hhhaaa, well a Tempest would turn me into a dave Smith instruments fanboi, tis true.

Kind of irrational I suppose, but I've always been wary of a box with no audio out. I was looking at Kore l, then Kore ll came out, which didn't seem like progress, so kind of glad I missed all that.
shall look into this further, lots of choices.

I've never owned an MPC of any kind, but people really do swear by them. I think there is a simplicity and lack of options in using one that makes for a very productive environment. I've viewed tutorials on them and to me they never seemed anything special, but I have to admit that I'm not much of a "sample" guy. I've never had the urge to chop up some beat I found on an old record. I don't care what a riff from Kashmir will sound like in my song. I'm very content in letting it stay in Led Zeppelin's song. Anyway, I know this isn't the only way to use an MPC but it seems to be very popular.

However, I did buy Maschine and there is a lot to like about having a well integrated piece of controlling software. I wish they'd revamp the song mode though, because it currently sucks. If it existed when I was shopping for such a thing I may have gone for this instead: http://www.akaiprompc.com/mpc-renaissance because in all honesty I think that while hardware is really good for some things, like tube distortion and analog synthesis, I never found a hardware solution that gave me the flexibility and speed of working in a DAW. The last "groove box" I owned was a Emu Commandstation and frankly it was very clumsy to work with compared to Ableton Live. I don't know Reason 6 but I've heard it's very fast and fun to work in as well.

So I doubt the Maschine or Renaissance is in your price range but they're worth keeping on your radar. Even NI's Maschine Mikro could be good and at $350 it's a bargain. The problem is that it won't integrate into Reason (in fact, nothing will, Reason is a closed system) so if you're sold on Reason maybe consider an Akai controller that has some keys and MPC style buttons. One of the audio guys I worked with had one and he liked it a lot.

No offense, but I find this kind of "ultimatum" advice to be a bit myopic, especially in modern times. While this might have been true in the 90s, you'd be unnaturally limiting yourself by taking this advice especially in these times where Moog is shipping analog synths that have a VSTi controller and Akai and Native Instruments are producing well thought out hardware-software systems that truly give you the best of both worlds.

Look at it this way. Say you like to do sample resynthesis. Sure, you could buy a Roland V-Synth but be prepared to sink a thousand dollars or more into that hardware and end up with something far less capable than Alchemy or Poseidon. Maybe the money is worth it if your sound is all about the resynthesis, but if it's just an occasional "spice" you add to your mix, then it's a ton of money to sink into that tool especially when for $200 or so you get a LOT. (I think they sound better too) Same thing with analog synth sounds. Maybe you just want that every now and then. Software will get you so damn close it won't matter that you don't have a Moog. Maybe that Minimoog sound is central to your style so of course nothing else will do. There's nothing wrong with having all your percussion be sample based and use hardware analogs for instruments... as you progress you'll find you gravitate toward some things and away from others. It's all a matter of personal preference and style.

In the end I think keeping an open mind and using a hybrid system is really the best way to go.

zerocrossing, before I start, I will just say this. I have read your posts and find you to be more experienced and knowledgeabe in general than I am. I therefore have the utmost respect for not just you, but your opinion.

I'm not sure if you got my drift though. Do you go back to the age of the 286 and the glory days of the 386?

This is not me saying this, this is other people that made records from that era, that I happened to know. I have nothing to do with them. I have no vested interest. I am TOTALLY SOFTWARE - well maybe a couple of akai controllers and an ensoniq kbd...


The clock on a drum machine, be it an Akai or an Alesis, is more solid than the clock of your general pc because of interrupt requests etc...

I did not make this up. It has been said to me by everyone at that stage of the game - they all use software as well - they love - they love it so much they buy it and use it.


Please don't take what I said as an ultimatum - I never said: Do this or I shall do that! I don't care what you do. And short sightedness is not what I am suffering from - maybe long-sightedness.


I was upset when they told me too. I thought: they can't be right: reactionary freaks. Until I measured it. It's not difficult, you don't need a TR-808, an Alesis SR-18 will do, for example.


Anyway, no hard feelings. Just trying to set the record straight.
My argument was about computer architecture.
About the very integrals of hardware versus software.
Maybe somebody that knows a whole lot more about this could help us out?

Please tell me where I am wrong and I will accept it with good grace.

Pun intended?


Did you know that the MPC2000 had a 386 for a processor?


You're both right. The MPC and other such hardware sequencers had great timing that isn't quite the same with DAW's AND there are factors beyond timing alone that are as important or more. I'm curious how the MPC Fly's timing will be, that could be a nice meeting in the middle.

Don't know about the fly, but over at MPC-forums.com the Akai engineers were boasting round trip (MIDI in -> audio out) latency of 2ms. They claim they had a custom high speed MIDI driver developed, the same tech used in their NS7/NS6/V7 Serato controllers.

By the way, can we please put the 'magic thing to bed? The 'tight' reputation is largely based on two factors:

1. being able to accurately record MIDI in real time unquantized. can't do that on computers without expensive workarounds.
2. the coarseness of 96ppq (in most models), and the way that coarseness interacts with the swing settings.

At that resolution, the distance between any two adjacent ticks is about 5ms (at 120bpm). Which means that whatever you play into the box (unquantized) gets repositioned to the nearest point on that grid. Which means you have a max variance of 2.5 ms (at 120bpm).

(sidenote: theoretically speaking, this is the situation with most well designed hardware sequencers (at least the standalone ones), as well as Atari's etc. Whereas on the computer, with MIDI jitter/latency and all, a given note may be recorded anywhere from 3-4ms up to 20ms or more late, depending on a variety of factors, and because of audio latency, you won't hear the audio for another 5-7ms at the very least. And if you send that MIDI note back out out, there's another round of MIDI latency/jitter to deal with.)

Thus when you apply swing quantize, what then happens is you are moving each of those notes around that 5ms (at 120bpm) 'grid', and IMO its a chicken/egg thing, where it sounds right because you've heard it on countless records (kind of like Lexicon reverb in movies). Nothing magic, just a well designed system and its inherent capabilities/limitations.

Cool!


Can you explain more what you mean by this?


That's a good point. I just checked and didn't see any way to do that in Maschine.

Well USB MIDI is subject to the constraints of the operating environment. Namely the interrupt system the CPU uses to determine which incoming/outgoing data to process and when. What that essentially means is that while one MIDI message may make it through in, say, 3ms, then next one may be held for 15ms or more before its passed on to the host. In addition, the host itself, depending on its internal MIDI design and the load its operating under, may or may not process that MIDI message immediately. Those are the primary sources of MIDI jitter (variable delay), AKA the drunken computer syndrome. The same problems apply to outgoing MIDI notes/CCs and incoming/outgoing MIDI clock and MTC messages. By the way, serial and PCI/PCMCIA MIDI interfaces fare a little better because they have a more direct pipeline to the CPU, but they still have some measure of jitter.

There are ways around the problem, but they all require purchasing rather expensive hardware: The Expert Sleepers ES4, the Innerclock Sync-Gen, or the Advanced Pro Gear MIDI Bridge 120.

MPCs, Ataris, and most other well designed hardware sequencers don't suffer from this problem, because they are designed to constantly poll for incoming/outgoing MIDI messages (as opposed to having to be informed that there's a MIDI message there for them to process), being that that's their raison d'etre... Once a MIDI message is received, its sent on its way almost immediately (but more importantly, consistently, i.e. without jitter), and the only delays are related to how fast it can either send it back out or how long it takes for the internal sound engine to produce a sound. Which, again, is the machine's #1 priority unlike a general purpose computer. So even if there is latency, there is usually very little discernable jitter.

You can work around latency (a predictable, consistent delay), you can't work around jitter (an unpredictable, inconsistent delay). Unless you're prepared to quantize everything (which is why step sequencers and groove quantize is such a big deal in computer MIDI land...). Nature of the beast my friend.

The Advanced Pro Gear unit seems like a good solution. There website isn't working, do you know if the company is still around?

It might be, or it might never happen in entire life of a man. Actually, I want to know where such numbers for small packets on usb bus are from? IMO all of them would have <1ms throughput time and unless there's 16 channels of raw audio on the same exactly bus there would be no visible latency for a messages.

Be carefull with maschine over promising

This was from about ten years ago:

http://www.soundonsound.com/sos/sep02/articles/pcmusician090 2.asp


USB is not well suited for time-sensitive data. This is a fact, not conjecture.

While that may be the case, it is only relevant should you be using a VSTi 'live' or if your sequencing a whole load of external gear. If your computer is triggering virtual instruments with its midi, the whole notion of MPC's having a 'tighter feel' or whatever you want to imply is irrelevant. If you record a 'take' of a couple of bars, any supposed 'usb latency' captured by your sequencer is easily rectified and a beginner will almost certainly quantize the data anyway.

As for the external gears timing, I guess you would have to record a 'master groove' track from your gear into your daw and rid the waveform of these 'usb errors' by using its sample accurate editing. Then its a simple case of quantizing the other 'sloppy' tracks to your fixed master groove track - that is if it bothered you that much....

The more I think about, the more ridiculous the thought of an MPC500 being considered over Reason 6 becomes - A controller and Reason is so much more powerful than an mpc500 that its laughable. Even if you just want to make beats Reason is way more powerful, not to mention its a complete daw now with audio recording etc.

Just MHO.

having read some of this last night(...), i'd say to the
OP, get an MPC...and get a DAW, both. the mpc500 is very
wicked piece of kit, going by the spec: flash ram to load wavs:
total luxury..maybe think of a mpc1000 with more outs and a
higher spec/feature set?(no battery jamming though).

and that way, you can GET AWAY from the bloody 'DAW'/PC/Mac for
a bit, and absorb yourself in the relative simplicity of a dedicated
operating system. you'll approach the whole thing differently, and it
may become the primary in your working method, leaving the computer
to its auxiliary role of manager/recorder, etc.

in any case, you'll generate material you can use in a ('DAW'-sic-),
and it is one of the wickedest bits of digital audio gear out there.
one thing you can say about akai is that they know sampling. not the
most feature-rich machines, but once you've been using one for a
while, you come to appreciate the directness and simplicity.

so, whatever you spend on your chosen computer sequencer, the mpc
won't be wasted money.

The whole point is that it's not capturing the performance accurately. You aren't always using quantize on an MPC. For example, hi-hats in hip-hip are typically recorded without quantize in order to retain the feel of the beat. Some producers won't quantize anything at all, and the accuracy of the performance becomes very important in those cases. If there is a lot of jitter in the signal, this is a problem. It's not something that is easily rectified or that you can ignore.

Now, I'm not arguing that someone should buy an MPC over software and a controller. I really don't think that the jitter issue is going to matter much for the way 90% of the people out there produce music. But you can't dismiss it outright. There are going to be cases where it makes a noticeable difference (like Araabmuzik's style for example). And for those people, an MPC may be a preferable solution.