#145

That’s pretty nice, though I think I would prefer remember what icons do what, rather than what number is what icon.

#146

In that case, the latency isn’t noticeable. I thought you were asking about the Lightpad for rhythmic performance. The physical button Blocks do not have noticeable latency.

As a heads up, though, the button Blocks require a Lightpad Block or Seaboard Block for Max use. The button Blocks do not have a USB port on them, which is needed for Max interaction.

1 Like
#147

That’s a good news bad news situation. I was hoping it could just be used as a programmable hand held bluetooth remote.

#148

Buy a single Nintendo Switch Joycon?

I think they are programmable via third party software

they are definitely bluetooth and handheld.

1 Like
#149

There should be many used and inexpensive Wiimotes out in the world now, as well.

Here, those are used to good effect by Jazari:

#150

Hi gang, time to bump this convo and revisit everything that’s happened since it went dark! :slight_smile:

A lot’s happened since June. To summarize (and answer Jason’s query along the way):

The new Lightpad M is lightyears ahead of the original Lightpad in terms of comfortable touch sensitivity. The original was so stiff to operate that it actually hurt my fingers to use for long periods, but the new one is softer, more responsive, more brightly lit, and more even in its response at edges and corners… a night and day difference. As a universal utility MPE-capable surface, it’s something I’m now happy to carry around with me. (And battery life’s better too, at least in mine.)

I haven’t gotten into the Morph yet, but I hope to soon. I’ll be writing about it for the magazine in a couple of months.

For folks who haven’t been keeping track, the Joué is finally ready to ship. They hit their completion deadline, only to discover that their factory had used bad paint on the overlays and they were cracking and peeling almost immediately. It took them a couple of months to get that sorted, but now they’re ready to start filling orders; I am looking forward to getting mine.

In other news, KMI still isn’t shipping the K-Board Pro 4. I am not at liberty to discuss what I was told at NAMM, but in very general terms, one of the two versions is basically ready to ship but the other one is still being tweaked. They’re very aware of the fact that they’re late, but plan to start filling orders in the second quarter. I have a lot of faith that this will happen and that the resulting units will be very good; the BopPad was late, but is absolutely fantastic, and the production samples of the K-Board Pro 4 that I played at NAMM were phenomenal. They’ve definitely got it working like a dream, they’re just ironing out construction in quantity.

I recently got a Linnstrument 128 as a thank-you gift from Kim Bjørn for my editing work on PUSH TURN MOVE, which has been a smashing success. I’m hoping to actually be able to do something with it soon, I’ve been too busy writing reviews…

Watch this space for future developments!

3 Likes
#151

This is good to hear. How is it in terms of latency? I am revisiting the idea of picking up a seaboard block, but very much on the fence.

Curious what you come up with on this one as well. I probably need to revisit Sensel in case they’ve released some updates, but last I checked, they need to up their game pretty hard on the software front. You either get full control, with nothing done for you (developer overlay or no overlay at all), or you get an under-developed, not-configurable-enough-but-not-quite-what-you-want-either MPD or keyboard. They would gain quite a lot in terms of community and functionality if they were to open source the editors for the official overlays, rather than only providing their very broad developer SDK.

1 Like
#152

I haven’t noticed severe latency problems when doing Bluetooth, and of course you can always use them with a USB-C cable…

We shall see. I need to dive in and poke around, and I haven’t really properly started yet…

#153

Just a few quick additions to this thread…

I am now in regular discussion with the folks over at Sensel about the Morph. They are keeping development moving forward as quickly as they can, and recently gave me a beta of their latest software, which I am still in the process of getting used to. I would agree with the assertion that the software side of things has a long way to go. I think they are very interested in making sure that hackers can do whatever they want, but they are taking their time [too much?] in adding more beginner-friendly features. I sent a fairly long list of fixes and upgrades I would like to see, and the beta should address at least some of them.

I should note, however, that I am very impressed with the Morph hardware. It’s light, well-designed, works well either with cable or Bluetooth, and really is a great platform for whatever the software people come up with. (I will admit to being a little bit frustrated by the music production template, because it seems to be backwards to the way I would want to use it, but that’s just me.)

On another note, the Joué arrived sometime ago, and after an initial flurry of hard use, I have temporarily set it aside. Once again, the hardware is quite impressive. The frame for the modules is incredibly solid and reassuringly well-built, a little heavier than I would like but that’s actually probably a good thing. The various overlays are comfortable to use and inspire a lot of creativity. However, the support software lacks a fair number of features that I really find essential for the way I design control surfaces, and I made up a large laundry list of them which I sent over to the builders. They responded that the vast majority of these will be addressed directly in the next upgrade to the software. In the meantime, it’s not getting much use, simply because the places where it doesn’t do what I want it to are too frustrating for me to enjoy the many things that it does well.

I am putting my Linnstrument 128 up for sale. It’s a genius piece and very well-built, but it completely does not connect with how my brain works. Very expensive lesson, but there’s really no other way you can learn this stuff without putting your hands on it.

And to return to where we started: my work with the ROLI Blocks has ground to a halt. I can visualize exactly what I want them to do, and there are some preset templates that come close in terms of UI. However, ROLI’s idea of the interface being hackable is to provide a low-footprint (and low-documentation!) code base for it, few if any examples, and basically no help.

As someone who does not code and has not coded since before the C++ language was invented, I find this astoundingly frustrating. This frustration is compounded by the software folks at ROLI telling me, in effect, that they consider this ecosystem completely bimodal… either you are a complete musical beginner, in which case they feel the obligation to provide a fully realized software system that you don’t customize at all and simply play with, or you’re a hacker, in which case they feel the obligation to provide the language to do the hacking, and their responsibility ends there. If you want to hack, go hack and don’t bother us. door closes firmly

This makes me sad, as there’s a fair bit of potential there, but they don’t seem interested in taking responsibility to make resources available for people who want to learn to do this stuff. Either that, or I’m just whining because I want someone else to do the hard part for me. :slight_smile:

I think I’ve said enough for one post. If anyone is specifically interested in any of these pieces, I don’t mind starting a new thread about any or all of them, as long as I still have the hardware in question.

6 Likes
#154

Same.

They were responsive to my first impressions, but never replied to my in-depth list of aftertouch fixes needed. Those were pretty obvious, though, so I expect they’ll still catch them.

The bigger problem, I think, is that there’s only one USB MIDI endpoint. You have 16 MIDI channels to distribute between each of your overlays, and if several use MPE, your apps won’t be able to differentiate between those. I suggested that, if creating more virtual cables wasn’t an option, OSC would resolve it for us. They suggested buying a second Joué frame.

So… once the software is fixed, I can pretty much find one or two configurations I like, and throw away the rest of my overlays.

That’s frustrating. But the build really is something.

#155

I recently purchased the Roli Seaboard Block, though I haven’t been that impressed with it so far, for both software and hardware reasons.

Part of the issue, which I should have foreseen, is that I use Ableton Live as my primary DAW, and it is perhaps the most poor DAW with respect to MPE support. For example, a multitimbral synth can usually be used with an MPE controller, by placing each voice on a different channel. This isn’t the full MPE experience, but it is at least workable. However VSTs with multitimbral support that place each voice on a separate channel, like the Korg M1 VST, can’t easily be used in Live because Live strips out the information about MIDI channels before sending it to the VST. The workaround is to create a separate instance of a synth for each voice, however any edits to a voice then need to be replicated across up to 15 copies of the same instrument, all channels need to be armed for recording, etc. The Seaboard Block works OK with the Korg M1 in stand-alone mode (outside a DAW).

On the hardware itself, I noticed in the store when comparing the Seaboard Rise to the Seaboard Block, that the Block, besides being much smaller, also has shallower ‘foam’ on each key. In practice I find my hands have tended to feel sore after using the Block for a bit, perhaps due to constantly applying pressure to the foam surface with my fingertips for extended periods of time. I also find the rubbery surface isn’t that pleasant on your finger joints to drag your fingertips across for pitch bends, since you also need to apply a bit of pressure to maintain the note.

I also think it would have been useful to also include a non-USB midi out port, either full-size 5-pin DIN or one of the newer 3.5 mm formats. The device has a rechargeable battery but it’s a bit useless to me since I don’t want to bother with MIDI over bluetooth, so it’s always plugged in via USB anyway.

1 Like
#156

Could you describe what sort of setup you’re trying to build that goes beyond what a single USB-MIDI endpoint can handle? I certainly understand the limitation, especially when used with MPE, but a detailed description of what you’re building would help me see exactly what you’re bumping up against.

(as background; I’ve designed hardware, firmware, and software for several USB-MIDI controllers and sat on the MPE Working Group. My goal here is to understand how to design products in the future that are less limited, with the caveat that throwing more features into a design is not always the best choice - I’m looking for clever design options that don’t necessarily add to the cost or create new limitations)

p.s. I just remembered that Windows has a lot of quirks because it historically has not implemented support for the full USB-MIDI specification. In some cases, adding USB-MIDI features will be welcomed by macOS users, but Windows users will be unable to use the device at all.

1 Like
#157

I suspect @greaterthanzero is referring to the fact that the Joue frame, could potentially have 2 or 3 different MPE ‘devices’ (Joue is split into 3 separate pads)

However, this does not need different USB endpoints (nor would they add value), this could be handled by MPE zones.(which are designed for this - basically MPE splits)

though in fairness, I think the MPE zone spec is not that great, and Ive seen little/no support for it in hardware or software - but lets hope now the spec is finally published developers can rally behind it.

1 Like
#158

Some synths let you specify which channel is the global channel and what channel range to look for events on. This makes MPE splits possible, assuming your controller is similarly configurable (linnstrument can do this).

#159

I know, I implemented mpe (and this) for the Eigenharps, but the mpe specs allows this information to be automatically done via an MCM, do you know of any hardware, or software synth/daw that handles this?

but as I said, hopefully now its published, we can all just ‘run with it’

btw: interestingly, the midi association published mpe reverted to the simplified mpe zone spec, which im thankful for (as Roli had updated the draft spec to a more convoluted multi zone (>2))

#160

I must say live midi implementation lacking proper multi channel support is beyond criminal at this point… I am basically forced to use logic for stuff live is better suited for.

1 Like
#161

OK so I totally agree about Live’s MIDI support. I do. I even bought Bitwig because I got so mad and I tried to go back to Logic, which I used from Notator days until 2002, when I switched to Live. It’s just no good for me–I need Ableton at this point, there are just too many things that I do with it that I can’t reproduce elsewhere.

BUT. Ableton + MPE is doable today. It is not ideal, but possible. And Live’s MIDI limitations themselves are work-aroundable as well. Not a defense, but just an indication that things are possible.

First thing first: If you are trying to use MPE with Ableton get and read all about this: https://github.com/juhot/MPE_Util

So while there is no MPE compatibility in Ableton right now, most plugins that support MPE also support a config where each voice gets it’s own MIDI channel, which can produce most of the results that real MPE support. ROLI stuff, for instance, supports this with a mode that round-robins notes through a set number of MIDI channels, so each voice gets its own midi channel. This works much better than a separate instrument on every channel for many reasons.

So far I use this to great effect with Equator (the ROLI synth), with Kontakt running the Auras library and a few setups I have made myself and with Repro-5. I can also get this to work kinda ok with Omnisphere. There are other synths that support this setup as well.

What sucks about it? Can’t really record MIDI to any effect. It just works better to record audio, at least for me. It is still more of a pain than it should be.

What is great about it? It works inside Ableton.

What do I think needs to happen? Ableton needs to support MPE. A little birdy told me that this necessitates a pretty deep rewrite of the central guts of Ableton and that’s why it is taking a while.

Anyway. MPE util. Basically you put an MPE plugin (or any plugin ready to work with multiple channels/CCs on a track. Then rename the track to -creatempe6 and it will automatically build out a setup like this through magic. Then you can group all the tracks that it makes and it just works.

As for the way MIDI works in Ableton. I don’t love it and I think it is dumb, but. In my project template I have a group that when expanded shows basically a MIDI mixer. All my inputs have dedicated tracks. I then take my MIDI inputs from those channels. This allows me to have 16 discrete MIDI channels per device. I have some Max4live stuff that I used to filter and route when I need more than that. So far I have no come up short and I’ll tell ya, when I go to logic and try ti figure out how to do some things that feel easy in Ableton my head often starts spinning. Ableton is limited but the modular nature of the mixer is very powerful and not very easily reproducible on these other platforms.

So, that should get you MPE-like behavior in Ableton without so so much pain.

Also, my plug. The Aura’s MPE Kontakt library is simply amazing. They implemented it as multi patches so it works pretty easy out of the box and the sounds are just phenomenal. They did a lot of instrument design for the Annihilation OST and many of those sounds are presented in this library. Can’t recommend it highly enough of you have a MPE controller. https://slateandash.com/collections/frontpage/products/auras

4 Likes
#162

Thanks for this, I was only aware of the create multiple vst instance approach…

#163

It depends on the plugin. The plugin needs to be able to work in this multichannel in mode.

In Kontakt and Omnisphere you make a multi. It can be a pita but at least it can be saved as a single preset.

u-he Repro-5 seems to take mod wheel and pitchbend per-channel into a single instance. People on the internets say Diva works the same way.

Audiodamage Quanta works well.

There may be others.

#164

This controller has capabilities beyond its capacity. That’s frustrating. More details below.

(I’d be happy to talk at length about more generalized goals and frustrations, but this might not be the right vehicle for that discussion.)

Short version is, they’ve released twelve separate modules and they’re planning to make more. And the only way to know which of those is producing a given signal is by assigning them unique MIDI channels.

MPE zones is a partial solution, but we can file that next to a dozen other improvements and workarounds they might implement someday. Supporting poly aftertouch would also help a lot, since right now, MPE is the only way to see pressure data for individual pads. This creates unnecessary incentive to consume extra channels, and heightens the problem of only having 16.

Really, I have a litany of fairly major complaints with Joué, each under the header of “they can fix that in software”. Like, if I put aside the controller for a year or two, it should be fine when I get back.

I’m an early adopter. I can forgive that stuff.

But this inability to differentiate which module a signal is coming from is more of a structural problem. Without suddenly adding OSC support or giving us access to the raw data, we’re limited to channel based module IDs.

And actually, yeah. Structuring a set of “devices” based on the three regions you can put a module into would actually solve everything for me. (That’s me, personally, as a developer, not representing the average consumer.)

Joué does send sysex messages identifying when a specific module has been added or removed from each region, and I could totally leverage those. So, if I knew which region a note originated from, I’d know which module created it, and could route things appropriately on my end.

2 Likes