Forum Archive

Let me start by expressing my great appreciation for the amazing Eigenstaff who are so active on these boards and provide an unprecedented level of support for this unprecedented instrument.

And also for the instrument itself, which is the most sensitive instrument I've ever played and which I hope to one day master.

However, I'm playing less in the last week and a half, and that's because I realize that I won't be able to take this instrument out into the Real World for at least three months if not six... unless of course I step in and whine a bit. :-D

There are many classes of possible players for this instrument but certainly one is the professional or (like me) semi-pro musician who already has an existing musical world which this instrument is just one part of.

Unfortunately, the Eigensoftware does not yet play well with others. I was hoping to get a gander at the software and see what I could do to tweak that but that won't be till 2010 so thought I could give you a hint of what I'm thinking of.

Let's start with "scene changes" - when I need to change the sounds and tuning of the instrument.

In a lot of music I'm called upon to play, I simply don't have time to press a sequence of buttons to change scenes. On the WX-7 I play as my main instrument, I have five program changes literally under my thumb. A keyboard player could easily have a dozen under a fingertip.

The Eigenharp, living as it is entirely in software, could do a much better job. Oh, boy. But this is the subject of a later article... :-)

But I always seem to have some controller program like a sequencer running anyway, and so do it seems 75% of DJs these days.

For now, I'd simply setting for a totally stupid, generic incoming MIDI interface where I could just directly set the parameters with continuous controllers. If you use the MIDI fine controllers concept, you could get 14-bit resolution, but I'd like with the bad 7-bit control you could get from just one controller.

You need to do this anyway!

That alone would go a long way toward having the Eigenharp appear at my gigs - or anyone else who already used a sequencer.

There's the other known problem of "producing sound in other programs". The Jack solution is simply not good enough for prime time, particularly since the target machine has Snow Leper on it. The Jack mailing list is full of issues, some of which I have.

I know and sympathize that this a difficult problem, but there needs to be a way for the Eigenharp to just "pop up" in every major sequencer.

If I recall correctly, the AU plug-in you have is the way to go, right? But many manufacturers don't support the advanced feature that you are using?

So, since I'm telling you how to do your job :-D why not do the following... go to *one* big sequencer company like perhaps Ableton and send them a Pico - if they promise to fix this feature as soon as possible? They might melt if they had one in their hands...

(Ableton's in a big push right now and might be a good candidate as they're supporting alternate controllers. If you got it all together they'd certainly feature the Eigenharp in their promotional material!)

Thanks for reading and again my appreciation for the developers and the community.

I´m beginning to use my pico in preparation for our 2010 gigs and I still have some issues with stability.

I´m positive that the eigenguys will be solve those issues pretty soon. Plus the upcoming scripting will provide us with a lot of new features that may be interesting.

There are certain things that are a real pain for a live situation and to my taste must be changed, one them is the fact that if you need to change a scale you are forced to re select the instrument and that cuts your performance, not to mention that we mostly play in a dark environment in order to project video.

Just imagine playing live and doing weird key combinations in the dark while you take care of other aspects, do I have the proper setting on the synth? do I need to change the patch on the pedalboard, etc etc.

At the moment I´m mostly using ableton live and I route the pico to a midi input obtaining really cool sounds and a decent playability (of course I do have an M-Audio firewire interface), so I´m using the pico as a controller which I know cuts a lot of features, but at the moment the only thing that I really dislike from this instrument is the software.

MIDI is an archaic protocol I agree, but why not just have a midi output configuration and when desired the whole eigend running? A new instrument protocol is needed, there is not doubt about that. There is a lot of research to be done and it will take some time, just ask Dave Smith ;)

This is not only is a matter of the protocol itself, but also the interface... these days you can see more an more controllers using for example RJ45 connections, but I guess that´s not the answer.

At the moment OSC is something that I would really like to have in eigend, I strongly believe that this is the secret behind the success of Monome.

Take into account that we are using something that is so new that these issues are in a way to be expected and also that we are witnessing the very first steps on what I believe is a musical revolution.

The near future will bring professional eigenharpists and even music created for the instrument.

As far as I know it is possible for other midi kit to control certain EigenD settings, it just requires them to offer a software factory setup with that stuff enabled (eg footswitch to change scale).

Better midi out to use the Pico as a controller is a good way to bridge the gap whilst waiting for more instruments that are eigen-aware. There is already some midi support for this and I dont think this is a bad thing.

I think the idea of teaming up with a sequencer/DAW company has its flaws - namely that there are quite a number of DAWs in widespread use and making the Pico work loads better with one in particular may put off users of other DAWs, Plus Im stuggling to think of many ways that the Pico needs much tighter DAW integration, beyond the issue of getting audio from EigenD instruments & loops into the DAW, which should be solved by future AU version of the software. There are some AU features that not many hosts support, such as midi out, but in general I think flaws in AU support that may have been mentioned are more to do with software instruments not being written to handle certain Pico/Alpha features such as independent pitch-bend for each key, rather than the AU standard being a big problem.

OSC has its uses, its no magic fix. Im sure it could handle all of the keypress etc info the Pico sends, and receive messages to set the LEDs, and as the monome has shown this would help open the Pico up to a world where people are coming up with their own software which defines what the hardware can do, and OSC is just used to communicate between this software and the hardware. But there very well could be quantity of data/timing/sync issues which lead Eigenlabs to use their own protocol for communication between hardware and software. A third way is possible, where the EigenD software still talks to the Pico/Alpha in the established way, but the software can listen to and send OSC as well. There are some scenarios where this may be useful, but again it doesnt solve anything on its own, people still have to write programs that use OSC with the pico to do something clever. Most OSC aware apps/max patches that exist now are for use with the monome, jazzmutant lemur, multitouch or other control surfaces that dont have that much in common with the pico, different paradigms (such as a grid of buttons) that dont really fit the pico. I would like OSC out from the pico so that I can visualise the state of every key & control, but this is a niche feature that does not belong at the top of the list of things to do.

OSC might have a use when it comes to trying to get other software instruments to play with the pico to its full potential. Fudging the comms using midi is possible for something with not too many keys like the pico, eg keypresses are sent as standard midi notes and then every direction of every key could be sent via different midi cc messages, which the pico-aware soft instrument can use to do individual note pitchbend etc. But with something like the Alpha you would probably start to run out of midi cc's, so this isnt a great botch. Alternatively Eigenlabs may want to work directly qwith specific instrument developers to make their apps pico-aware and communicate data from the pico to the instrument soft using a proprietary protocol. This is fine if Eigenlabs can get many partners on board, and if the Eigen instruments remain the only hardware which has a notion of individual pitchbend, modulation etc for each key. But thinking ahead, if this paradign takes route and becomes a norm, then you probably want a new open standard for communicating this data between apps and hardware. Thats where OSC could have a use, someone just has to settle for a spec on what the OSC messages look like (eg /instrument/key no pres yaw roll where pres, yaw and roll are floats containing info on the full state of the key no). I cant imagine such things happening right now, any impetus towards this would require a sizeable number of pico & alpha users to seed the idea that independent pitch bend etc is a great way forward for instruments.

So returning to the present, I think the priorities should be stability, increased flexibility for doing stuff with midi in & out messages, and an AU plugin that removes the audio routing issues some people have. It seems like these are all being worked towards, at which point broader issues can be looked at again. Need to walk before we can run, and I think Ive seen a few comments on the net that suggest some people think the Eigen instruments are a solution in search of a problem - certainly the EigenD software presently creates some new problems, but only time will tell if the masses come to see what the Eigen approach offers. Im perfectly happy to put up wth the flaws because the sensitivity & multi-directional capability of the keys is so brilliant. If boatloads of people come to the same conclusion, then a great foundation is formed to really move the future of instruments on, and other parts of the musical software equation are going to want to come to the party, and all will be great. If this does not happen then we'll remain fully reliant on eigenlabs to provide all the solutions via their own software & instruments, and users like me will be hoping that eigenlabs would add some form of OSC support so that I could hack together my own solutions.