The Gear Forum designs a next-gen digital modeler!

Completely irrelevant. I can connect any USB-MIDI device to both send to the Helix and receive from it.

Yeah, with Helix in host (slave) mode. In your own example, the USB-OTG adapter is turning your tablet into the main downstream device - and drawing power from it.

If i understood correctly the point made by others above, USB MIDI controllers like the Novation Launch Control will never work plugged directly to the Helix, because Helix cannot power, nor control, a USB link to begin with. And if you need external power, well, might as well use the standard MIDI port anyway.
 
Last edited:
Apparently.
Samsung Tab A (or Samsung S10) running an app called "MIDI Controller" (pretty cheesy, but working...) -> USB-OTG adapter cable from USB C plug to USB A jack -> standard USB A to B cable running into the Helix. Works.
The Tab and S10 are USB hosts. Of course hosts can connect to Helix. People have been sending MIDI to Helix from their iPhones (with camera kits) and iPads since 2015. What you cannot do is connect a USB device (like a MIDI keyboard, MIDI pad controller, or QWERTY keyboard) directly to Helix. For that, you need a USB A port on Helix and that port needs to support class-compliant MIDI (or a HID driver for the QWERTY keyboard).

Host <—> Device = Yay!
Device <—> Device = Boo!
Host <—> Product that can operate as either/both a Host and Device because it has both USB B and USB A ports (like, say, an Akai MPC) = Yay!
Device <—> Product that can operate as either/both a Host and Device because it has both USB B and USB A ports (like, say, an Akai MPC) = Yay!
Besides: You can send MIDI controls from your computer to the Helix via USB, too.
Of course. Computers are hosts.
 
Last edited:
because Helix cannot power, nor control, a USB link to begin with.

Yeah, this.
And fwiw, I'm perfectly aware of all of that, I was only talking about the MIDI side of things, which really don't need any "hosting" as there's a bi-directional communication going on via USB anyway.

People have been sending MIDI to Helix from their iPhones (with camera kits) and iPads since 2015.

Sure - and that's basically all that has been asked for, namely being able to connect external MIDI knob controllers. That's perfectly possible already, just that most newer knob controllers these days need USB hosting, which still is a pretty different thing as they basically only need that for power supplement.

Anyhow, as long as bidirectional communication isn't a given and as long as (ever since 3.5 at least "only" by default) the HX series is using the same internal communication data flow as snapshots, things won't be very satisfying IMO.
Personally, with the Stomp I don't need snapshots anymore and it seems that I'm only ever running one single patch per gig anyway, so I might give that MIDI Controller app another go. But for any bigger setups, external MIDI knob control isn't there yet.
 
Yeah, this.
And fwiw, I'm perfectly aware of all of that, I was only talking about the MIDI side of things, which really don't need any "hosting" as there's a bi-directional communication going on via USB anyway.



Sure - and that's basically all that has been asked for, namely being able to connect external MIDI knob controllers. That's perfectly possible already, just that most newer knob controllers these days need USB hosting, which still is a pretty different thing as they basically only need that for power supplement.

Anyhow, as long as bidirectional communication isn't a given and as long as (ever since 3.5 at least "only" by default) the HX series is using the same internal communication data flow as snapshots, things won't be very satisfying IMO.
Personally, with the Stomp I don't need snapshots anymore and it seems that I'm only ever running one single patch per gig anyway, so I might give that MIDI Controller app another go. But for any bigger setups, external MIDI knob control isn't there yet.
It has nothing to do with bi-directional data and everything to do with A) the physical port and B) whether the product acts as a USB device/peripheral, USB host, or—and this is still exceedingly uncommon—both. If a product can act as both, it almost always has separate USB B (to act as a device) and USB A (to act as a host) ports. There are a few examples of products whose USB C ports can switch between host and device/peripheral roles (mostly tablets and phones), but Helix has a USB B (device/peripheral) port. USB C is very flexible, but that also makes it a huge mess when connecting products that may not subscribe to very clear roles.

I'm not aware of any product that has only a USB B port but can act as a USB host; that is, can connect directly to bog-standard MIDI knob controllers, keyboards, etc. that also have only a USB B port.

As a general rule:
  • If a product has only a USB A port, it's a USB host. It most often does not have USB B ports. Examples: Macs, PCs. This does not include USB thumb drives, as their USB A connectors are not ports as much as they are sorta... integrated cables.
  • If a product has only a USB B port (or USB A cable attached), it's a USB device/peripheral designed to connect directly to a computer or perhaps a tablet or phone. It most often does not have USB A ports. Examples: MIDI keyboards, QWERTY keyboards, mice, printers, external drives, Helix, HX Stomp, BOSS GT1000, Moog Sub 37, Focusrite Scarlett, etc.
  • If a product has both a USB A and USB B port, its B port allows it to act like a device/peripheral, but its USB A port may still not necessarily allow it to act like a Mac/PC-level host. For example, it may only support USB drives, and not class-compliant MIDI for keyboard/knob controllers, HID for QWERTY keyboards/mice, drivers for printers, etc. That's up to the physical jack and how the firmware and if necessary, drivers are developed.
  • If a product has a USB C port, it most likely is intended to act as either a USB host (computers, tablets, phones) or a device/peripheral (almost everything else).
So what you want to ask for is, very specifically:
  1. An entirely new Helix hardware design.
  2. That new Helix hardware to have an extra USB A port on it (because otherwise you wouldn't be able to connect a Mac/PC at the same time)
  3. The specs of that USB A hardware to support class-compliant MIDI, not just storage for thumbdrives, etc.
  4. Line 6 to develop class-compliant MIDI hosting for the USB A hardware and test it with common MIDI keyboard controllers and knob boxes.
Perhaps doable if we ever get around to making a new flagship multieffect, but a lot of work just to support MIDI controllers without proper MIDI DIN ports.
 
Last edited:
See, I'm pretty much familiar with all that. And yet, that's not exactly what this discussion is about. It was about how to add MIDI controllers to the HX family. And while it's impossible that the HX devices will ever supply bus power to any MIDI controller (for obvious reasons), it's still possible to use them - if with some workarounds. And if I had a USB B to USB B cable (I doubt something like that exists, again for obvious reasons...) or any kind of adapter, I'd instantly check whether my BCR2000 would work via USB (I'd almost bet it would, might however need some fancy cable with an x-over).
Whatever, for the next HX generation, I think it'd be a clever move to simply add a USB C port providing both host and slave functionality.
 
I'd instantly check whether my BCR2000 would work via USB (I'd almost bet it would, might however need some fancy cable with an x-over).
I'd take that bet, because I reckon there's no hardware to support USB host mode in the Helix, and certainly no software! ;)
 
See, I'm pretty much familiar with all that. And yet, that's not exactly what this discussion is about. It was about how to add MIDI controllers to the HX family. And while it's impossible that the HX devices will ever supply bus power to any MIDI controller (for obvious reasons), it's still possible to use them - if with some workarounds.
You simply cannot connect two USB devices/peripherals together with a USB B <> B cable. Not only because USB B <> B cables don't exist, but because two devices/peripherals have no way to talk to one another:
  • A USB game controller can't talk directly with a printer
  • A USB hard drive can't talk directly with a Strymon Iridium
  • A QWERTY keyboard can't talk directly with an HX Stomp
  • A MIDI keyboard can't talk directly with an audio interface
And if I had a USB B to USB B cable (I doubt something like that exists, again for obvious reasons...) or any kind of adapter, I'd instantly check whether my BCR2000 would work via USB (I'd almost bet it would, might however need some fancy cable with an x-over).
Emphasis mine. It doesn't work like that. None of this works like that.

In the end, a device is a device and a host is a host, and never the twain shall meet. The notion that USB just magically sends and receives MIDI like bi-directional MIDI DIN cables... has never been a thing.
Whatever, for the next HX generation, I think it'd be a clever move to simply add a USB C port providing both host and slave functionality.
The problem is that again, one couldn't connect both a Mac/PC and MIDI peripheral at the same time; plus, USB C data/power management is a much bigger mess when a port can switch between device and host duties. You also don't want two USB C jacks, as it makes it way too easy to connect to the wrong port. If you want to communicate with something that has a USB B port (like a MIDI controller) but no MIDI DIN ports, you really want a USB A port on your product. And that USB A hardware needs to support class-compliant MIDI and your development team needs to take the time to program your product to be a class-compliant MIDI host, which is far from simple and three galaxies away from "I bet it just works; you need the right cable."

IMHO, that's a lot of cost and effort to simply avoid using MIDI DIN cables or supporting MIDI controllers that only have USB.
 
Last edited:
You simply cannot connect two USB devices/peripherals together with a USB B <> B cable. Not only because USB B <> B cables don't exist, but because two devices/peripherals have no way to talk to one another:
  • A USB game controller can't talk directly with a printer
  • A USB hard drive can't talk directly with a Strymon Iridium
  • A QWERTY keyboard can't talk directly with an HX Stomp
  • A MIDI keyboard can't talk directly with an audio interface

Emphasis mine. It doesn't work like that. None of this works like that.

In the end, a device is a device and a host is a host, and never the twain shall meet. The notion that USB just magically sends and receives MIDI like bi-directional MIDI DIN cables... has never been a thing.

The problem is that again, one couldn't connect both a Mac/PC and MIDI peripheral at the same time; plus, USB C data/power management is a much bigger mess when a port can switch between device and host duties. You also don't want two USB C jacks, as it makes it way too easy to connect to the wrong port. If you want to communicate with something that has a USB B port (like a MIDI controller) but no MIDI DIN ports, you really want a USB A port on your product. And that USB A hardware needs to support class-compliant MIDI and your development team needs to take the time to program your product to be a class-compliant MIDI host, which is far from simple and three galaxies away from "I bet it just works; you need the right cable."

IMHO, that's a lot of cost and effort to simply avoid using MIDI DIN cables or supporting MIDI controllers that only have USB.
Thanks for the clear and concise explanation. Your rationale makes good sense. However as we begin to see devices like Korg's Kronos and Nautilus sport USB Hosts I wonder if in the longer term they won’t just become table stakes for most digital musical instruments. Admittedly synths tend to live in a far more networked environment than modellers so the same impetus may not apply. But as an exercise in blue skying, having the ability to easily (simplified 2-way learning maybe) extend a modeller’s UI into the realm of faders and knobs would be a really attractive proposition.
 
Thanks for the clear and concise explanation. Your rationale makes good sense. However as we begin to see devices like Korg's Kronos and Nautilus sport USB Hosts I wonder if in the longer term they won’t just become table stakes for most digital musical instruments. Admittedly synths tend to live in a far more networked environment than modellers so the same impetus may not apply. But as an exercise in blue skying, having the ability to easily (simplified 2-way learning maybe) extend a modeller’s UI into the realm of faders and knobs would be a really attractive proposition.
It makes a lot of sense in the world of Kronoses, Nautiluses, Akai MPCs, etc. because MIDI keyboard/knob controllers are the expected method of input. That's not to say there's zero utility in connecting keyboards, knobs, etc. to floor-based multieffects, but in our world, it's far more common for MIDI-equipped guitarists to just use DIN ports. In fact, I'm struggling to think of a guitar-centric, non-product-specific MIDI controller that doesn't have 5-pin MIDI DIN ports (or 1/8" TRS MIDI ports, which is the same thing); I'm sure someone will correct me.

EDIT: KMI Softsteps don't have 5-PIN or 1/8" MIDI but they sell a mini USB-to-5-PIN DIN adapter? The site doesn't show a proper photo of it, so I have no idea if it's an interface itself or they're using mini USB as their multipin connector.
Because a product shouldn't act as a USB device and USB host through the same port. There are some rare cases where power management can be configured to work between, say, a laptop and a tablet where the tablet acts as the data host but the phone acts as the power host (generally when the phone is connected to an outlet but the tablet isn't), but that's far beyond the scope of this conversation and doesn't apply to MIDI.

Besides, how would you connect two devices to the same port anyway? If you say "USB hub!", note that USB Hubs are used to connect multiple peripherals to the same single host, not connect multiple hosts. iConnectivity MIDI does make interfaces that can intelligently shuttle MIDI (and audio!) from, say, a tablet or phone to a computer and back, but that's accomplished via two USB B ports (because tablets, phones, and computers are USB hosts; any peripherals [like MIDI controllers] are connected via their USB A ports).
 
Last edited:
But computers and smartphones do that all the time.

And as far as connecting multiple devices go, yeah well, why not add another USB C port?
I already explained this:
You also don't want two USB C jacks, as it makes it way too easy to connect to the wrong port.
Also, it's much easier to find bog-standard USB A <> B cables. "I need a USB printer cable" is generally all you need to say at your corner drugstore.
 
Come On Reaction GIF by GIPHY News
 
But computers and smartphones do that all the time.

Um, no, they definitely don't :D You might be able to transfer data back and forth between them, but the USB spec is strictly tree-network.
There's always a host, devices cannot interact with one another except via the host, and hosts cannot communicate directly.

Note this is true even for USB OTG - it just allows devices to change roles on the fly.
 
Note this is true even for USB OTG - it just allows devices to change roles on the fly.

Well, then be it that way. Fine with me. Really, everything that could be made easier should be made easier. Anything else but USB C connections should vanish from earth.
 
Well, if both USB C ports served the same purpose (which they could), there'd be no wrong one to plug in.
USB hardware and its required tangential components vary greatly in capability, complexity, and cost.

Line 6 could spend 5x the cost to put two robust role-switching USB C jacks in a product (and by extension, add 5x more to that functionality's cost to the user), spend 5x the development and testing cycles ensuring it works perfectly, and then deal with the fallout of CS calls from people trying to connect two computers or a computer and tablet/phone and one not working.

Or...

We could save money, charge less to the customer, and use a USB B (or normal USB C) port for host communication... right next to a USB A port for peripheral communication, like Korg and Akai do.

Or...

We could save even more money, use those development/test cycles for features that more than three users care about, and ditch the second port completely because the vast majority of guitarists and bassists who use MIDI are perfectly fine with 5-pin DIN jacks... and nearly all guitar-centric controllers have 5-pin MIDI jacks. It's not like the masses will let us get rid of MIDI DIN jacks anytime soon anyway.

Or...

We could do something completely different.
 
I really hope the next gen uses USB-C for connection to the computer.

For MIDI, I don't care much as I don't use it. If it's a full size device, just use the standard 5 pin. If it needs to be more compact, a breakout cable from mini jacks is fine.
 
We could save money, charge less to the customer, and use a USB B (or normal USB C) port for host communication...

Ok. At least use decent ones next time then (plenty of reports of the "noses" falling off from the USB ports).

and nearly all guitar-centric controllers have 5-pin MIDI jacks.

Not knob controllers, though. I have been looking around a lot. Went back to my clumsy, way too huge BCR, only to ditch that as well, simply because the Helix Floor wasn't up to it - so that's another thing possibly. We could have a deal: I'm fine with 5-pin MIDI the day L6 implements things properly. Or so.
 
Back
Top