Who is all geeked out on analog logic?

Post your patches here! Put the format(s) in the subject.

Moderators: Kent, luketeaford, Joe.

User avatar
KSS
Super Deluxe Wiggler
Posts: 6902
Joined: Mon Jan 25, 2016 7:28 am

Re: Who is all geeked out on analog logic?

Post by KSS » Wed Sep 22, 2021 7:29 pm

Dolphinwolf wrote:
Wed Sep 22, 2021 10:50 am
When I started this thread I had a preconception that was not communicated well. Min and Max were all that I considered to be analog logic. Now I understand that some people consider this term to be a misnomer and other modules are important to this concept like comparators and slew and such.
You have cut to the core in this post! There are indeed two different applied meanings for 'anaog logic or analogic. One which is truly digital-aka binary- 'underneath' analog inputs and often outputs too. The other is what you had in mind based on your post here. More akin to the fuzzy logic you mention in the next paragraph. They are truly different things and the results are different too. But fortunately they may be combined for even greater effect than used singly.
Here is the gist of my concern and curiosity. If Min and Max are functional generalizations of And and Or gates then why the hell are we not able to stack and combine these analog gates to extract all kinds of significance from musical signals?
You'll need a NOT aka inversion too. <--One reason I try to champion making and using moog style V-trig to S-trig patchcords. Add a couple diode patchcords I also regularly champoin and you have the means for both analog and binary logic.
This thing I am typing on uses a whole bunch of boolean gates to do all kinds of things, wouldn’t it stand to reason that combining analog gates could yield enormous profit?
Sure. Serge T also recognized what you've typed here about the massive utility of functional AND and OR and agreed! That's why there's a peak and trough module in his systems. Those are simple circuits. Why not make up a bunch of them and see where it leads? that may give you more and quicker answers than we could share here.
I think I just need to research fuzzy logic and see if it can be applied to music processing equipment.
Yes. This would be a useful path, I think. I would expect that looking into NLC's range of modules will also bear *much* fruit. Probably no one has ever done as deep a dive into this arena -for synths- as he.
Also patching things into other things with only the smell of rationality hovering above my head.
The ARP2500 is the queen of this. The matrixswitch patching sets up all kinds of inherent non-linear mixing, aka analogic. While you can patch it in a style of a 'patchcord' subtractive synth, there is much to be found going outside that approach! In your own synths, stop relying on the -mostly!*- outdated advice to avoid patching outputs-to-outputs and your world of sounds grows widely and immediately.

*It was mostly an outdated understanding of the electronics needs -that OPAs would be broken by this**- and the Serge module owners manuals that started this now-held-as-sacred advice. But it doesn't hold up in actiual practice outside Serge. And some Buchla too. Some early 'digital' like the circuits of the EDP Wasp. Be careful with modules based on those ancestors. But in the large, almost all other synth modules handle output-to-output patching just fine. Alan Strange was talking about it way back when his book was published, ARP was doing it with both the 2500 and 2600 patching examples and advice. Not patching passive mults freely -to avoid some inherent 'danger'- is some of the most limiting and 'stupid' advice that ever rose to supposed fact in the modern synth world.

**Besides opamps which *did* need care to prevent various problems. In the 60s. There was also the *huge* growth of TTL logic in the early 70s and especially its 'totem-pole' output structure which fed this fear of putting outputs together. Because if you do that in TTL, the outputs DO try to fight each other and that *can* lead to high current draw, heat and failure. But that doesn't really apply to most synth circuits. By the early 70s, the op amps used in synths were mostly protected and the fear of output to output no longer was based on much truth.

This 'analogic' ground was well plowed and seeded in the past, with a bountiful classic harvest of music and other results that led to some of the lust we see today for clones. But then became overgrown and thick with unruly brush in the fear brought on by the first wave of digital and the fresh memories for opamps of the elders.

It has lain fallow too long! Take up your plows and hoes! The ground is rich and with tender care it will repay and not forget evermore. ;)
Treat utility modules as stars instead of backup singers.
Treat power supplies like Rockstars instead of roadies!
Chase magic sound, not magic parts.

User avatar
Mood Organ
Veteran Wiggler
Posts: 657
Joined: Wed Dec 17, 2008 2:52 pm

Re: Who is all geeked out on analog logic?

Post by Mood Organ » Thu Sep 23, 2021 10:14 am

brandnewmath wrote:
Wed Sep 22, 2021 7:16 pm
It's not being built specifically for music, but I think this new, inexpensive analog computer might be of interest to some in this thread.
An affordable Analog Paradigm computer. That’s great. Thank you for posting.

User avatar
EATyourGUITAR
has no life
Posts: 8022
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Who is all geeked out on analog logic?

Post by EATyourGUITAR » Sat Oct 02, 2021 1:10 pm

Dolphinwolf wrote:
Wed Sep 22, 2021 10:50 am
When I started this thread I had a preconception that was not communicated well. Min and Max were all that I considered to be analog logic. Now I understand that some people consider this term to be a misnomer and other modules are important to this concept like comparators and slew and such.

Here is the gist of my concern and curiosity. If Min and Max are functional generalizations of And and Or gates then why the hell are we not able to stack and combine these analog gates to extract all kinds of significance from musical signals? This thing I am typing on uses a whole bunch of boolean gates to do all kinds of things, wouldn’t it stand to reason that combining analog gates could yield enormous profit? I think I just need to research fuzzy logic and see if it can be applied to music processing equipment.

Also patching things into other things with only the smell of rationality hovering above my head.
Analog logic is a misnomer indeed. Min and max are not in any way related to and or gates because gates are for %100 digital systems. Min max are for %100 analog inputs. Read my explanation about this earlier in the thread. Min max functions have analog comparators at the input always. Notice these are analog comparators. There are circuits that do it with diodes on many inputs simultaneously but if that was invisible to the user then the analog comparator explanation is equivalent. The reason we use diodes and opamps to do min max is because it is less parts therefore cheaper to implement. You could make a 3 input min max with at least two analog comparators and some logic gates behind it. Search for what logic gates actually means. It is for boolean mathematics. This is mirrored in the function of digital circuits. This is why we can design complex digital computer systems because we already have a fully developed area of boolean mathematics to find complex systems of logic gates that are equivalent to other different complex systems of logic gates. There is a lot more information contained in an analog signal but only if that information is of interest to the user. Anything above 20KHz is usually ignored. But also analog signals can be broad spectrum. Digital communication has a constant frequency but computers get around that problem by operating the logic gates at 5GHz. Now there is some work in analog computers using photonics that use different wavelengths of light superimposed to perform faster analog computation. New fiber optic communication has recently been invented to use the rainbow light for transmission but it must be packed in a spiral to prevent unwanted interference or decoherence.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
KSS
Super Deluxe Wiggler
Posts: 6902
Joined: Mon Jan 25, 2016 7:28 am

Re: Who is all geeked out on analog logic?

Post by KSS » Sat Oct 02, 2021 7:50 pm

@EYG. You are incorrect. Min and max do not always have input comparators. See the Serge peak and trough module schematics for examples which peform min and max.

These circuits do perform AND -n- OR. You can decide what is a one and zero -externally to the peak and trough- to test this. Rest assured that these will function as gates if you read their output levels associated to whatever your logic calls 1 and 0. They become gates based on your *external* definition of their output, not because they do or don't have comparators. Diodes -or any PN junction- might be considered a comparator of sorts but the devlish detail is that its threshold is not fixed as it would be in a digital system. Instead it changes with current, light, temperature and voltage. The removal of this variable threshold is what drove the development of early logic families.

There is no need to constrain yourself to 100% digital or 100% analog.

User avatar
metamorphmuses
Veteran Wiggler
Posts: 630
Joined: Wed Sep 12, 2018 12:02 am
Location: Oakland, California
Contact:

Re: Who is all geeked out on analog logic?

Post by metamorphmuses » Sun Oct 03, 2021 4:56 am

I have fairly limited experience with analog logic in modular synthesis, but I have been getting more interested in it.

The first module I got that has some in it is the Pittsburgh Modular Mod Tools, but (1) I got it primarily as a combination envelope / function generator and LFO, with some basic logic and S&H along for the ride and (2) I have frankly found the logic aspect of it fairly inscrutable.

However, then I got the Nonlinearcircuits Let's Splosh, which I find to be a never-ending source of fun CV and audio jumbling. Among its main functions are comparator and logic. This got me more interested in other NLC modules like the Chopper, which I got and also operates on comparator and logic. This has kept the ball rolling with NLC for me and gotten me interested in the 8-Bit Cipher, which brings logic and sample & hold together in a kind of Source of Uncertainty type affair.

It's turning out to be kind of addictive, and that's definitely a problem as I have no more room in my NLC rack unless I relocate the Rossum Assimil8or to another rack, and I am reluctant to do that because all these NLC modules play very well with the Assimil8or.

User avatar
PhineasFreak
Super Deluxe Wiggler
Posts: 2199
Joined: Sun Apr 10, 2011 3:49 am
Location: South Wales
Contact:

Re: Who is all geeked out on analog logic?

Post by PhineasFreak » Sat Oct 09, 2021 12:33 pm

ryo have recently [re]released the 2xVCX - a festival of analogue logic - dual 4 qudrant multiplier, dual vc xfader, dual vca, dual attenuator, dual vc polariser!

following soon to coplement the analogue logic range are a max/median/min, a triple comparator and dual performance latching/momentary switches.

also considering a slope detector/envelope follower.

if there's any cruicial analogue logics ive forgotten please yell at me!

User avatar
Whelm
Veteran Wiggler
Posts: 720
Joined: Fri Sep 11, 2015 7:27 pm
Location: West Coast Ish

Re: Who is all geeked out on analog logic?

Post by Whelm » Sat Oct 09, 2021 12:53 pm

PhineasFreak wrote:
Sat Oct 09, 2021 12:33 pm
ryo have recently [re]released the 2xVCX - a festival of analogue logic - dual 4 qudrant multiplier, dual vc xfader, dual vca, dual attenuator, dual vc polariser!

following soon to coplement the analogue logic range are a max/median/min, a triple comparator and dual performance latching/momentary switches.

also considering a slope detector/envelope follower.

if there's any cruicial analogue logics ive forgotten please yell at me!
:love:

I've got a 2xVCX waiting to be built. Looking forward to the max/med/min and the switches are appealing too.

What is the difference between a VC polariser and a 4 quadrant multiplier?

User avatar
PhineasFreak
Super Deluxe Wiggler
Posts: 2199
Joined: Sun Apr 10, 2011 3:49 am
Location: South Wales
Contact:

Re: Who is all geeked out on analog logic?

Post by PhineasFreak » Sat Oct 09, 2021 1:01 pm

Whelm wrote:
Sat Oct 09, 2021 12:53 pm
What is the difference between a VC polariser and a 4 quadrant multiplier?
vc polariser/4 quad mult/ring mod/bipolar vca are p much all names for the same thing when constructed using these cricuits - so i was gettting a little hyperbolic

technically you can make ring mods using transformers etc so there's differences between some of these things sometimes. 2xVCX can be all of those things in the same sense because the bipolar vca is not specifically only a vc polariser or specifically only a ring mod etc.

edit: i just finished playing with vc seq/trig xpander for the first time in a while - the ability to turn any input waveform, cv shape or modulation source into a string of output values that though stepped are cv's in a kinda continuous manner - notjust on/ooff 5v but actual varying analogue voltages means this couyld kinda sit in the analogue logic functionality land too;
by the tme it recieving feedback loops in the clocking and sending trigs from the expander to digital logic in patches it gets compex and deep real fast...

User avatar
EATyourGUITAR
has no life
Posts: 8022
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Who is all geeked out on analog logic?

Post by EATyourGUITAR » Wed Oct 13, 2021 4:27 pm

KSS wrote:
Sat Oct 02, 2021 7:50 pm
@EYG. You are incorrect. Min and max do not always have input comparators. See the Serge peak and trough module schematics for examples which peform min and max.
I already mentioned this. I said even if a min max module does not have analog comparators it is functionality equivalent in the black box sense. I mentioned that it can be done other ways such as diodes. You probably skimmed through my post. I say this because all we really need is a conceptual model. We don't need to be electrical engineers to understand how to model it mentally for the desired results. If you show someone the circuit with diodes they may not understand it at all. If you use the analog comparator sequential switch model people can immediately grok it because they probably already have a doepfer analog comparator and a doepfer sequential switch. I'm just trying to teach the concept not the electrical circuit.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
EATyourGUITAR
has no life
Posts: 8022
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Who is all geeked out on analog logic?

Post by EATyourGUITAR » Wed Oct 13, 2021 4:36 pm

Whelm wrote:
Sat Oct 09, 2021 12:53 pm
What is the difference between a VC polariser and a 4 quadrant multiplier?
Nothing. They are the same. We could get into semantics about what is technically not exactly the same but for a fast reference you can equate

Fully balanced modulator
Gilbert cell
Ring modulator
Four quadrant multiplier
VC polarizer
Bipolar VCA

Technically a ring modulator requires a diode ring and a transformer. Because of the transformer a diode ring modulator only works on AC signals.

A popular choice for these circuits is the AD633. It does not require a transformer or a diode ring.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

Post Reply

Return to “Synthesis Techniques”