Moog Music Theremini Reviews

"I imagine that opening the theremini and photographing its guts (let alone putting scope or other probes into them) might just invalidate the warranty (?)  ;-)"  - FredM

Not sure I'll probe it, inspecting the components and waving a frequency counter around the general vicinity will likely be enough to sufficiently understand things.  I'll have to play with it for a while before opening it as it is a return unit.  If the patient doesn't make it through surgery it's totally on me, but you make a good point.  If ESD or something nails it later and I can't repair it then I might be perusing the warranty with a fine-toothed comb.

=> It's in Fort Wayne, IN (~600 mi away) with an estimated delivery of Tuesday - woo hoo!

Theremini arrived via FedEx about 1/2 hour ago.  Looks to be in great shape for a return.  I plugged it in and ran through the setup and presets, but haven't spent quality time with it.  Took some unboxing pics that I'll post later along with some observations.  Unfortunately I'm smack dab in the middle of leaving-tomorrow-for-vacation last minute rush.

First impressions: the speaker doesn't get very loud at all.  I noticed the speaker "bass" ports go all the way through to the bottom of the unit.  The pitch antenna gets really scratched up by the contacts when you stick it in the socket, the knobs feel a bit wobbly, the tuner is likely of limited use, pitch correction could probably be just on / off with no middle ground (and use the dedicated knob for something more important like calibration).  Pitch field seems a lot less linear than the EWS.  Except for the preset knob, the other knobs are pots, so when you switch to another preset you have to move the pot a bit for the current knob position to be meaningful (a pet peeve of mine - just use encoders already!).  The encoder is a 12 click/rotation type so the detents are kind of spaced out.  Not trying to trash it, just keeping it real.  More later.

Theremini Unboxing / Disrobing Pix

Pix from my uploaded gallery here.  Right click and open in separate browser window for higher resolution.

Figure 01.  The box.  The dealer I bought it from actually double boxed it, which is always nice (except for the trees).

Figure 02.  The quick start guide and a promo for winning free stuff if you make a musical submission to Moog Inc. (and somehow manage to win).

Figure 03.  That's the power supply box on the left, the power cord on the right.

Figure 04.  A catalog of Moog products on heavy weight newspaper.

Figure 05.  The Theremini going for the Heaven's Gate look.

Figure 06.  The power supply.  12V @ 2A and doesn't weight much.  Gotta be a switcher.

Figure 07.  The bottom where we see the antenna storage and labels and such.  This one is dated "7 14" which is last month.  Not shown just out of the picture L&R are 4x exposed phillips head screws.  Also not shown are 4x phillips head screws which are under the rubber feet.  The feet have a long molded in rubber piece that fits down the hole, but they are also weakly glued in.  You have to remove all 8 screws to open the Theremini, but you don't have to snap it apart or fish around for other fastening methods or anything difficult after that.

Figure 08.  Here you can see the deep scratches the antenna electrical contacts make on the antenna when you stick it in the hole and when you remove it.  That's gotta hurt for the guys in the returns department, and it doesn't make me all that happy either.

Figure 09.  The Theremini opened up and with L&R swapped.  The main board down front has the processor, power connector, and all the back I/O.  The control board is above.  The pitch board is on the left.  The volume antenna on the right is a surprisingly substantial plate inside the case.

Figure 10.  A bit better view of the cabling between the control and main boards.

Figure 11.  The main board.  At the upper right are L2 & L3, two inductors in series for the volume antenna.  I think they are 103's (10mH) like on the pitch board.  Not sure what L22 (470) on the left side is doing though it might have something to do with the pitch side.  I'll go through the other components when I have the time.

Figure 12.  The pitch board.  L3 is a single 103 (10mH) ferrite inductor with zero ohm resistor R3 installed to bypass the L4 position.  A small L2 is in series with the antenna.  U1 is an active 6 terminal device, can't read the marking but tracing out the circuit might give some clues.  I felt around on the bottom of the board and there are no components under there.  L1 and L5 are also present, perhaps as power supply decoupling.

Holding the antenna of my frequency counter near the volume antenna I get a reading of ~213 kHz.  Holding it near the pitch antenna I see ~290 kHz.  The ratio 290 / 213 = ~1.36 which is very close to the square root of 2, which isn't too surprising for a 2:1 L ratio in the pitch and volume side inductances.

High res pix here: http://www.mediafire.com/download/sod93aqq5999pog/Theremini_pix.zip

Ye Gods!  Dewster, those are great pictures!

4 battery clips for the antenna! LOL ;-) Makes the Nano look like craftmanship! - Truly horrible!! - and is that hot-melt adhesive on the 3 wires to the pitch antenna board?

Anyway, we now know theres an oscillator on this board, +V GND and output on the wires, U1 could be a dual BJT, looks like L3 is part of a parallel resonant circuit including the antenna via a tiny HF rejecting ferrite (L2), no ESD protection (with such a long GND lead any ESD components on the board would be a waste anyway), so whatever U1 is it, and the passives, are at risk... 3 year warranty? I wouldnt take that risk! - perhaps, with luck, if you only play it once every few months.. or perhaps they think its price is low enough that people will just throw it away when it fails.

You say the linearity felt worse than the EWS, looking at the AFE that wouldnt surprise me at all! - I guess they just shoved a crude oscillator on the unit and thought they could get away with digital linearization.. Looks as if nothing remotely approaching your attention to detail with regard to getting the best from the front-end has even been looked at.

And unless theyve located some inductors with rather special ferrite, I see nothing to mitigate thermal drift.

So far it looks like just about every AFE presented here by digital theremin designers/builders is probably as good as or better than that nasty little board!

Sorry - I really dont want to be critical - The main board looks serious - but to me the antenna board looks like part of a cheap toy.

There is a lot in it for the money - makes the price of the EWS seem unjustifiable. I suppose costs had to be cut somewhere. Perhaps there will be a market for up-market replacement pitch antenna boards...

Fred.

Added -> I cannot see an analogue reference oscillator, nor any trimmable inductors.. I therefore suspect the reference is digitally generated and tuned to match whatever frequency the analogue oscillators happen to output.

"Ye Gods!  Dewster, those are great pictures!"  - FredM

I added a link above to the native res pix from my camera.

"4 battery clips for the antenna!"

Is that what those are... hmm.

"You say the linearity felt worse than the EWS, looking at the AFE that wouldnt surprise me at all! - I guess they just shoved a crude oscillator on the unit and thought they could get away with digital linearization.."

It's possible I'm doing something wrong with the setup (though I'm following the on-screen instructions) but the linearity is the worst I've ever experienced (not that I have a ton of experience with other Theremins).  The pitch field is useable for playing tunes and such for maybe a foot out, after that it's so stretched nothing much happens.  (To me the totally amazing thing about a good Theremin is getting good pitch change way far away from the antenna.)  So maybe it's not too surprising we're not seeing a million videos of people playing tunes on these things?

And it gets worse, the Theremini has sticky pitch:

You can see it above quite clearly and hear it here: http://www.mediafire.com/listen/yjydd6exab7b1dg/Theremini_20114-08-06.mp3

This is the first patch "01 CLASSIC THRMN" with the pitch correction and effects knobs turned all the way counterclockwise, and the min / max pitches set to C1 / D8 (the max the Theremini will do).  I'm holding my arm out and sweeping it in a semicircle by twisting my body so that my hand and the pitch antenna are closest about 1/2 way through the sweep, so as to produce the smoothest change I can.  If I sweep my arm quickly the Theremini responds with a noticeably smaller pitch change than when I sweep my arm slowly, so they are likely filtering in there way below 1kHz.

I shouldn't jump the gun as I've only had a chance to play with it for an hour or so, but so far:

Pros: The voices and effects are fun, the control panel is nicely angled, the built-in speaker is handy, and the price is right.

Cons: Everything else you might expect from a good Theremin seems to be MIA, but it could be pilot error I suppose.

The calibration could be a lot more interactive, or at least give some kind of feedback other than a number on the screen counting down the seconds until the cal is done.  There is one step I don't quite get either.  Here is how the whole thing goes (with on-screen instructions):

1. Press SETUP button until you get to the CALIBRATION screen.

2. Press EFFECT button.

3. Press SETUP and move away 4', wait for 5 sec countdown.

4. Stand at arm's length and press SETUP, but no countdown happens, just the next screen.

5. Press SETUP and place hand near pitch antenna, wait for 3 sec countdown.

6. Press SETUP and place hand far from pitch antenna, wait for 3 sec countdown.

7. Press SETUP and place hand near volume antenna, wait for 3 sec countdown.

8. Press SETUP and place hand far from volume antenna, wait for 3 sec countdown.

I suppose in step 4 they are just asking you to get back into playing position, but it's a little confusing.

====

For 290kHz and 10mH, I calculate 30pF for simple LC resonance on the pitch side oscillator.

====

Gotta hit the hay, probably won't be posting much in the next few weeks, but will try to check in if/when I can.

[EDIT] Just tested the power supply brick with a DMM: there is a short between the outer 12V barrel contact and the AC ground (which is how they establish the operating ground for the Theremin circuitry).

Thanks Dewster, those high-res pictures are great. The antenna connectors to me looked identical to battery clips (C or D cell) I have used, but close examination and the metal on the theremini's clips seems thicker, so they could be bespoke parts - fairly sure though that at least the idea came from battery clips.

Your tests seem to show awful linearity, pitch sticking, and long latency. Linearity could possibly be improved with a better antenna board, but this isnt going to fix the other issues.. The only people who will want better linearity will be those bothered by the other issues as well.

Everyone seems to have gone silent about CV / Pitch tracking, which isnt a good sign (I imagine a Moog distributor would come back quick if they found CV did track pitch at 1V/Octave).

So it looks like all serious users (particularly those who want it for "tunes" not "effects") will be disappointed and/or find the instrument of little or no use.. Thereminists and Synthesists - for these two major groups in Moog's customer base, the theremini isn't going to be the 'bridge' between these two worlds it could have been, or satisfy the requirements of either world.

For the cottage industry theremin developers this is good news! ;-) ... Moog have taken the low road, mopped up the toy and gimmick market - and left the high road clear for "us" - Yeah, they've got the biggest slice of the total market, (some cottage industry theremin developers - Nano for example- could nibble into this and compete on price ..particularly if their performance on matters such as linearity is better than the theremini, and it seems that it would be hard to achieve anything worse.. ;-)

But the market for playable theremins, and the market for theremins that interface with synthesizers (and particularly the booming CV modular market), is, AFAICS, untouched by the theremini.

Fred.

"Just tested the power supply brick with a DMM: there is a short between the outer 12V barrel contact and the AC ground (which is how they establish the operating ground for the Theremin circuitry)." - Dewster

Yeah! - and they've put a grounding screw on the back! Big + to Moog for at least doing that! - Shows perhaps that some of the posts here are observed ;-) .... Mind you, it could just have been added in desperation when their horrible AFE prototype only worked over perhaps 1' due to low antenna voltage combined with no player grounding...

Looking at your MP3..

IF you are doing a full sweep (getting as close to the antenna / providing the same max capacitance on each sweep) then the latency is just awful! - If we look at the peak frequency 'round 6.55s, we get the fundamental at about 4kHz in a ~1 sec sweep, this drops to about 1kHz (@14.65s) for a sweep of about 0.3s.

It is real difficult (particularly with an extremely non-linear theremin) to accurately sweep ones hand to a consistent distance - but even allowing for this, we are talking about 2 octaves error! - even if we allowed 1 octave error due to hand 'misplacement' in a cramped non-linear section of the field close to the antenna (and such non-linearity  would be dreadful.. looking at the sweep / peak @ 9.5s, there is indication that a 1 octave error could be due to non-linearity, as the sweep is just a bit shorter than the sweep at 6.65s, but the maximum fundamental frequency is ~2kHz) this still leaves a 1 octave error due to latency - this must equate to latency in the 10s of ms - perhaps even up to 100ms.

If all that effort they put into "pitch correction" and visual tuning had been invested instead on making the damn instrument playable - linear with low latency - they might have produced a musical instrument.

But if the performance I see and hear is an accurate representation of the theremini's musical capabilities, then its probably the worst most unplayable baguette of rubbish ever to have appeared on this sad planet.

I strongly suspect the resolution is low, and had been 'overcome' by a "permanent portamento" (integrator) with TC of 10ms or more - on a rapid change of hand position, this 'portamento' doesnt have time to catch up to the hand before the hand goes back to its home position.. On the above one can see that, when going from low frequency to a position which should at least produce >2kHz, in 150ms it can only attain 1kHz, this severely limits the instruments playability - If it takes ~300ms to go from the lowest to highest pitch (as seems to be the case) then for all practical musical purposes the instrument is utterly useless for anything except drones.

"IF you are doing a full sweep (getting as close to the antenna / providing the same max capacitance on each sweep) then the latency is just awful!"  - FredM

Nice charts!  My sweeps weren't 100% scientific, and I had to do them without being able to hear the audio as my PC was hogging the headphone jack, but yes, that kind of latency behavior was pretty obvious when I could hear what I was doing.  There's no way you could do anything even vaguely percussive on it but you could likely play regular stuff (if you don't mind staying up close and personal with the pitch antenna).  The stickiness of the pitch could be an incomplete masking of the pitch correction, but more likely it is a numeric / resolution problem in the software or with front-end data acquisition.

The direct approach (who knows what's going on in the Theremini without more probing and analysis) requires all the sensitivity you can muster from the oscillator along with a high sampling frequency, and even if you do everything right the far field is still somewhat marginal for a professional Theremin, though it can be "fixed" to a large degree through filtering, but then you start seriously trading off responsiveness.  The offset heterodyning method is IMO the only real way to go for a pro digital Theremin.  The digital offerings lately are giving the digital approach a black eye, I'm wondering if soon there will be negative demand for them.

Thanks Dewster for the detailed pics, analysis and walk through!

 I don't have very high hopes for this version of the instrument (assuming there will still be a 'pro' upgrade option), but I'll still have to play the thing for myself to be sure.  I've yet to get an ETA on when I can receive the instrument I'm borrowing... sometime in the next couple of weeks hopefully.

Seems the topology of the pitch and volume oscillators are the same:


The U1 is probably a dual inverter like the SN74LVC2G04.

The following is a short review of the Theremini from thereminist Wilco Botermans - reposted here to Thereminworld with permission.

Wilco Botermans writes:

Today I spent a bit more time with the new Theremini theremin. I wish I could say I like it, but I can't. There are several problems which prevent it from being a good theremin. The sound (timbre), the linearity, the tuning mechanism and, maybe the most annoying, the delay. With the delay I don't mean the built in delay effect, but the delay in response to the movement of the hands. I guess it takes about 15 milliseconds to respond to the players hands. It is extremely tiring when the sound happens after you've played it, as not only the pitches, but also all expression is out of time. With playing vibrato or a trill I noticed that the extreme pitches of the movement are not heard, as the hand is faster than the engine. Combined with the complete lack of linearity, it makes the response extremely untrustworthy.

About the sound, well, it sounds like a digital representation of an etherwave theremin in a fixed and unchangable setting. And that is of course what it is. But more annoying because an analogue sound is alive. Not as much as an acoustic instument maybe, but the theremini sounds dead.

The tuning is just one long annoying button pushing screen staring activity. And if a mistake is made, or the environment changes, you have to start again. No quick small adjustment of a big knob thigh where you want it, and very visible, but buttons that need to be pushed and make the very light theremini shake on its stand.

The autotune does not make up for this, as with autotune, the machine decides when the new note is heard, not the player. Vibrato, tone bends, and other things that make a theremin a theremin are taken away from it. It may be easier to play 'the right note' in some instances, but it takes away all personality of the individual player.

For me, playing a theremin means to be in direct contact with the sound, with the music. No keys, buttons, strings or anything else stands between me and the sound I hear. This is the truest, sound interface. The purest contact with sound. The digital engine of the Theremini destroys this completely.

I did have some fun playing with the animoog sound presets, but after hearing them all, I heard them all. I prefer hooking up my Etherwave PLUS to a Slim Phatty and have the SONIC INFINITY that Moog Music really does offer.

I have heard the remark that it is not for what is called precision players (players that only play classical melodies) but that it could be marvelous for experimenters. Well, after 17 years playing both melodies and using the theremin in all sorts of experimental setups, both analogue and digital, my opinion is that what goes for precision players also goes for experimentalists. They, or at least, I, also want to rely on a direct response and reliable pitch arc when I am drawing an air castle near the pitch antenna and process the sound in 6 different sound paths, or use it to pick up the movement of a dancer to process sonic accompaniments.

So I am not happy with the Theremini. Not for myself, as a thereminist in the largest sense of the word, and what is more, not for the theremin community. Because of the lower price, I think many potential theremin players will prefer to buy the cheaper Theremini, but they will not experience what a theremin truly is, and quickly set it aside as a toy for Halloween or at a party, but they will not consider buying a ''real' theremin afterwards.

Lastly, I very much wish Moog Music would not present the Theremini as a serious instrument, for the new generation of theremin players. The world has enough conformity as it is. And listening to the Clara Rockmore CDs as they suggest in the manual does not help, because even she wouldn't be able to play a theremini like that.