Post your questions, opinions and reviews of the MPC1000. This forum is for discussion of the OFFICIAL Akai OS (2.1). If you wish to discuss the JJ OS, please use the dedicated JJ OS forum

By mpc3000 Fri Jan 23, 2004 12:54 pm
No, this is not my forum. I would not want that repsonsibility.


"which have the ability to place audio on a sample accurate level"

I didn't say anything about sample level placement. I speak midi.

By tiger vomitt Fri Jan 23, 2004 4:34 pm
OOPS my bust, then the props goes to you MPC TUTOR, genius ruler of all things educationally MPC! i salute you.

sorry for talking out my a$$ homey!

By jigginz Sat Jan 24, 2004 3:45 pm
Hi
Sorry to drag all this back up again, but i just wanna get some opinions on the midi timing issues as discussed earlier in this thread.
I know this doesn't directly apply to mpc's coz they trigger their sounds internally, but how is it that higher sequencer resolutions don't make any difference to midi note timing when they are quantized?
I mean, i have been experimenting with a simple 4/4 kick/snare combo running at a fairly high bpm sequenced on an Mmt8 and when i double it's resolution it just sounds much tighter, kind of a more rigid quantization. But how can this be if these two simple beats are quantized hard to a 4/4 time sig playing on top of each other.
I mean surely i shouldn't be hearing a difference.
I know there's been much discussion on this forum about resolutions picking out playing nuances but i just don't get it when it comes to quantization?
User avatar

By MPC-Tutor Sat Jan 24, 2004 4:12 pm
I believe the forum member 'MPC3000' has pointed to some scientific source that says humans are incapable of differentiating beyond 96ppqn resolutions - any sequence that is recorded at a higher resolution will sound the same to our ears.

As you say, quantising a beat to 1/16 is quantising to 1/16 no matter waht sequencer or sequencer resolution you set your machine to. That snare is coming in at 01.02.00 no matter what. Does it sound different if you double the BPM and adjust to compensate? No.
User avatar

By Lampdog Sat Jan 24, 2004 4:57 pm
mpc3000 wrote:"however i also use my well trusted Logic Audio Platinum 6 setup with the EXS24 soft sampler, Korg Triton, etc, and the fine tuning in Logic is by far the best in any software setup, better than the MPC 4000, DP, Pro Tools, pretty much anything. "

Roopesh, I find it hard to believe that you actually use and understand all that gear well enough to come to that conclusion, but you still cannot figure how to edit a sequence in the MPC1000 to move a sample one tick.


I've noticed that also from previous posts.

By jigginz Sat Jan 24, 2004 5:49 pm
Hi Mpc Tutor

As you say quantizing is quantizing and i don't dissagree with that so why can i hear a difference?
I even tried it backwards to see what happened and after going back twice, effectively 24ppqn, and then swithing to the originally converted pattern at 192 the difference is comical!
You might want to see if you can borrow one from somewhere just to have a play with coz I really don't understand what's going on with this sequencer.
I just want some answers coz would this be the same for all sequencers?
User avatar

By MPC-Tutor Sat Jan 24, 2004 6:17 pm
Why not record three 1/16 quantised sequences for us? One at 90bpm, one at 45 bpm and one at 180bpm, compensating for the tempo change each time so the notes fall in the same place. I'm interested to hear the differences, because mathematically, there should not be any differences - if the snare comes in one the second beat, it should always come in on the second beat, even at double time or half time.

By jigginz Sat Jan 24, 2004 6:53 pm
Hi Tutor
I would like to do this to see if you can hear a difference but i'm quite illiterate in the uploading of Mp3's department so i'm gonna need some tutoring on how to get the loops over to you.
By jksuperstar Sat Jan 24, 2004 6:53 pm
MIDI is an asynchronous (sent whenever, not synced to anything!) serial interface, runs at 31.25k bits per second. The MIDI spec allows 24ppq over this serial line as standard. That's 24 clocks per quarter note no matter what the internal ppq is. To quote the midi counsel directly:

"However, MIDI data being sent from a sequencer can include a number of different parts. On a given beat, there may be a large number of musical events which should occur virtually simultaneously. In this situation, many events will have to “wait their turn” to be transmitted over MIDI. Worse, different events will be delayed by different amounts of time (depending on how many events are queued up ahead of a given event). This can produce a kind of progressive rhythmic “smearing” that may be quite noticeable."

Anything higher is time-stamped when a serial packet gets recieved above & beyond this 24 ppq clock. But it takes a full millisecond to send 1 "note on" for instance (3 bytes at 31.25kbaud). Having higher PPQ just sets when that 1ms begins. (at 80 bpm and 96ppq, that's at best 7ms quantization...equivalent to 1/128 of a second (128Hz), definitely audible to the physical ear, but questionable to the subjective brain).

Since MIDI is serial, as is USB, etc, ...only 1 thing happens at a time. So when a PC plays something over your 8 port MIDI device, things that are intended to play simultaneously, simply don't.

I'll agree with the "Mmt8" being tighter *sometimes*(I think you meant Emagic Amt8)...when used with Logic the Amt8 pre-loads the MIDI data for a given clock (ppq tick) into the Amt8, then sends out ALL PORTS at the SAME TIME...so all your notes ARE tighter. This is a fact, and is why Emagic made the Amt8 series...having a studio with 24 MIDI ports driving an orchestra of modules gets fuNked very quickly. MOTU came out with something similar, but again your sequencer needs to support it. But for 1 port, this really makes no difference. Rememer though, high bpm means less time for each clock tick(ie- more resolution)...things don't get noticable until you slow things down. This is why nobody daisy chains their instruments...they have multiple ports instead (like 2 on the MPC).

As said before though...you quantize, and your quantizing to the 24ppq typically (unless you go over 32nd triplets - ie: 8 32nd notes per quarter, but to get triplets you need *3 (to get the accuracy needed), so 8*3=24ppq, get it?). But the multiple ports still stands even when quantized.

By jksuperstar Sat Jan 24, 2004 7:02 pm
Sorry for the ramble, but point is, if you really need higher ppq, up your tempo and but play your notes at 1/2 tempo, or similar. That'll give you 192ppq in the MPC...defintely below the brains 5ms limits. In the MPC that'll work up to 150bpm, if your tempo is over that, I doubt you can hear the 96 - 480 ppq difference.

By jigginz Sat Jan 24, 2004 7:17 pm
Whoa Jk, you're outta my league!
I'm just talking about a late 80's Alesis mmt8 Harware sequencer!
I just can't work out why the same quantized piece of music(compensated for at different bpm's obviously) sounds tighter at double resolution and even looser when reversed back a couple of times.
That Amt8 program your talking about does sound nice and tight though, might have to take a peek a that one!

By jksuperstar Sat Jan 24, 2004 8:35 pm
the Amt8 is actually a hardware MIDI interface for Mac/PC. Used to need Logic to be tighter, maybe some other software takes advantage of it too(can't see why Emagic would the the world how!)

as for your dilemma, that's a good one. If you've got a PC & MIDI interface, you can run a MIDI-Monitor, which typically timestamps everything it gets. That way you can see where the notes fall with respect to the MIDI clock ticks...might give you clues if it's that important.