes up closer to 16. additional percussive material must either be put into a 2nd pad bank OR a new program. limits instant accessibility of different samples.
-requires a ton of parameter tweaking. velocity-level must be set appropriately for each pad used.
User avatar

By Antonym Thu Nov 22, 2007 4:46 pm
idea two - simult + velocity->level.

simult 1 snare pad (hard hit) to another snare pad (med hit) and another snare pad (soft hit).
set snare pad 1's velocity-level to 100, set snare pad 2's velocity-level to 70, set snare pad 3's velocity-level to 30.
hit pad hard. note: all 3 snares play, hard hit is predominant due to louder sample.
hit pad medium - hard snare is relatively quieter due to broader velocity-level. med snare is (should be) predominant
hit pad soft - med and hard snare are comparatively quiet. soft snare is loudest due to short velocity-level range.

PROs:
-similar setup to original velocity-range technique, removed in os2.
-in some ways better, due to there not being sudden changes in which sample is playing but rather a more gradual shift toward current predominant sample based on hardness of strike
-since you can simult to additional pads in d bank, this doesn't hog the pads like the first idea did.

Cons
-polyphony hog. 3+ voices are eaten with every strike of a drum pad.
-appropriate velocity-levels will be different with every sample. will require a lot of tweaking.
-may be difficult to avoid phasing since sounds will always be layered, albeit at different levels.
User avatar

By Antonym Thu Nov 22, 2007 5:08 pm
**CURRENTLY MY FAVORITE SOLUTION, actually more powerful than os1's velocity layers**

idea 3 - selective playback. (synthesis of idea 1 and 2)

think of your drums in terms of 2s. when chopping from break, get 2 kicks, 1 soft and 1 hard. upbeat, downbeat. get 3 snares - 2 hard, 1 soft. get 3 or 4 hats with a variety of force.

assign you drums, group them with midi note assignment. (capital/lowercase = harder/softer, _=blank pad or other sample)

K S H _
k S h _
_ s H _
_ _ h _

simult your kicks/snares/hats with SEPARATE layers. those simulted layers can be the same sound (example - same snare additive for the SNARE pads) but change the velocity-level and velocity-filter of the simulted pads so that striking the snares will provide different at different velocities.

PROs
-good synthesis of the above
-easier to play
-moderate use of pad real estate
-about as easy on polyphony as the original os1 velocitative layers
-due to velocity-level and velocity-filter of simulted pads, a more varied sound is performed than in os1.

CONs
-still not as easy to play as os1 velocitative layers due to additional hat pads
-takes more pad real estate than os1, but not a serious con.


*****for now, this is my favorite idea.

By tikivablinki Thu Nov 22, 2007 10:34 pm
Nice ideas! Interesting.

I dont use velocity-layerswitching that often, since I usually work with single electronic one-shot samples, not chopped from breaks.

Here is my way of how to make low-velocity-hits sound a bit more realistic, or in the end different from the original signal. Maybe it´s more for electronic samples, that don`t sound that realistic anyway.

Velocity -> Sample-start (Adjustable. Found in the amp-envelope page)

Lower velocity pushes the samplestart forward by selected amount.
This just cuts the attack from e.g. a snare with decreasing velocity. The ghostnotes still have a little attack, but the sample gets shorted also.
The result depends strongly from the sound of the sample that is used.
(Meaning sometimes it works, sometimes not so nice)
May also make clicks on certain startpoints. Especially with bassdrums or other material containing low frequencies.

If its clicking, try adding a tiy bit of velocity-modulated attack. This function is also found in the amp-envelope page. Right above the starttime-velocity-parameter.

Well, maybe not that kind of sound one is looking for.
Just a suggestion...

Since I cant see this feature in the pictures from OS2, I´m afraid this may be gone aswell. Hmm - than this technique wouldnt help either ofcourse.
User avatar

By Antonym Fri Nov 23, 2007 4:29 am
oh man, badass! that's SUCH a good idea

velocity-start and velocity-filter really can be powerful, especially with simulted layers

this kind of thinking is what i need to do, man. it's like the old way, except way more powerful.
User avatar

By Antonym Fri Nov 23, 2007 4:36 am
err...scratch that...velocity-filter and velocity-start are both going to get deleted...in favor of bringing range layers back.

Image

it is sad that either one or the other must go. i enjoy both range layer AND velocity-filter, and i feel like i have only just begun to appreciate velocity-start.

oh well. these ideas still work, and will be useful

long live range layers. amazing how exciting even the pre-game to os2 has been.
User avatar

By mikolo Fri Nov 23, 2007 12:59 pm
hey nym,
I checked this thread yesterday but didn't have time to reply!
I get the impression JJ's reintroducing layers at the expense of velocity start etc? this would be a shame cause i use that too, although not as much as layers.
Using simult as a work around would be ok, but backwards compatability is important for me. (I've been making layered drum programs since i got my mpc- i use them for live stuff).. although i could spend a few weeks/months converting everything..

is it due to processing power that the sacrifices have to made? .i.e the mpc can't handle a program with layers/adsr/etc all included.

Could JJ make more options on the type of program you want to make
for example a drum program would be like the current program with layers etc, but without the adsr, an instrument program could have adsr without layers. you could use one for multisampled layered drums etc (which in my opinion an mpc should be able to do well!) the other for more sound manipulation synth stuff?
User avatar

By stan steez Fri Nov 23, 2007 4:04 pm
On the real: stop waisting your time guessing about production techniques for an OS, which is not even laid down yet! And that goes for anybody: the os2 feature requesters, the how-can-I-install-OS2-ers... everybody! It's just annoying. Until this OS is out, everything you do is simply speculating on the basis of imagination and of course never to be forgotten: rumors...

Out
User avatar

By undersound Fri Nov 23, 2007 5:24 pm
imagination and hope wasnt part of your childhood then?

Good planning nym :) alot of typing n effort went into that
User avatar

By Antonym Fri Nov 23, 2007 9:40 pm
stan though i understand and agree with your concern for the rabble, i'm by no means among the panicked.

regardless of how this OS is going to turn out, i'm equipped to assimilate it as i need to. planning ahead for the use of new features with existing is simply part of the way i must do things.

is it due to processing power that the sacrifices have to made? .i.e the mpc can't handle a program with layers/adsr/etc all included.


mikolo, i don't know the specifics here, but it appears that is the case. my idea was to possibly go down to 2 layers and preserve as much existing control as possible while adding ADSR. again, i don't know if that'd do it, but because of simult i'd be interested in that way of easing the load.

I think that we have to come to grips with the fact that this OS will not be backwards compatible. as jj said, this is a true new OS. the cool thing is this "quick swapper" ability gives us both os's at the touch of a button. it'll be like having 2 mpcs. similarly, audio tracks really kind of give us a form of reverse compatibility.

however, i do think that while this post was useful just as a brain-buster to look for other useful and usable techniques even in jj os1, it came "too soon." that's part of what stan's saying, we don't know enough to even talk like this. to an extent that's true but like i said, good discussion came out regardless.

mikolo, i'd like to talk more about velocitative drum layering with you. what did you think of my third option?
User avatar

By The Grublet Sat Nov 24, 2007 4:39 am
Um. I missed the boat on this OS2 thing, but what is worth trading for layers?

A Short List would do.
User avatar

By Mr modnaR Sat Nov 24, 2007 9:25 am
The Grublet wrote:Um. I missed the boat on this OS2 thing, but what is worth trading for layers?

A Short List would do.


1. Your spleen.
Image
User avatar

By mikolo Sat Nov 24, 2007 2:56 pm
my idea was to possibly go down to 2 layers and preserve as much existing control as possible while adding ADSR. again, i don't know if that'd do it, but because of simult i'd be interested in that way of easing the load.


so is this linked in with your third idea?

I understand it like this:
each pad has two layers,adsr,velocity>amp/filter/start and simult.
and you create 4 layers by simulting to another pad.

If this were possible that'd be great, i suppose we'll have to wait and see.
User avatar

By mikolo Sat Nov 24, 2007 2:58 pm
I use your first method from time to time anyway, usually if I just want to start making a beat quickly.

I understand your second method as refering to JJ's original plan:
layers but without velocity ranges.But still including vel>amp/filter/start.

This wouldn't work for me i don't think.Not unless JJ could create a velocity trigger range for each pad rather than each layer.
If that were the case id be very happy too cause you could use simult.(similar to your 3 idea but with a different workaround on JJ's side).

anyway..im gonna stop speculating now don't want to annoy Stan... i know his steez... :wink: no disrepect stan i know what your saying! :D