MPC 3000/60 Swing

I'm curious to know how the swing actually affects the note placement when applied to a quantized beat.

Could someone please post a copy of a midi grid displaying the before and after of how the note placement changes when swing is applied to the drums.

I'd like to see where it moves the notes and if it actually moves each note to its own specific spot, or just slides the backbeat around together as group like other sequencers.

I'm not sure if what I said makes sense, but if you could just type out what the grid from your microscope edit screen displays I'd really appreciate it.

Thanks alot

The grid edit is completely different then what you are talking about , its not really that visual. I just tells you what pad has been recorded at what time in the sequence.

distantb wrote:The grid edit is completely different then what you are talking about , its not really that visual. I just tells you what pad has been recorded at what time in the sequence.

Thats actually exactly what I was talking about. I just wanted to know at what time in the sequence that the pads/drumhits are recorded depending on the swing applied, and whether or not every note is moved differently or just groups of notes moved together.

For example, on my fantom in a basic, 4/4 Kick/Snare/Kick/Snare pattern with 16th note HH's, and no shuffle/swing applied the grid would like something like this:

Note: (my fantom has 480 ppqn, but I used 96 ppqn in this example since the MPC has 96.)

(Midi Ticks)
0 96 192 288
K S K S
___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___

0 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360
HH HH HH HH HH HH HH HH HH HH HH HH HH HH HH HH
___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___


When I apply 70% 16th swing, the Kicks and Snares stay exactly on the 1-2-3-4 counts, but groups of the HH's are shifted together the same amount throughout. For example,

(Note/Pad position)
0 96 192 288
K S K S
___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___

0 33 48 81 96 129 144 177 192 225 240 273 288 321 336 369
HH HH HH HH HH HH HH HH HH HH HH HH HH HH HH HH
___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___


You see how the 2nd, 4th, 6th, 8th, etc. HH's were all shifted equally by 9 ticks.

I guess my question is does the mpc 60/3000 apply swing in the same way, or do the 2nd, 4th, 6th, 8th, etc. HH's get shifted each a different amount, like maybe the 2nd moves 8 ticks, the 4th moves 10 ticks, the 6th moves 9 ticks, the 8th moves 12 ticks, etc. which would create a more natural, realistic feel since no drummer could play perfect like my example. Also, does the MPC swing shift the 1-2-3-4 counts around as well?

I always thought that is what gave the MPC its unique, "realistic" swing in that it moved every note a different amount, unlike all of the other sequencers that move each note the same amount leaving an unnatural, "robotic" feel.

Could somebody post a grid following the one in my example showing how the swing affects the pad/drumhit position within the sequence?

Unfortunately the spacing didn't show up the same after I submitted the post, but you can still tell by the numbers I think, the position of each drum.

Sounds like your on your own on this one I don't think anybody wants to do all that.

jk81 wrote:
I guess my question is does the mpc 60/3000 apply swing in the same way, or do the 2nd, 4th, 6th, 8th, etc. HH's get shifted each a different amount, like maybe the 2nd moves 8 ticks, the 4th moves 10 ticks, the 6th moves 9 ticks, the 8th moves 12 ticks, etc. which would create a more natural, realistic feel since no drummer could play perfect like my example. Also, does the MPC swing shift the 1-2-3-4 counts around as well?

I always thought that is what gave the MPC its unique, "realistic" swing in that it moved every note a different amount, unlike all of the other sequencers that move each note the same amount leaving an unnatural, "robotic" feel.

Could somebody post a grid following the one in my example showing how the swing affects the pad/drumhit position within the sequence?

We all know that 'swing' works by taking every other note from a 1/16 or 1/8 pattern and delaying it by a set 'amount'. All the moved notes affected by the 'amount' control are moved an equal distance from the original programmed position. Swing does not have any effect whatsoever on 1/4 notes or 1/2 notes.

But on the MPC sequencer there are slight variations to the placement of notes whether swing is applied or not (I have done some tests which you can find in a post I did a year ago on synchronising the MPC with Pro Tools viewtopic.php?f=11&t=80943 extract below). These slight variations are what I refer to as 'the MPC swing' and have nothing to do with the actual swing controls.


Now how good is my sync I hear you ask? Well it does the washing up and its always there…. No seriously, very good, reliable and predictable. I ran a few tests to see exactly how the PT sequencer would behave when slaved to the MPC4000. There were some interesting results.

OK so I have midi timecode sync setup and ready to go. An empty sequence is loaded in the MPC and a new session in PT. I call this THE STEP TEST and used it to see how the sequencer of PT and the MPC interact. The MPC swing setting is set to 50% (off) for the purpose of the test, timing correct is set to 1/16 and is activated.

On both the MPC and PT I have matched the tempos and set loop to ‘off’ as if I were tracking from the MPC. On the MPC I hit ‘go to’ and ‘start’, this ensures both sequencers begin the test at bar 1 beat1.

I now use the MPC’s <step> button to advance the MPC sequencer one 1/16 step (240 ticks) at a time. Each time I do this I record the position of the PT sequencer shown in the transport display.

TIP: the resolution of the MPC sequencer is 960 parts (ticks) per quarter note (ppqn). These are the numbers shown after BAR-BEAT- XXX in the sequencer display.

MPC4000 PTLE7 OFFSET

1-1-000 1-1-000 0
1-1-240 1-1-230 -10
1-1-480 1-1-461 -19
1-1-720 1-1-691 -28
1-2-000 1-1-922 -38

1-2-240 1-2-192 -48
1-2-480 1-2-422 -58
1-2-720 1-2-710 -10
1-3-000 1-2-941 -19

I performed the same test using the <<bar>> button to advance the MPC sequencer.

MPC4000 PTLE7 OFFSET

1-1-000 1-1-000 0
2-1-000 1-4-922 -38
3-1-000 2-4-941 -19
4-1-000 3-4-902 -58
5-1-000 4-4-922 -38

The results show that there are some very minor differences between the two sequencers timing. The offset figures show that PT is always slightly behind the MPC although the nature and degree of the inaccuracies appear to follow a pattern. Take a look, in the first test we have offsets of -10, -19, -28, -48, -58 and then the cycle repeats itself. The same can be recurring pattern is seen in the results of the second test, we have timing offsets of -38, -19, -58 and then the cycle repeats. And the same offset numbers appear in both tests (19, 38, 58...) Very interesting…..

The same results were obtained using midi clock instead of midi timecode.


The outcome of these tests can be summarised by saying that the MPC4000 used in my experiment had a natural tendency to drift, ie to play notes earlier than their programmed position. The amount of negative drift I recorded was up to 58 ticks early as shown above and was cyclical in nature (not random). As we all know drums that hit early add a sense of urgency to the rhythm and it is this phenomenom that give the famous MPC drum sound IMO.

jk81 hope that info helps a little. Dont let this 'swing issue' become an obsession, use it or dont use it. Simple!

Thanks Clint. I really appreciate you taking the time to experiment with it. I've been a musician most of my life, but I'm also going to school for computer science, so I have an interest in the technical side of music as well.

My main reason for wanting to know about all of this is to decide whether or not I could use an MPC in my setup.

I currently record all of my music live w/o any 'quantize' or 'swing' used, but that can become very time consuming getting the timing how I want it. I'm really opposed to using quantizing/swing because every sequencer I've used has such a rigid, monotaneous result. When I heard about the MPC 'swing' and its natural/realistic timing, I thought that could be extremely useful and speed up my workflow.

The only thing I could think of that would make it different from the rest would be that each note got shifted a different amount much like a recording of a drummer in real time where each hit was slighty off by a little, but not all off by the same exact amount.

In other words, the MPC must be set to templates of actual recordings of a live drummer. If that is how it works, it would be definitely something worth investing in.

Thanks for all your help. I think I'll try one out myself and see how I like it.

Do they all work this way? I've heard that the 3000/60 had the most realistic timing. I would probably look at getting a used 2000 or 1000, maybe even a 500.

well as far as the swing and offset of my 2500, i love it. you can tweak it to do some funked up stuff. that's all i know. sorry i can't directly help. honestly i imagine the 60/3000 can be emulated on a 2500. you just gotta know what you're doing. and of course its not the "real thing" but come on...

Maybe you should actually listen to the device in question to see if you like it because the term "realistic" is totally arbitrary.

Regarding the "test".. Is that midi or audio? I've also noticed slight regular timing differences when recording a "straight" MPC 1000 pattern into a computer sequencer but I can't really draw any conclusions from that because the computer and midi interface were low-end w/no midi timestamp system to make sure I was getting an accurate result.

as far as the "Human feel" goes or swing... for me its a bit too sloppy.... i come up with better/cleaner drum patterns on the 2k or XL then on the 60...

elektrik_muz wrote:Maybe you should actually listen to the device in question to see if you like it because the term "realistic" is totally arbitrary.

Regarding the "test".. Is that midi or audio? I've also noticed slight regular timing differences when recording a "straight" MPC 1000 pattern into a computer sequencer but I can't really draw any conclusions from that because the computer and midi interface were low-end w/no midi timestamp system to make sure I was getting an accurate result.

The tests I carried out were based upon midi timecode and midi clock pulses transmitted from my 4000 to my Digi 002 based Pro Tools LE system. The MPC and Pro Tools were synchronised using the above methods. No audio was involved.

The results were consistent regardless of the method employed enabling me to reach an accurate conclusion.

The results only serve to validate something which is often confused with the 'magical' effect of the MPC's swing function. The swing on thew MPC has exactly the same effect as the swing on your DAW (delaying 1/16 & 1/8 notes by a predetermined amount); it's the timing drift, slightly ahead of the beat that provides the famous MPC groove.