By
Jamon
Fri Jan 06, 2012 4:07 am
The big change is that the computing is outsourced to the PC.
It used to be, that to edit video, you needed dedicated hardware that did nothing but edit video. Companies had to design and manufacturer that hardware, which was costly. Each year to keep up with advancing technologies, new machines had to be created, which meant a lot of work because each one had all the internal eletronics it needed to work on its own.
To edit audio, the same, first with tape, then with digital and embedded electronics. For video games, the same, from arcades to home systems. Even listening to music, was done on dedicated devices with embedded chips acting as the "brain" of electronics.
But now, since the Personal Computer revolution, there are generic brains everywhere. Instead of needing someone to use niche kits that require learning all the specialties of how to program for it, and designing an expensive dedicated unit, they can outsource the deep "brain" stuff to the PC, and focus on the app.
Now video, audio, games, and more are all moved to the PC. Instead of a very small team of dedicated engineers, we have a huge landscape of connected developers and users. This has several advantages, from lowering cost, to speeding up development, to giving users more choice and power to customize their tools.
There's disadvantage to this too. It's possible to get tighter efficiency in a dedicated system built specifically for one purpose that will never change. This means, if you build a drum machine and design it tight, you can get low-latency performance that is impossible today with a PC. It's because the PC is too generic, having to go through lots of extra steps, to be all things to all people.
Another disadvantage is portability. You can unplug a dedicated drum machine and take it with you anywhere, and it's all self-contained. A good example of when this is important is with cameras. We have USB cameras that can plug into a PC, where it's mostly just the light sensor, and the PC does the rest. Yet, most people don't want to carry a PC outside to snap a pic of a kitty. So, digital cameras are mostly still built as small dedicated devices, that have the processing, memory, storage, display, controls, and OS all built in.
But look at the situation we're in with the MPC firmware. Although there's advantages to it being self-contained, it's too niche, and Akai dropped the ball, so instead we rely on a rogue developer. People praise JJOS, and compared to AkaiOS it's great, but what if that were opened up to hundreds of rogue developers? It can't be done, because none of them are wanting to release control, and even if they did, it's too niche, so people would have to learn this specific system to code for it.
But look at the MPC Ren, which has no brain, yet has the controls of the MPC, and a USB interface. This taps into the entire PC environment. If they published the API, there'd instantly be hundreds of developers who could code for it. This is because, people already have PCs, and already know how to code for them, and there's huge libraries of existing code to start with. So the step to integrate software with the controller using a custom protocol is tiny compared to having to develop a dedicated standalone machine.
You know how you were finding clever ways to expand the MPC, doing things like voice-control by tying software that speaks MIDI to the MPC through USB? Well, you could do way more, with basically the same process, with something like the MPC Ren. Like how JJOS had Tetris before. You could find some Doom code, and a way to bridge its graphical output to the graphical input of the MPC, and the MPC button outputs to the Doom inputs, and then you're playing Doom on the MPC, something way beyond the capabilities of the dedicated MPC hardware.
Now of course that's ridiculous, you have no reason to play Doom on the MPC. But, it's just an example, of the power and possibility, which is very cheap when we share libraries of code and learn the same common PC environments. Maybe you love Ableton software, and just program a GUI for the MPC's LCD and method to interface with it using the MPC buttons. Then all you need to do is run Ableton, and you've got your MPC acting like a realtime performance machine. Then you can close that, and run the Akai software, and it's back to the familiar MPC paradigm.
That's just the theory. In practice, no one gets it yet, and it's taking a long time for the manufacturers to experiment with controller hardware, and the software developers to figure out how to interface with it, and the PC GUI is still stuck in the windows+mouse metaphor, with menus and all kinds of distractions that work against realtime creative workflow, and the latency is still higher than it should be, and most PCs still use loud fans and motors and take half a minute to cold boot, etc.
So, it's understandable why it's not obvious the benefit, and it seems kind of like reinventing the wheel to end up worse than it was. One of the things slowing it down is that the hardware makers don't seem to push the idea of publishing open source and specs. If these machines are using common protocols, then it's easier to integrate them with other software. For example, you can get a controller from Mackie or Behringer, and since they both speak the Mackie HUI protocol, which is published and coded into most major DAW apps, they both can work with your favorite software.
But will the MPC Ren have 3rd-party apps to choose? Or are you going to be locked into using Akai software? If they publish the spec, and encourage people to develop for it, making it free without some restrictive costly license, then you'll see it being integrated into major DAW software. You also may see neat rogue apps, like maybe JJ Sequencer with full LCD integration so it looks and feels just like JJOS on the MPC, but uses your PC RAM, HDD, and CPU, plus has Windows GUI for those times when you want to zoom in tight and do complex curve edits with mouse.
Without that, I think a lot of the benefit is more for Akai, and the choice to make a controller + software instead of a full machine is partially out of their own desire to make the most money with the least amount of effort. They probably even outsourced the PC software development, so instead of having to have a dedicated core team of engineers who understand the underlying architecture, they just wire some money to some firm and upload the spec, so they can quickly develop for cheap, and they can stick the update on their website.
Eventually someday, you could plug your MPC into your pocket phone/camera/player/browser/PC, and use it exactly as you do today, where it looks and feels identical and you can ignore the pocket PC completely. Then unplug and go to your friend's house, or the local club, and plug in there, to have their hardware instantly recall your MPC world, to play or perform without messing with their files.
But I don't see that happening any time soon, since the landscape is such a mess. Just trying to explain some of the theory, as to why we're moving in that direction. There are advantages to have a common electronic brain, with lots of appliance shells. There's disadvantages too, so some things, like a fire alarm, will always need to have some stable core of standalone functionality. But with audio and video, if you're designing a new music machine, it's easier and more appealing to design it around a PC today than some obscure electronic kit you ordered from some niche company only engineers know about.