Why Shuthi-1 can't be 100% analog?

Hi all! Big day today after Ambika has been finally revealed.

I have successfully built 2 Shruthis and i will probably built Ambika as well. But even though i am a big fun of the sound character M.I. synths have, i need an answer on the title.

Which are the reasons for not having a fully analogue synth by Mutable Instruments? Is it because it was initially designed to have that unique sound or it is technically a lot more difficult?

Sorry if it sounds stupid but it has been in my head for a while.


To give some perspective-take a look at this, the Elby ASM-2. It’s a monosynth without patch storage based on a 1976 DIY design. It has 2 VCOs, 2 VCFs, 2 EGs, 2 VCAs, 2 LFOs and an S+H/Noise

The two blocks at far left are the VCOs-each one has 8 trimmers…the results sound great, but it’s a lot of work to build…

If you’re looking to get into the the diy analog game, look into going the modular route. That’s where I am now… careful tho, it’s way too addicting and get ready for a serious back log of projects. :wink: I currently have something like 20 PCB’s waiting to be stuffed. I purchased them when I saw them pop up on the Muff’s or E-M forums and thought “OMG I so need that![](”. BTW… OMG I so need to pick up the Ambika boards)

yes and it’s getting more expensive. we use an 8 track tascam. it’s great but the tapes are really not cheap. my friend mentioned that some company was going to keep them in production (again?) but i can’t recall the name at the moment

if you need only stereo you need good old tube revox

i was looking for a 16 or 24 track fostex…

I imagine they sell that stuff cheap on ebay, but setting it up and maintaining the heads etc it is a skill you need to learn if you are used to digital.

Tape was never cheap either.

no way, i’m saving money to try to get an ambika
i bet oliver’d rather get my coins than see my face (in all due respect)

For me the whole vintage synth sound is a bit of a myth (as I’ve said on here before no doubt) if you are listening to old 70s and 80s music wanting to recreate those sounds.

The way they recorded back then was different, they used lots of outboard gear, multitrack tape, bouncing etc. If you want vintage sound then use a fully vintage set up.

It’s no good to use an analogue synth and then use protools, cubase or logic.

Hey, i have (with exception to MI gear) a full vintage Setup. Im looking for a TapeMachine indeed…

Still, i assume that, while being a mutable instrument, the “little miss sunshine” synth is a 100% analog thing? :slight_smile:

Maybe take a Trip to Paris this Saturday and find out . . .

If you want analogue then buy analogue. The last big analogue polysynth released was the Alesis Andromeda, I suggest you hunt one of those down.

The big boys can design and have big custom ICs made which simplifies the product. Smaller companies have to build things from common ICs.

Moreover, the oscillator isn’t as critical to contributing to the “analogness” of a synthesizer, as a filter is. I’m not saying that analog oscillators don’t have character, but if you think about it, a patch will highlight the filter 90% of the time. Plus, as Luap said, you don’t get awesome, rich-sounding wavetables!

also i guess especially in a polyphonic design you heavily rely on highly integrated oscillators just for a reliable temp compensation over all voices. those chips are all very rare to say the least

I think i’ve got more information than i needed at first place. Thank you all for clearing this up for me.

A VCO takes 5 or 6 ICs. Unless smart software techniques are used, it needs careful manual calibration (4 or 5 trimmings on a multi-waveform VCO).

LFOs, mixers, ADSRs etc. are easy to do in the analog world (2 or 3 ICs) if you accept the following constraint: the only way to control them is through a knob directly interacting with the circuit (ie: no MIDI control, no CV control, no memory). As long as you want CV/MIDI/memories/polyphony, a LFO becomes as complex as a VCO ; an ADSR becomes as complex as a bunch of filters ; a mixer becomes as complex as an array of VCAs. This is why most “full analog” synths are mono, full of knobs, and with poor MIDI control - because that’s the easy and low part count way of doing things.

So in the end, an all analog equivalent of the Shruthi-1 in terms of modulation possibilities and MIDI control would be 4 to 5 times more complex (at least 30 ICs and of course dozens of parts around each).

Yay, the VCO myth. Being analog doesn’t translate into superior oscillator sound per se. Case in point: Aka VX-90. Sounds thin. Waldorf Q or Microwave or a whole host of other digital synths: Fat.

Sure, VCOs are nice but there’s a reason that full VCOs in Polys are rare! Complexity. Most of those full VCO polys use VCO chips (JP6/JP8/MKS80, Prophet 5, Matrix 12/Xpander, Chroma, Chroma Polaris, Akai AX/VX series, Korg Polysix etc). The rest of the polys are DCOs in some sense, i.e waveshaping a timer wave like the CEM3396 synths from Cheetah or Oberheim, or the Junos. Some are even more digital DCOs like using short samples or timers in combination.

There are very few fully discrete VCO polysynths due to the horrible cost/complexity.

Sure, some of those chip-based or discrete VCOs sound nice but there are things that are only possible using digital or hybrid schemes. In analog they would require way too much circuitry. Try to build an oscillator with independent slope control over the rising and falling edge using analog or fully discrete solutions. You’d have to compromise or use some very clever trickery. Now do it digitally…

@Luap That was a good answer. Thanks!

@fcd72 Not in any case but maybe it could be better under certain circumstances.

Because it would probably be 4 times the size/cost, and there are things you can do with digital that you can’t with analogue. Like… Wavetable synthesis.
Still, being a Hybrid, you get the best of both worlds, with little of the drawbacks of either :slight_smile: