Moog Music Theremini Reviews

Very nice review Synthguy, thanks!

The information we have regarding control over pitch quantization seems to be a bit conflicted, but your description is very clear.  Perhaps people are confusing "sticky" tuner behavior with pitch behavior?

Kind of strange you can't save edited patches via the front panel.  I wonder if they don't have any EEPROM in there?  Flash is often a bit scarier to write to, and can't support as many writes.  (I'm hoping FRAM or similar really takes over and soon.)

Sticky tuner behavior...good description! I think that this is what is confusing the issue. The tuner only reports the notes of the selected scale, with the deviation bar showing the difference to the next reported note, but if the quantization knob is turned all the way down, the pitches are continuously produced.

I was using the Peterson Stomp Classic to look at pitch, which claims .1 cent resolution. My crappy technique certainly can't hold any pitch ANYWHERE near that close, but by moving farther away at a lower pitch and anchoring myself to a chair, I was able to freeze and move the display smoothly back and forth. So, how accurate? Looks pretty good, but I'll try to see later if I can get a better idea at higher frequencies. I usually keep the quantization knob at about the 12 to 3 o'clock position, which gets me on pitch, but still lets me add vibrato, although my technique is pretty poor ( but improving slowly).

BTW, the Theremin preset is saved, at least on my unit, with the quantization all the way down, but with the Ionian scale selected, so that scale is imposed when the knob is turned up. Not sure if it was that way originally, I might have altered and saved it with my Midi experiments. Eh.

A few of the presets make use of the built in filter, using the volume antenna to shape the filter as well for wah effects, and Animoog wavetables can have the scan rate modified by volume or pitch antennas as well, which can be interesting.

Not sure what is transmitted, haven't gotten around to playing with that yet, but I imagine certain data is 2- way to allow for interactive knob updating for an editor, if we're lucky.

"I was using the Peterson Stomp Classic to look at pitch, which claims .1 cent resolution. My crappy technique certainly can't hold any pitch ANYWHERE near that close, but by moving farther away at a lower pitch and anchoring myself to a chair, I was able to freeze and move the display smoothly back and forth. So, how accurate? Looks pretty good, but I'll try to see later if I can get a better idea at higher frequencies."  - Synthguy

Sticky pitch can be very elusive.  Depending on the mechanism, you might find it anchored at a particular pitch or physical location, or it might pop up during certain kinds of movement.  The easiest and most foolproof way I've found to detect it is to record the audio of smooth arm movements in and out, and then examine it using the "spectral frequency" view in Adobe Audition (this is of my early digital Theremin prototype):

If you want to record a quick MP3 of you smoothly moving your hand nearer then farther from the pitch antenna, and doing so at different speeds, I could easily analyze it for you and post the results here.  Any stickiness should stick out (ha ha) like a sore thumb.

Thanks Synthguy for the feedback!

I would have to experience this myself to be totally convinced, but I think there will not be any detectible zippering. If the Animoog engine is the heart of the sound creation in the Theremini, then I imagine it would be quite easy to clean up noise by implementing glide.  The area of focus would then shift from sound quality to playability, because latency would be introduced by any masking done by the algorithm.  The length of time of a single oscillation of vibrato at approximately 5-6Hz would require a very low degree of latency without totally ruining the realtime dynamic capacitive experience of theremin playing. 

As a general comment, I think it's important to note (if it is not already obvious) that people who have played the theremin before now, have expectations that need to be met before any stamp of approval is given.  People who are owning a theremin for the first time, in this case if it's the Theremini, will very likely be satisfied with what they are presented. Without expectations or requirements that need to be met, it's more than acceptable from the new player's point of view to hear the great sounds and feel like they are receiving a high quality product. Neither the veteran theremin player or new theremin player can have the same experience with the new product because their needs are very different.

"I think there will not be any detectible zippering. If the Animoog engine is the heart of the sound creation in the Theremini, then I imagine it would be quite easy to clean up noise by implementing glide." - Randy

Thanks for mentioning this important point, Randy -

The issue with this sort of "smoothing" (effectively a form of portamento) is that it can still only operate with the data it is given.

If (and this is just an example, and may be completely irrelevant) the data increments in 10 cent steps, it would be possible to glide between these steps in a musical manner and "hide" any "zippering" - but it would not be possible to "settle" on any interval between these steps - the playing would still be constrained to some microtonal intervals, even if the glide isn't.

In practice, unless these steps are extremely large, it is likely that this trick will be acceptable for many players - most synth "emulations" of theremins rely on portamento - and many (most?) people couldn't differentiate between a "real" theremin and some keyboard pretending to be a theremin.. Those "into" theremins however can instantly tell the difference!

In practice, micro-portamento could be almost good enough - If the intervals are reasonably close together (resolution is low but not appalling) and if a good predictive engine is employed to dynamically adjust the portamento time constant to match the playing (rate of change) and if the overall latency is low.. A lot depends on how it is implemented.. Sensitive vibrato is the thing most likely to fall foul of any smoothing - it will (must) impose a distortion of the vibrato 'shape' and depth - so at best one is triggering set points and the engine is attempting to produce a facsimile of these movements - the lower the resolution, the poorer this facsimile will be. 

All the above is, however, pure speculation - this is the most frustrating aspect of this thread - we are depending on subjective evaluations*, when all we need is a few specifications that are in the Moog Theremini technical reference folder. I have absolutely no doubt that some of the Moog technical team are reading the posts here - they could clear up all the issues in one post! (The fact that they have not commented anywhere and / or refuted potentially damaging speculation leads me to think the product might even be worse that we are guessing - I know if I had designed something and people were trashing it based on false information / speculation, I would be jumping into the thread giving data to correct matters)

Its starting to look like the only way of really finding the true resolution will be with a MIDI analyzer - this data wont (I guess / assume) be processed by the Animoog engine, so one should be able to see the raw data the Animoog sees.. Anyone owning a Theremini should be able to do this using some analyzer (something like this) and setting the theremini to transmit 14 bit data - sweep the whole range slowly, record the data stream, and all should be revealed!

(it still wont answer questions about latency etc)

Fred.

*I am not implying that subjective evaluations aren't useful - In the end its all "subjective" - one either likes a sound or feel of an instrument, and it either serves ones requirements, or it doesn't  .. But in many cases, particularly for those who are thinking about buying a product and know what they want / need, a full technical specification helps to eliminate products which are just going to waste their time - Its a bit like buying a computer - If you know you need 4GB of Ram and a 3GHZ processor for some application, you wouldn't buy a "revolutionary" new computer if the manufacturer refused to disclose its speed or memory! - Even if it was 2/3 the price of a standard computer which had the specification you need.. Or you wouldn't unless you were a risk taker!

Here you have a video, one shoot without editing, i guess that some trolls are waiting for something like that, let's going to see what happen, i just want to share some features of this instrument with you all, my theremin-family, thank you.

https://www.youtube.com/watch?v=o6x-DsWjl5A&feature=youtu.be

Thanks for the video robonil!

Unfortunately the audio is ambient rather than direct out, so analysis with Audition is fairly inconclusive:

You can see from the timeline that this is around the 1:30 mark in the video.  Too much room bounce and noise and such to get a clear view.  When recording for analysis I do line-out to line-in.  Waveforms with a lot of harmonics seem to work best as they give you lots of sharp lines in the spectral plot.

When you do the octave thing it appears the octaves get increasingly cramped the closer you get to the antenna.  Is that true?  If so, is it a problem for you, and is there any adjustment you can do to improve it?

Thank you for posting the video, robonil.

The pitch correction doesn't seem to help much as a learning tool, does it. :-)

As I recall (and here is my standard disclaimer - I am not a melodic thereminist, just repeating what I have heard skilled classical thereminists say) the two big advantages of the RCA were good linearity and a limited pitch range (3.5 octaves?) giving good spacing between semitones. This suggests to me two things other than turning on the pitch correction which would be good for melodic players. 1. Set the highest note further away from the pitch rod - the range is very compressed near the rod. 2. Use a limited pitch range, say three octaves rather than seven.

Many thanks for that demonstration, Roboni !

Far, far better than I had expected! - Why the hell has Moog not done any demos like this ?!

Playing any theremin with 7 octaves coverage is difficult IMO! - But as GordonC implies, pitch correction really isnt impressive! - To me, when turned up, it just sounds like a nasty toy.. and worsens the playing rather than improve it!

But when turned off, it does sound like a theremin is being played - couldnt get that from a keyboard with portamento!

And when turned off, it really does seem to be turned off!

Fred.

(I have inserted your video here to make it easier to view)

oh.. about trolling ;-) I must just qualify my comments by saying that I really dont like that particular theremin sound - But I have heard worse.. Have heard worse from some analogue theremins!  Also, by the looks of it, one could adjust this sound quite dramatically if you wanted, something most analogue theremins have limited ability to do.

Great demo, Roboni!

Very brave of you to be the first to actually show some of what the instrument is capable of. THANK YOU!!!

It will take time for us to realize just exactly what the potential of this new performance instrument is. Any new set of musical tools will need time for those talented and daring individuals to find a way to incorporate them into their musical vision.  

For instance, if you really see the pitch correction as a "nasty toy", then that potentially creative avenue might be closed to you. As I've been experimenting with the Theremini, I've started to realize just how deep this thing is, and how much control the good folks at Moog have given us in such an early realization of this technology.

Myself, I really like the quantization, and after messing around with it for a while, I realized that there are a number of parts of it that are accessible via Midi (and then the lightbulb went on). Using a Midi controller pedal ( I use the old Digitech PMC 10, but there are many others with similar capabilities) I started creating setups where I could control the quantization continuously using a volume pedal, and found that this allowed me to really concentrate on the antennas, and define the pitches more or less sharply using the pedal, or just bring the quantization in for a particular passage or effect IN REAL TIME. 

Also accessible via Midi is the defined musical scale, the root note for the tuning, and the low and high notes produced, which allows you to press a switch, send the appropriate Midi data, and update these things on the fly during the performance, so you can actually carefully mold the ranges you want to use at any given time while you concentrate on the antennas. There are lots of other parameters that can be instantaneously modified as well, making the possibilites for expression expand exponentially. And this is all before we even touch a preset (which can of course be accessed via Midi as well); these are just parameters that can be applied to the currently selected preset, which makes the fact that there are only 32 presets available a bit more tolerable. You can even shape the Animoog engine tones produced via Midi, so the concept of presets, is a way, almost becomes unnecessary with the proper Midi controls. If you send the proper Midi data, you could essentially create a complete preset from scratch!

So, as a new tool, I think that there are more than enough possibilities to keep new explorers busy and creative for quite a while (I can certainly see I've got a LOT of work ahead of me...). Now, I realize that there are probably a lot of people here that may not be thinking in the broader Midi possibilities that this instrument presents, but I'm hoping that I might be giving some of you some ideas about what could be accomplished here. In the bigger picture I see that you could even have the Theremini parameters under sequencer control (for those with larger setups or those who might already be using backing tracks). This would allow for complete parameter control linked to the musical score. Now we're gettin' somewhere!

Last night I was watching a Youtube video of an early ELP performance of the song "Knife Edge" where Keith used the early modular Moog to mostly make swooping glides, burbles and whistles for the solo. Sometimes the sound would cut out while he was changing patch cords, and then he brought out the the ribbon controller to make, well, more swoops and glides. It took time for him to imagine, create and realize his vision of this new instrument to make the impressive music we would begin to hear a short while later, and which have eventually become a part of this instrument's lexicon.

The takeway here; using the Theremini from the front panel is nice, but not much immediate control. Tapping into the many tools available through Midi gives a MUCH larger spectrum to work with. Like any good instrument, the biggest variable would be your imagination, and how you see using the available "knobs" to shape YOUR musical vision. Although this is a Theremin, it can also be much, much more than that.

What possibilities do YOU see?

On a side note, is it me, or does the volume antenna not make the most sense? I would much prefer to have the volume increase as I move my hand closer to the antenna, kind of like using a volume pedal, and not the other way around.

On my list of things to try is creating a patch in Midi to accomplish this somehow. Can it be done? Maybe!