Gyratec G21 Magneto-Dynamic Infundibulum review

[Note: I have written an addendum/update to this review, about a year later. You can find it here.]

I haven’t exactly been asked to write a review of the Gyratec G21 Magneto-Dynamic Infundibulum, but there have been a lot of questions about it from interested people. It made sense to write a review that I, or Jakob at Gyratec, can direct people to for more information on the unit in use. So here it is.

As I write this I’ve had the unit about 6 months, using it on a daily basis in my mastering room. By way of an abstract, I’ll start by saying that it has become a firm favourite in my studio. It is utterly unique and it has its idiosyncrasies, but it is extremely useful, sounds great, and has a lot more versatility than the ‘passive clipper’ description suggests. I’m a big, big fan.

I’m not going to do the Sound on Sound description of all the knobs and so on; I’m going to talk more specifically about how I think of the unit, and how I use it. Anyone interested in the inner workings of the unit (and it really is an incredible design) should take themselves over to the excellent Sound on Sound review. I may cover some of the same ground here, I haven’t read it in a while and I’m writing this on the hoof, so apologies if I do. The same applies to the manual, I may duplicate some of that. Apologies.

Right then.


The G21 has a high pass filter, a tilt switch, a full band clipper and three frequency-selectable band clippers. These siphon off a portion of the signal to be clipped, filtered, and passed back to the output.

After that, probably the first thing to say is that the G21 is totally passive; there’s no power required at all. There’s not even a power inlet. It’s powered by the very fabric of space time itself, or something (if ‘Maxwell’s equations‘ means anything to you, you’ll love this box).

As such, once it’s in, you’ll notice a drop of level due to the expected insertion loss. This loss depends on what’s driving the unit, where it is in the chain etc (all of which also has implications for the action and sound of the box, of which more later). The unit’s bypass leaves a load in line, so the sound level doesn’t jump up when switching it in and out, enabling you to make a fair comparison.

When you do take it out of bypass (with no clippers engaged), the signal passes through the transformers, you smile, and you quietly thank your chosen deity for all that is good in the world.

Now I admit that I get a bit romantic at moments like this. But this unit sounds absolutely glorious. Immediately obvious depth and width, with a seriously classy character.

Stupid comparison alert: I’ve often described the G14 eq (a tube unit) as sounding ‘golden’ – to me it sounds like a warm summer’s late afternoon sunshine. Well, the G21 (not a tube unit) sounds like Vega, or possibly Rigel. It has a brilliant, clear (not bright!) crystal quality, with a very slight blueish tinge. Stupid I know, but the first time I sat with the unit and switched it in and out, Vega popped into my head. So there you have it. But whatever comparison you make, the transformers in this thing sound the absolute business.

They know how to roll ’em at Gyratec.


The box has a selectable out/25/35Hz high pass filter. I have to say that for the music I deal with mostly (I think they’re calling it EDM this week), it’s rather heavy-handed, or obvious. I get more mileage out of the Dangerous BAX high pass filter, which is much more delicate. Even at the closest comparable setting (24Hz), the BAX rolls off without sounding like it’s rolling off. But, when something really does have a ton too much low low end, the G21 filter does sound very nice indeed.

Engineers working on less sub-centric music may find their mileage varies though, it has to be said. But it’s a high pass filter, so there’s not much more to say here.


It’s worth discussing the meters a bit. They are a beautiful thing to look at, but in truth, they don’t really mean a great deal. Or anything at all really. That’s not to say they are pointless though. In fact I’d say they are more important to the internal functioning of the unit than they are useful for getting any useful information about what’s happening. I’ll try to explain how I see it.

When the unit clips a signal and the distortion gets filtered out, that energy has to go somewhere. This energy drives the meters, so, no power required for those pretty dials. Genius. But it’s the fact that the meters give the energy somewhere to go other than to audible distortion, that’s important – not whether the meter moves much or not. That’s my take on it anyway.

In use, the meters move in mysterious ways. They show current being removed, not dB. So with a low input signal you may have to press your eyeball against the glass to detect any movement (put your coffee down first), and with a higher input signal they move a lot – but both could be doing the exact same amount of work on the signal. I’ve also found that when driven by electronically balanced units, the meters move about even with the clippers completely disengaged. Madness. Why does this happen? Nobody knows, not even the designer. It serves as another reminder that this box is something totally different to what you’re used to.

In any case, I’ve been assured that if the clippers are out, they are out – I’ve not heard anything and apparently no effects are measurable, so it’s just another reason to close your eyes and open your ears.


I reckon the thing most people ask about the G21 is ‘yeah yeah, magnodibbly infundiwhatever, but what does it actually do?’.

Simply put, it makes dynamic stuff that pokes out, disappear. Often completely invisibly. This could be a ringy 808 hi hat at 6kHz, that flappy rubbish in the kick drum at 135Hz, overall transient wildness. You can just move them back.

In no particular order, here are some things I’ve used and regularly use this box to do;

– removing aggressive plectrums in heavy rock at about 3.3kHz. You don’t eq or compress them, hurting the guitars… The plectrums just gently disappear back into the mix.

– moving too-up-front shakers at 11kHz back into the mix. I do this a lot.

– as a kind of vocal compressor for a wild vocal performance where certain notes jump out. They don’t any more.

– using the full band clipper early in the chain to make wild transients calm down a bit (think analogue tape here, with the full band clipper).

– using the full band clipper last in the chain to take the load off the limiter. This combination is very handy if you need ‘loud’ but hate that unnatural, flat limited sound.

– using the mid band in low pass mode to tame a jumpy low end.

– using the mid band in band pass at 270Hz (this setting is quite broad) very gently. If there is a better ‘auto demuddy-fier’ than this in the world, I need to know about it now. It’s absolutely incredible. It’s so invisible that you have to be extremely careful.

– possibly the most often used function here: the low band between 95-200Hz to fight tubby, dominant kicks caused by the poor implementation of sidechaining. I really cannot stress the following enough: In combination with something the Elysia Nvelope (review pending), you have the ability to work with a track in ways that have hitherto been impossible. You can re-eq individual elements in a track on an as-needed basis in an almost completely transparent way. You can shift energy away from problem areas on an as-needed basis, and put the energy back in a better balanced way. You are, essentially, able to dynamically re-eq the track and/or elements within it on an only-as-needed basis, without messing with the integrity of the track. No crossovers, no time constants. Often it does all this so invisibly that it’s not until you take it out, that you realise you were grossly overdoing it.

[A note as an engineer here: some of this sounds very invasive for mastering, but sometimes it’s necessary. I get a lot of bedroom mixes. So there.]


‘But this box is 3000 quid and I can do all this with a dynamic eq or multiband plugin. Are you having a laugh?’

This is a key point. It just doesn’t sound anything like any of those options. At all. Sometimes, of course, that means it isn’t the right tool. But doing the ‘same’ thing with the best other options available, set up as closely as possible – and believe me, I’ve tried – just doesn’t sound even remotely the same. It’s an infinite-knee band-specific compressor. It’s a stupidly musically-aware dynamic eq. It’s both. It’s neither.

Again; no crossovers, no time constants. Because it is controlled by the signal itself, not listening and responding to it, it ‘follows’ the music absolutely perfectly. And it doesn’t meddle with the phase integrity or do anything at all, in fact, in the time domain. No crossovers, no time constants. You can hear that stuff.

A couple of expanded examples;

You have a mix with a really scratchy shaker. It has a real excess at, say, 12.4kHz. You eq it out. That’s fine, but it detracts from everything else. So you try a tight band compressor/de-esser. You fiddle with the threshold, amount of reduction etc and get a balance of all the settings so that it’s OK. In many cases you might end up with a bit of eq and a bit of de-essing. It does the job.

Now you approach it with the G21. You find the irritating frequency (the frequencies are very well chosen), set the band to clip mode and simply turn the shaker down in the mix. That’s exactly what it feels – and sounds – like. Like turning the shaker down in the mix. It’s genuinely mind-blowing . I’ve sat there turning the level knob up and down with my mouth hanging open, listening to shakers go forward and backwards in the mix.

Similarly for tubby, over-compressed kicks in electronic music. Eq that flappy rubbish between 110-160 out, the whole tune gets thin. Multiband it out and it usually sounds a bit better down there, but it can still suck the balls out (I might rewrite this bit). G21 it out, it’s like the kick was better eq’ed in the mix. Just two days ago I had an ultra tubby kick with loads at 200Hz, it sounded rubbish. I set up the G21 and again, sat there listening to just that part of the kick go forwards and backwards in the mix, like I was eqing it. Ridiculous. I’m a jaded engineer these days, but that stuff still excites me.


I’ve mentioned once or twice about how you can unknowingly overdo things. Sometimes though, you need to be a bit more assertive, and make compensations elsewhere. The range switch allows for this. Basically, you can use it to tilt the mix a bit. If you’re being forced to control the low end but it changes the tone somewhat, you can switch from red to pink, or pink to white, and it re-tilts the mix. This can be very handy when you need to change things without sounding like you’re changing things.

This also opens up extra sonic options with the full band clipper. Each ‘threshold’ is a different clipping material, and therefore each has a distinct sound. As you’d expect, generally the lower the threshold the softer/smoother (more ‘tapey’) the sound. But in conjunction with the range, each threshold now has three sonic flavours. Sometimes it’s a toss-up between the highest threshold with the white setting, and the next one down with the pink. So, more options.

You can also get radically different sonic results placing it elsewhere in the chain, a result of different units feeding and being fed by the G21, as well as effectively having different thresholds and therefore different amounts of processing occurring.

Perhaps the last thing I want to mention is how the parallel design works in use. I don’t mind admitting that it took me a bit of time to get this straight in my head, but it’s actually fairly simple.

All the clippers – including the full band clipper – operate in parallel. I had been thinking about each of the three band clippers operating on their bit of the signal, all combining and feeding the full band clipper. Sort of seems logical. But no, they’re all in parallel.

When considering this in use, it’s essential to remember there’s only so much energy to go around. This becomes quite useful. If a certain band feels or sounds like it’s working too hard (sometimes it has to), you can share the load and take off some of the strain by giving the overall signal a little nibble with the full band clipper. By soaking up a little energy across the whole spectrum, you don’t have to work so hard in a specific area, and can thus keep things more transparent. You can balance between the two to get the most effective and transparent solution. Many hands make light work.

This doesn’t come up often, but it is useful, and worth mentioning. Again, I don’t know of anything else that can do stuff like this, and while it sounds complex it feels incredibly organic and natural in use.


It’s not all sweetness and light. This is, after all, a clipper.

You’ll note that the front panel has ‘Use with caution’ on the front. So, I’m going to talk for a moment about the more… ‘dangerous’ aspects of the unit.

Make no mistake, this is a distortion unit – albeit so brilliantly designed you often don’t know it. But sometimes, maybe in 5-10% of cases, it reminds you in no uncertain terms that it is a monster underneath. Certain sounds, certain tracks, will simply scream with distortion. It’s really very nice screaming distortion, from a creative/artistic point of view. I’m sure it would come in very handy for sound design and the like. But I tend to try and avoid that sort of thing on clients’ masters.

In this regard, I tend to think of it a bit like a woman (please note that this is intended as a lighthearted stereotype rather than being deliberately misogynistic!). That is, it will be absolutely fine the majority of the time, and then go completely bonkers and scream at you for apparently no reason. There’s no point arguing, there’s no point trying to reason with it, you just have to accept that it isn’t happy about something, and move on.

In all seriousness though, I’ve found that it’s generally pure/resonant tones that kick the unit into ‘scream’ mode. I’ve started checking very carefully when a track contains things like Rhodes keys, very resonant filter sweeps, or very prominent pure tones. It’s not all such cases, though – I should definitely stress that. It’s not like it goes mad at the suggestion of a Rhodes, but… it might.

So, as the front panel suggests, use with caution. I’ve considered this ‘negative’ side of the unit a lot, and it’s almost got me into trouble once or twice when something has nearly slipped through QC. But in truth, it’s not so different from any other unit. With a more run-of-the-mill unit, you might switch it into the chain and it changes the sound a little bit; could be better, could be worse. You might sit there and flick for a while and decide ‘hmm, maybe it’s best to leave it out’. With the G21, it screams its displeasure, you take the point, and that’s the end of that.

On a similar but less extreme note, you can think of this as a highly tweakable tape device, and sometimes in quieter parts you can get a little ‘hair’ on things. In such cases I’ll switch out the offending band for that section, and splice it in after. I mention this a) because while I love the box it does sometimes require ‘special treatment’, b) I want to be honest and share all the information I can, and c) because even with things like these, the unit has become so often essential to the best result that it’s worth working around.


I’ll try to summarise. I’m aware how lengthy this review is, but this just isn’t like anything you’ve used. It’s not another flavour of eq; it’s not another compressor, that does the same thing as a million other boxes in a slightly different way. It is, in every way, something else.

This is a great, unique, exciting box to work with. It sounds great, and as I continue to get to know the unit better, I’m getting to know when it won’t sound great. In that way it’s no different to any other box; it’s just not subtle about it. It’s idiosyncratic, it’s demanding of special attention that a more typical unit just doesn’t require. It’s beautiful, it’s terrifying.

Above all though, it’s just incredibly useful, and there’s no way it’s leaving my studio. I can 100% say that it has allowed me to do things that previously would have been impossible, or at best nowhere near as good. I’ve tried to explain these things in this review, but I’d recommend trying it yourself.

It’s improved the quality of my work, and when it comes down to it, that’s what a piece of professional equipment must do if it is to stay in the rack – no matter how quirky/and unusual it is.

  1. #1 by Farhad K. DadyBurjor on June 15, 2014 - 5:12 am

    Bob, this is the best, most honest, and well written review of any piece of gear I have ever read. Ever. I would so love to see your reviews in more, um, mainstream magazines. I’ve been curious about the G21 since it was announced and not having access to trying it out have been scouring the Net for reviews that could help. Yours nails it. Thank you for writing, and looking forward to reading many more such reviews.

    • #2 by bobmaccsblog on June 15, 2014 - 8:29 am

      Glad you enjoyed it. Every time I use I think, ‘I should have mentioned this, I should have mentioned that’. But I got most of what I wanted to into the review. Cheers!

  2. #3 by kris keogh on July 8, 2014 - 6:56 am

    Yep, great review. You’ve hit the musical practicality vs tech talk ratio right out of the ballpark. If only all gear reviews were this thoughtful and useful. Much respect.

  3. #4 by Suade on July 22, 2015 - 4:19 pm

    That’s a really useful review Bob. One of the best I’ve ever read because it conveys so much meaningful experience. Thanks for sharing your thoughts.

    • #5 by bobmaccsblog on July 22, 2015 - 4:58 pm

      My pleasure – I keep meaning to do an addendum to this review, I still deeply, deeply love the box and use it every day. I really don’t get why it doesn’t get more press!

  4. #6 by Olivier Zahm on September 1, 2015 - 10:03 pm

    This review has gotten my piggy bank to subscribe itself on a mission 🙂

  5. #7 by Peter on October 7, 2015 - 8:25 am

    Superb review! Thanks!

  1. Unashamed Gyraf fanboy post | Bob Macc's Blog

Leave a Reply to Peter Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: