Moog Music Theremini Reviews

Thanks so much Cristian!  The video can tell us a lot about the instrument.  I'm also unsure why Moog Music would not want to make basic demonstration videos similar to this format. 

When you set the pitch range to cover as many octaves as you have demonstrated,  it does seem to make it more difficult to play.  With such small intervals, the Pitch Correction feature seems to not be as useful when playing melodies. The note that you want to be 'in-tune' may technically be in tune, but the steps in between are still audible and the smooth transitions between notes (which we are accustomed to with traditional theremins) are of course not there at all.

Cristian, could you perhaps record a video of the pitch field when it has been calibrated for only 3 to 4 octaves, so we can hear what the stepping/locking sounds like at 0% and 50% of the pitch correction?

The experience of playing any theremin will always be unique to every person depending on a number of factors. player body type/size, theremin tuning, environment.  These details are nothing new to those of us who play the instrument, but the new first time theremin player... not having a firm understanding of all the details affected by capacitance, then adding pitch quantization and elaborate tuning/calibration processes, equals a sure fire guarantee for more confusion. Any one hoping to become a better theremin player by playing a Theremini can kiss their dreams goodbye.

The playability of the instrument should be the highest priority of ANY musical instrument manufacturer. In the case of the Theremini, it seems the sound engine was most important. It seems as if the theremin control was "added on" to the Animoog Sound Engine.  (again, these are impressions... I need to play the instrument to be sure.) Bending the expressive control interface of a theremin around an existing MIDI instrument is backwards to me... , but for good reason.

I won't go into much detail in this thread, but I have for the past three years been working on software that will convert audio signals to MIDI. With it, a traditional theremin could be used to control the full iOS app of Animoog running on an iPad.  A computer and Audio MIDI interface is needed, so the complexity of the setup doesn't compare at all to having a self contained all-in-one Animoog theremin (the Theremini).  It's quite possible that our friends at Moog Music are aware of my project, and their silent treatment is the result of a what they may imagine is a conflict of interests (I'm just speculating).  I've been working on my software (and a version of it has been publicly available) a lot longer than the Theremini has existed.  Plus, I am not interested at all in any sort of competition. 

Back to the video...

As far as I am aware, from my experience with the prototype, the reason for the pitch not changing all the way up to the antenna has to do with how the range is derived from the control signal.   There is a starting point (or upper limit) in the control signal's range which determines where the highest pitch will be.  This was most likely done this way because if there exists the possibility of defining the highest MIDI note, it should always be accessible.  The only way to guarantee it is always accessible is to implement some predefined buffer distance from the antenna.

There are definitely two approaches to consider for the Theremini. One is using it as a creative tool, to explore new control possibilities with the access to its powerful sound engine opened up via MIDI.... The other approach is to use the Theremini as a traditional theremin, to play melodies and derive some satisfaction from the dynamic nature of the capacitive interface.

It seems to me, and it should be obvious to everyone by now, the melodic approach is not going to be successful without some kicking and screaming along the way.  The new and powerful feature set that the Animoog engine offers has caused more challenges in other aspects of the instrument, but these deficiencies are not apparent to people who are buying a theremin for the first time.  It's not even important for the new buyer, because if they have decided they are are pleased with sounds of the product, that is the bottom line. Not everyone can be pleased all of the time... it seems Moog Music decided to go with features what made most business sense to please the target audience: synth players (for creativity) and children for (for some kind of fun musical scale lesson).

"For instance, if you really see the pitch correction as a "nasty toy", then that potentially creative avenue might be closed to you. " - Synthguy

Yeah - Its not so much about "see the pitch correction as a "nasty toy" its that, to my ears, it sounds like a "nasty toy".... But this is just my subjective reaction, others may think it sounds wonderful!

For me, its not the concept of pitch correction which is the problem - its the implementation, or more specifically, the implementation on a theremin - I think that pitch correction turns the theremin into something other than a theremin (or at least it sounds like that to me - on this implementation of pitch correction anyway)

And so far, I haven't liked anything I  have heard from this "other instrument" - But as I said, that's my personal subjective "taste".

Then there is a big question in my mind about what use pitch correction on a theremin  really has even if it sounded ok to me - I dont see that it can improve on a keyboard! A Synthesizer with keyboard is easier to play, doesn't have blippy unintended interim notes, can have portamento to facilitate whatever degree of glide one wants, and one can add vibrato and/or modulate pitch and volume on wheels or ribbons or pads.

IMO (apart from possibly the "heterodyning sound" which is debatable and the theremini doesn't produce anyway) the theremin has ONLY one advantage as a controller / instrument, and that's its infinite freedom for the player to control every minutia of pitch and volume.

Throw this away, and IMO one really is throwing the theremin away! - ok - the fact that the theremini seems to retain this in theremin mode (no pitch correction) qualifies it IMO (which aint worth a damn ;-) as a theremin - but turn pitch correction on and IMO its a different, and IMO extremely inferior instrument.

And I cant see pitch correction being much help as a learning aid either.

Having said the above, I agree that some musicians may well manage to use the instrument, with pitch correction, in musical ways - That even if what they produce wasnt "musical" to my ears, it may be "musical" to others .. Musical taste is a strange thing! ;-) .... I dont even like the music produced by Peter Pringle using his Ether Vox to drive MIDI - The wash of tidy harmonies churned out by the algorithm just bores me - But I love everything else he does.. I also love GordonC's "Beat Frequency" stuff, and am kind of worried that if he gets a theremini his music will lose its grim enharmonic madness!

Fred.

I am not a musician or thereminist, but it looks to me like the major technical problem with the kind of pitch correction implemented on the theremini is the pass-through notes.. On rapid passages played on a conventional theremin, the (skilled) player is able to move quickly from one interval to another, no "emphasis" is applied to the intervals between the "source" and "target" interval.

With the theremini's correction, any and every interval (in the selected scale) between "source" and "target" is played and, because its made "distinctive" by the pitch correction, produces (to my ears extremely, intolerably annoying) "blips" - The only way the player would have of getting rid of these 'blips' would be by "pumping" the volume antenna - And I doubt that this will produce acceptable results, and suspect it will limit playing to stuff at quite an elementary level and slow pace.

I was working on "pitch correction" some years back - entirely different scheme - the pitch itself was unaltered and always infinitely adjustable, but the volume was increased (to an adjustable degree) as one got closer to a "valid" interval.. Say it was set to chromatic, the volume would go down between the notes, and be at the level set by the volume antenna only when "on" the note - one could set the "sharpness" of the attenuation between notes etc..

The above (IMO) had the advantage that one would never accidentally switch notes when doing vibrato etc - but it still had the same problem that all notes between "source" and "target" were emphasized - I called this scheme "On Key Emphasis" - But sadly, it could never do "On INTENDED Key Emphasis" because, until we develop a telepathic interface, the theremin cannot know what key you want to play next!

I heard the annoying "blips" on my crude knock-together proof of concept rig (I used a CV quantizer and some synth modules to crudely test the idea) and promptly abandoned that idea (and the telepathic interface idea because I realized the defense industry would steal it even if I did develop it, LOL ;-) (only joking on the latter by the way - I have absolutely no idea at all about how to implement a telepathic interface.. )-8

Synthguy. The volume field as it is permits one to snap one's hand away from the side of the loop to make a staccato note without needing to be cautious about hitting the volume loop.

I have some ideas about using MIDI to make an arpeggiator, and to sweep filters over the waveform very slowly.

Fred. I thought it sounded alright on Dorit's demo. 

I am hoping it will increase the amount of madness, and perhaps occasionally contrast the grimness with something less enharmonic. Who knows? Mostly I buy stuff with a strong idea of what I want to do with it, and then discover that what I imagined it did is not really what it does, but what it does do is good too. :-)

 I think On Key Emphasis would be better. 

In talking with the people at Moog, it seems like they have a lot of plans for development of the firmware, but are short on engineering resources, so things will certainly be happening here, but slowly. Good things, like Midi note number transmission, and other fun and useful stuff. The word is that there is still lots of memory left for future upgrades. I personally applaud the decision to get it working and get it into the hands of those who are really interested in this new technology...and future upgrades are always welcome.

Currently, the system only operates on Midi channel 1, but this will likely change.

Another first impressions review

--> http://www.muffwiggler.com/forum/viewtopic.php?p=1583197#1583197

"Another first impressions review" - GordonC

Thanks!

Wherein we are told the Theremini:

1. Needs to warm up a bit before the analog section and tuner are stable.
2. Has CV outs, but they don't do quantized 1V/octave.
3. Can't be used as a MIDI controller.
4. Delay effect is mediocre.
5. Has a great tuner, and every Theremin should have one.
6. Volume control doesn't control the line-out level.
7. Isn't a serious Theremin.

#1 suggests they are using ferrite coils or other temperature sensitive component(s) in the oscillator(s).  And I wonder what "unstable" here means?

If true, I'm kind of surprised at #2.

I don't understand #3.  Are they complaining about lack of DIN connectors?  Because I'm pretty sure the Theremini can transmit both 7 and 14 bit pitch and volume MIDI information (though as MIDI over USB).

#7 is fairly damning.  Moog likely targeted the dabbler market with the Theremini, and may have upscale offerings in the works, but did they have to hobble it so bad that it isn't a serious instrument?

I'd like it if someone made a video showing how large the pitch field can be made.  Everything I've seen so far is rather small and cramped near the antenna, and if it could be made larger then perhaps the linearity in the far field is better.

Also, Moog may very well release a software upgrade for the Theremini which addresses the various shortcomings, but buying one now and counting on them to do so at some point in the near future is a recipe for madness.  SW development takes time and is expensive, and so a SW upgrade might very well never happen - which would be a shame because it seems that it is mainly poor SW that is holding it back.  What they should do is release an SDK for it and let enthusiasts take the reins (but I feel this way about most SW centric products, particularly musical ones).

I've never noticed the unit needing any time to warm up, or display any signs of being unstable. Not sure what's being referred to here.

Both antennas are transmitted as Midi CCs  (2 and 20 IIRC). Makes a very nice Midi controller. On the Ipad, I've patched them into a variety of destinations on different synths.

I haven't played with the CV outs, but like any CV source, they can be scaled however you like. My guess is these are designed more for general purpose controller duty targeted for the Moogerfoogers. These are probably able to be rescaled via firmware.

The delay effect has 3 set ranges (short, medium and long) from the front panel with just an effects level adjustment, but is completely adjustable via Midi. (Delay time, feedback and level) The delay time can be more than a second, IIRC, and is stereo.

The front panel volume does only control the speaker and headphone jack. However, there are two Midi volumes available...one for preset level matching, and a global for overall level control. If the front panel control is analog, then it is what it is. If its digital, then it might be reassignable in the future. I'd like to see that happen.

The review doesn't say it's not a serious theremin, in fact just the opposite. ?

I'm not counting on software updates, but after several conversations with the people at the factory, I feel fairly confident that there will be some interesting developments in the future. Patience.

In the meantime, plenty is controllable right now, you've just gotta actually get in there and do it.

Hi Dewster,

As I understand it..

1. Needs to warm up a bit before the analog section and tuner are stable.

 Its an EW front-end (probably modified to produce offset heterodyne frequency). As such, its thermal behavior will be like an EW - and it will use ferrite, both for EQ and tank inductors.

2. Has CV outs, but they don't do quantized 1V/octave.

Yeah - this is obscure.. I interpret it as that the CV is unquantized (as in, pitch correction doesnt operate on it) . I suspect the CV is derived directly from the heterodyning, as is MIDI... As with MIDI (next) I believe that if Moog couldnt provide a selection of modes for this without extra expense, they made the right choice in selecting this mode of operation - I suspect they had little choice - I suspect that the pitch data from the front end is processed early (linearized / scaled or whatever), and this resulting data is packeted for MIDI and output via D/A tor the CV output.. I think this data then goes to the quantizer and voice.

3. Can't be used as a MIDI controller.

I suspect that to the reviewer, if Note-on / note off MIDI messages arent transmitted, to them MIDI isnt transmitted. To me, this is just silly - Its a goddamn theremin not a keyboard!

IMO, Moog got this right!  IF the theremini transmits >= 12 bits of CC pitch data, then that should be all you need. It should be (will be) simple to take this CC into a software quantizer and produce note-on/off if one wanted (perhaps this is what Moog plans to do) and even to combine pitch and volume CC's to produce running-status MIDI note messages combining them as note data with embedded volume data that could be read by any MIDI instrument or software synth.

The above scheme seems, to me, to be a good way of getting the required data from a theremin reasonably quickly (if one mixes CC and note messages, one bottlenecks the stream) ... The quantity of data (say 1 packet volume and 2 packets pitch) streaming continuously is just on the edge of acceptable in terms of speed.

I do understand the reviewers objection - MIDI has been mainly seen as a way to connect keyboards to expanders, and the assumption that one should be able to plug your theremini into your D110 or whatever, and have it control the D110, is understandable / justified to some extent IMO.. But in the wider context, MIDI is (or was for its time) a comprehensive communications protocol - You wouldnt expect to be able to play a D110 using any other (non keyboard) MIDI controller (like the Korg Nanokontrol) designed to send CC messages.. Once again, I think its primarily Moogs fault for not clearly specifying the instrument - They make it look like something that it isnt, they mislead people by saying things like "it uses a heterodyning oscillator"  - perhaps they dont tell lies, but I suspect they will pay a high price in returned units for not being really up-front .. They do say, in small print, that "Note data is not yet supported" - But really, they need to say loud and clear that the MIDI doesn't yet drive any standard external MIDI equipment.

Also, a big let-down for the modular synth crowd must be the lack of separate CV outs for pitch and volume - With both, you could wire the theremini up and drive VCO VCF VCA to create a custom theremin with your rack.. But without both, you are limited to pitch only.

But hey - it is cheap! - If it was £500 these comments would be more valid.. And despite my critical assessments, for its price it does seem to have a lot of potential -

Fred.

More research is needed concerning the CV out. In the manual, one of the things listed in the menu for it is to apply pitch correction to the CV, which would imply that it is scaled. Hmmm. I don't currently have anything around to test this, but I'll report the facts once I can check it out. The CV can be assigned to either the pitch or the volume antenna. Also found in this menu is the capability to assign the pitch and volume antennas to any Midi CCs you want. The default on mine was CC 2 and 20. You can also select 7 and 14 bit resolution here. This is a global setting. I bet that this stuff is also accessible via Midi (seems likely) but is currently undocumented. 

The quantization being controlled via pedal works very well. You can go from pure Theremin with no pitch detection ( just pure glissando) to just note to note of the selected scale, and anywhere in between in real time just by using the pedal. This, in conjunction with selecting the range, scale, and root note makes the daunting task of defining a usable workspace for a melodic passage much more manageable.

Regarding the tone of a standard Theremin, what wave form is actually produced by one? I've always thought that it was a sine wave. The Theremini produces sine, triangle, Etherwave, Supersaw, and 3 different Animoog wave files. 

Two of the wavefiles are dynamically scannable by various means, Animoog 1, and oddly enough, Etherwave.

"Regarding the tone of a standard Theremin, what wave form is actually produced by one? I've always thought that it was a sine wave." - Synthguy.

Yeah - Lots of folks think the theremin produces sine waves.. Perhaps its because the original Tannerin on "Good Vibrations" produced pure sine waves, and this was mistakenly called a theremin, and the idea stuck.

On the classic analogue theremins (particularly Lev's tube ones) the waveform can be really complex and frequency (pitch) dependent - describing the shape doesn't do much to describe the sound, as (on the classics) its source waveform has both odd and even harmonics which are 'shaped' by resonances in the instrument - The source waveform harmonic content varies with pitch, but the resonances are static (and complex) which, on some instruments, imparts vocal and/or other qualities with similarities to the human voice or bowed strings.

Modern instruments tend to have less complex harmonics and virtually no static resonances, but (apart from those which use true multipliers, like the S/C) dont tend to produce sine waves -

The EW uses a diode mixer, and the core waveform (which is deliberately distorted by a LM13700 OTA) is similar to a rectified sine wave (this is an oversimplification) and contains odd and even harmonics. I personally dont like the EW (or perhaps the 'core' harmonic distribution of sound from a diode mixer) much - but it can be adjusted on the EW to be acceptable to my ears - but again, this comes down to taste. Personally, I think that to get a theremin that sounds like a classic theremin, the waveform needs to change as the pitch changes - and not just in the way one can achieve by using a VCF or single filter (I might even find that the theremini can be adjusted to produce a sound I like more than the EW).

The following is a bit OT:

I also think that the addition of a buffer to the EW prior to the mixer changes and improves the sound -I suspect that this may be due to a slight "error" in the design of these, whereby the signal from these buffers actually drives the mixer "harder" causing slight modification to the waveform. (as well as providing its intended function of buffering the oscillators which improves bass response)

But this is Wave Crafting.. IMO, something of a lost art - IMO, the thing which gives the early theremins their "organic" quality was the "inferiority" of the components used - The developer was forced to select (by ear) the transformers and capacitors etc that produced the "best" sound in combination with the selected amplifier + loudspeaker - It was a complete instrument he was crafting, there was no such thing as a "flat" frequency response - so a developer like Lev Theremin crafted all aspects to get it to sound the way he wanted.

These days, we think "flat" - We try to iron out any 'kink' in the response, so that it will be 'flat' when it drives a 'flat' amplifier and 'flat' loudspeaker... Its a bit of a joke really! ;-) .... Oh, when recording acoustic instruments (including the human voice), we need 'flat' to accurately reproduce these - But neither acoustic instruments nor the human voice are 'flat' - we need 'flat' to accurately capture their "kinks" and resonances.....

Yet when we design an electronic musical instrument, we think* in terms of flat (or deliberately controlled) responses - we think "ramps" (with mathematically precise distribution of harmonics) and squares and supersaws (ramps with phase modulations) and triangles and sines - but pay little regard to the fact that in an acoustic instrument ALL of these waveforms are of little comparative importance - they are the "excitation" signal, they must contain the required harmonics (bowed instruments are excited by saw waveforms and variations thereof, which are rich in odd and even harmonics, woodwind instruments tend to resonate with predominantly odd harmonics, so triangle excitation or similar is required) - but the real character of the instrument comes from its body, its complex resonator - and the response of this most certainly isn't flat!

* I should just qualify the "we think".. I can only speak for myself, so perhaps it should be "I think" or "I thought".. This may also apply to what follows..

I am mainly talking about those involved with monophonic "synthesizer" and theremin type instruments, It seems that us "electronic music" instrument designers have (or have had) an "anti acoustic" perspective - The idea of using a synthesizer to "emulate" an acoustic instrument seems (and IMO mainly is) a bit daft.

But with the theremin its a bit different - because the theremin never was an acoustic instrument.. It can never be a perfect replacement for a human voice or a violin, but it can "add" some of these qualities to produce a unique beautiful (to my ears) tone.

And this has led me to thinking that, in the process of "going electronic" and deliberately avoiding the idea of "replicating" acoustic instruments, we (or at least I) may have thrown the baby out with the bath water.. Its not "replication" if one applies acoustic type resonances to electronic instruments.. Even if one is designing for "electronic music".  (None of these issues apply for those deliberately designing keyboards etc to replicate acoustic instruments - For them. resonances etc must be of primary importance - and  they should be factoring resonances, formants etc into their design comprehensively and at an early stage, and this is why most;y digital technology is used - there are just too many factors to synthesizing acoustic instruments realistically for analogue computation to be economic - Techniques like acoustic modeling are needed. )

Fred.