[Editor's note: in this article, we're taking a closer look at MIDI 2.0—a significant update to the MIDI protocol over 15 years in the making. It's likely that you're heard of MIDI 2.0; it's been a hot topic in the electronic music and production world for years. But…what is it really all about?

In order to highlight some of the shortcomings in the original MIDI specification that led to MIDI 2.0's inception, we recently published the colorfully-titled article What's Wrong with MIDI? If this is the first time you're hearing about MIDI 2.0 altogether, we'd suggest starting there for some perspective. In this article, we're diving deeper into the specifics about MIDI 2.0—including a technical overview, as well as perspectives from the people behind its development. Let's get started.]

The Current State of MIDI 2.0

For around the last 40 years, the MIDI protocol has served as a mainstay, allowing electronic musical instruments and computers to freely exchange information. During that time, MIDI has been so successful that it has hardly been updated or added to—remaining largely unchanged since its introduction in 1983.

Recently, in a press release from the MIDI Association, it was announced that Focusrite Group CEO Tim Carroll has been named president of the MIDI Association. During his two year term, it is hoped that even more developments and expansions will be seen with the further implementation of MIDI 2.0. For those unfamiliar, The Focusrite Group of brands includes, among others, Sequential, Oberheim, and Novation—and of course, Focusrite themselves. In a recent press release, Carroll noted that he is “...proud to accept the president’s role and lead the MIDI Association for the coming two years. This is a critical inflection point to the MIDI Association’s history. As an industry we’ve only begun to realize the potential of MIDI 2.0, and I’m excited to work with the MIDI Association board and all its members to create MIDI’s next expressive, musical, and creative chapters.”

To get a better understanding of what a massive undertaking this has been, I spoke with Rob Rampley of Media Overkill, a key developer of the new MIDI 2.0 protocol. He shared that, “I wrote the first draft of MIDI 2.0 in August of 2005, we called it ‘HD-MIDI’ back then...and then it got voted into reality in 2020. That is not a negative statement—the obstacles were there for a very good reason—they just weren’t visible until we actually began the work.”

Mike Kent of MK2 Audio was also a key contributor to the development of MIDI 2.0. Kent noted they were “Creating something modern while keeping backward compatibility to a 37 year old legacy.” According to him, some of the challenges were as follows:

“MIDI 1.0 is really good. We worked hard to keep existing MIDI concepts so that the existing understanding and knowledge retains its value. Many MIDI users have a personal and long-term connection to their musical instruments. Existing users do not want a new MIDI that does not connect to the products they already own and love. We had to find a balance between honoring the past and adding new capabilities.

The design we have come up with stays close to MIDI 1.0, but I think has enough flexibility to add more, step by step, in the years to come—providing powerful features, but recognizing that MIDI devices don’t all have high powered processors.”

He also elaborated that, “There is a wide range of MIDI products, both hardware and software. Hardware ranges from $50 gadgets to $5000 powerhouse synthesizers. Software ranges from simple apps on a phone to DAWs running on mulit-core processors. MIDI 1.0 delivers great musical power without requiring heavy processing power to use it. It has been a challenge to find new capabilities that allow simple devices to do new things while also delivering powerful new features for high-end instruments. The members of the MIDI Manufacturers Association have a broad spectrum of product types with different needs; agreement wasn't always easy. But we found compromises that best serve the whole MIDI world. MIDI 2.0 has features to make MIDI easier for all users, while also allowing demanding artists to expand their creative expression.”

I also spoke with Brett Porter (of Art+Logic) ,who also worked on the MIDI 2.0 efforts. He said, “I was lucky—Art+Logic was brought into this effort at the point where the new specification was nearing completion. Our involvement was focused on developing a prototype implementation, so the challenges were mostly stubbing my toe on parts of the standard that were not described clearly or where different parts of the document contradicted each other, and working to clarify those issues. At that point, the Working Group putting the standard together was still in the process of merging two different proposals for 'next generation MIDI,' and at a larger scale, the challenge was getting two sets of very smart and very opinionated designers and users of MIDI to work through the compromises required for that unification."

He also added that going forward, “Longer-term, the real challenge will be getting all of the required pieces in place for the entire ecosystem to work—new drivers for interface hardware, operating system updates, software updates.”

MPE (MIDI Polyphonic Expression) has been one of the only significant additions to the standard MIDI protocol in recent years, allowing MIDI to work more seamlessly with polyphonic gestures. The consistency of MIDI has allowed a lot of older hardware to stay in use, communicate with DAWs, and allow for developments and applications beyond music. So…why update it now? What can we anticipate from changing MIDI?

So, Why MIDI 2.0?

Let’s start by revisiting some of MIDI’s shortcomings. Until now, MIDI has communicated primarily in 7-bit messages, which only allow a range of integers from 0–127. For many control-oriented purposes, this communicates enough detail—but ultimately, not a lot. This means that the resolution of information, such as communicated by a hardware controller, is relatively un-detailed. This can significantly inhibit expressive performance…but somehow, despite that, we have more controllers on the market than ever before, offering a wide range of tactile performance gestures.

×

Still, these low-res MIDI messages lack a lot of nuance in things like gesture information, and are relatively slow for modern data transfers. Although not all MIDI data still needs the MIDI specific hardware connection (as much can be done over USB today), there’s still room to improve on this. The updates in MIDI 2.0 are really a jumping off point, as there are so many unknowns of what could be created next.

Digging into what MIDI 2.0 will offer, we can look at the key features of the MIDI 2.0 environment. These are outlined in two groups by the MIDI Association and the MIDI Manufacturer’s Association. Chiefly, we have The Three Bs and The Three Ps—which helps make all this information a little bit easier to grasp.

The Three Bs

The Three Bs describe the implications of a new MIDI expansion known as MIDI Capability Inquiry (sometimes referred to as MIDI CI or just CI). MIDI 2.0 uses MIDI CI to establish Bidirectionality (our first B), which assumes bidirectional communication. So, MIDI changes from the previous protocol—it was previously a monologue, but it can now be more of a dialogue. In application, this means that connected devices can talk to one another and find out what specific protocol they are going to operate off of before proceeding.

The second B is Backwards Compatible. This core design principle ensures that old devices and new devices to always allow fallback to MIDI 1.0 to maintain compatibility. So in essence, yes—your old gear will still work, and new technology will be able to quickly establish how it needs to talk to older hardware.

The last B of the three is Both, which speaks to a key goal of MIDI 2.0—the goal to enhance MIDI 1.0 features where possible.

So, again, our three Bs are: Bidirectionality, Backwards Compatibility, and Both.

The Three Ps

The three Ps pertain to future and possible MIDI expansions, which are enabled by MIDI CI enhancements. The first is Profile Configuration, which defines sets of rules for how a MIDI device may send or respond to specific MIDI messages to achieve a specific application. In other words, this helps declare compatibility.

Property Exchange is the second P. Property Exchange involves messages that both receive and set device profiles, including but not limited to things like product names, configuration settings, controller values, and various other forms of meta data. This type of information exchange is very helpful for things like exchanging parameter information or batches of presets.

The last of the three P’s is Protocol Negotiation, which is where MIDI-CI will define how it will negotiate between old and new protocols. In instances where devices do not support MIDI 2.0, MIDI 1.0 will continue to be used.

So, What Does This All Mean For You?

So, what does this all mean in practice? We see MIDI 2.0 as offering a framework for more detailed communication between hardware and software—and the potential for expansion upon the present state of gestures and information exchanged. As MIDI 2.0 is fully adopted, more innovations will continue to emerge.

MIDI 1.0 should work as normal as long as DAW makers continue to support 1.0. From a software perspective, DAWs may gain the ability to harness new ways to edit and notate MIDI—as well as new methods for capturing, producing, and reproducing gestural information within a software instrument or DAW.

Brett Porter noted that, “After a few years exploring the expressive capabilities added to MIDI via MPE instruments and synths, I'm most excited personally to add even more per-note control to MIDI at a higher resolution and more polyphony. Even without thinking about per-note control, the increased resolution in the new MIDI 2.0 voice messages will address a lot of the issues that I've been frustrated by for decades.”

This also paves the way for new ways of communicating and executing things like MIDI edits as software develops.

Mike Kent also noted that, “the auto-configuration features we are developing will make the biggest difference to most MIDI users. Imagine things like having hardware recall in a DAW project [as it] happens now with plugins, auto-mapping of controllers for every connection, or a device self-generating its own editing software in a DAW.” These are astounding possibilities.

Of the new capabilities of MIDI 2.0, Kent has said he’s “interested in the new pitch control capabilities. I know there are many musicians who work outside of the 12-tone Western scale, and I am happy that MIDI 2.0 will meet those interests. But I make Western music. I’m interested in how pitch can be controlled over the life of a note to add acoustic instrument type movement. I’m also interested in how MIDI 2.0 can be used to escape equal temperament, to have Just Intonation in chords and intervals, or to arrange music that has chords with different colors based on flexible scale tunings.”

×

When asked what they see happening next, Brett Porter said that, “we need to wait for all of the various manufacturers to release their individual pieces of the puzzle, which puts us into a small chicken-and-egg situation—a MIDI 2.0 synth isn't fully usable until there's an official way to connect to it (we're waiting for the USB-Implementers Forum to approve the proposed transport details for MIDI 2.0 Universal MIDI Packets), and as a software developer, there's only so much I can do until Microsoft/Apple add support for MIDI 2.0 to their operating systems, and until device drivers for all my IO boxes are updated to get that data into my computer.”

Mike Kent weighed in to describe some of the potential ways that the MIDI 2.0 rollout will unfold, adding, “It will take time to see the potential of MIDI 2.0 realized. We need all parts of a system to support MIDI 2.0. […] Many users won’t really be able to use MIDI 2.0 until the DAW they use is updated to add MIDI 2.0. Many current MIDI products could be updated to add some component of MIDI 2.0 support. Some companies will ship brand new products designed from the start with MIDI 2.0 capabilities. It will probably take up to a few years for MIDI 2.0 to spread to a wide set of MIDI devices.”

While some may worry about compatibility, or being forced to make pricey upgrades, each of those things are dependent on manufacturers. Brett Porter noted that, “An astonishing amount of work went into making MIDI 2.0 interoperate with MIDI 1.0—all MIDI 2.0 equipment will power-up as MIDI 1 devices, and only make use of the new features when both pieces of gear have negotiated what the other one is capable of. People make fun of me sometimes because I still play my original Yamaha DX-7 that I bought new in '86, but I have a weird emotional attachment to it, and I plan to keep it in my rig until it gives up.” He added, “the Capability Inquiry facility in MIDI 2 should mean that users won't even need to be too aware of what each piece of gear they're using can do because the system will figure those things out by itself.”

This is truly an exciting time for music makers, MIDI controller designers, and software developers. In closing, Brett Porter mentioned, “the opportunity to create radically new kinds of musical systems and interfaces is going to result in an explosion of invention, I hope. I'm really looking forward to creating some of these new things, and getting a chance to play with the things that other folks are probably starting to work on right now.”

MIDI 2.0 was and is a massive effort on the part of so many. In Porter's words, “This wasn't the first time I've been involved with a group creating a technical standard, but it was definitely the first time I've done so where everyone involved was motivated as much for personal creative reasons as I've seen with MIDI 2.0.”

Of course, MIDI 2.0 contains many new features we didn't discuss here—we’ve only just begun to scratch the surface, and can't wait to see where this will lead next. So, for now, hold on to your controllers everyone!