Software vs. Analog in 2025 – Has the Balance Shifted?

[wikter]

They still cannot properly replicate Stradivari’s work to this day.

It's weird to hear a stradivari's violin from the player's position.

[whassup]

[pdxindy]

Here's an example of a bonkers plugin that no hardware maker would likely make... maybe in Eurorack

Yeah, Eurorack is where hardware matches software for the wide range of "hey what if we did this" ideas implemented

I’d love for a developer to make a new standard of modular. No cables. A rack system that would plug into a central brain that came in various sizes. All mod connections would be made and stored in the brain via a mod matrix, everything done at very high refresh rates. Touch a button, then touch a source control, then a destination and adjust to taste, sort of how the Prophet 12 works, including being able to modulate modulation amounts in the matrix.

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

[zerocrossing]

Here's an example of a bonkers plugin that no hardware maker would likely make... maybe in Eurorack

Yeah, Eurorack is where hardware matches software for the wide range of "hey what if we did this" ideas implemented

I’d love for a developer to make a new standard of modular. No cables. A rack system that would plug into a central brain that came in various sizes. All mod connections would be made and stored in the brain via a mod matrix, everything done at very high refresh rates. Touch a button, then touch a source control, then a destination and adjust to taste, sort of how the Prophet 12 works, including being able to modulate modulation amounts in the matrix.

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

You could easily maintain quality, if you did everything at very high refresh rates. I mean, this kind of thing has to be happening in digital modules now, if they sound as good as you claim, eh? The benefits are huge. Full recall, and no mess of wires that you can no longer easily tell what’s going on. It would ultimately be more like a highly customizable fixed architecture synthesizer with modular style modulation. Hot.

[wintoid]

thats how we got the bass, some ham fisted chippy, tried to knock up a viola, and voila, the bass was born!

10 points for using voila and viola in the same sentence!

[SamDi]

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

Why? You could have the module connections analog controlled by digital.

[_leras]

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

Why? You could have the module connections analog controlled by digital.

If the switching is even possible and the multiple analog single connectivity could be done it could be quite interesting. A modular with no wires or front panel connectors and just a patch bay. Interesting concept.

[IvyBirds]

Here's an example of a bonkers plugin that no hardware maker would likely make... maybe in Eurorack

Yeah, Eurorack is where hardware matches software for the wide range of "hey what if we did this" ideas implemented

I’d love for a developer to make a new standard of modular. No cables. A rack system that would plug into a central brain that came in various sizes. All mod connections would be made and stored in the brain via a mod matrix, everything done at very high refresh rates. Touch a button, then touch a source control, then a destination and adjust to taste, sort of how the Prophet 12 works, including being able to modulate modulation amounts in the matrix.

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

you would be able to do it all in the analog realm rather easily using similar technology to what was used in electromechanical telephone exchanges for 100 years

All you are doing is linking audio circuits like we're used in the microphones and speakers on the handsets of analog phones, with DC circuits for the CV like was used in analog phones where 48 volts of phantom power was used to power the phone itself and the 90 volts used to make the ringer ring

You would be able to have the modules on a small thin rack on a desk/wall/rack and have the switchboard in a separate box you could have on the floor with cabling running between them

Then have a very simple low cost computer like a Raspberry Pi run and control the relays and switches. Pretty much everything would be cheap off the shelf parts. You could even have the raspberry pi in the switch board and control it wirelessly with a phone or tablet via Bluetooth or WiFi, or just have a small touch screen module on the rack

Not sure how much of a market there would be for it, as a lot of modular users like the spaghetti and plugging things into each other

One thing that is quite fun however is using VCV rack or Cardinal in Windows on a large touchscreen monitor and using your fingers to draw the connections

[jamcat]

thats how we got the bass, some ham fisted chippy, tried to knock up a viola, and voila, the bass was born!

Well actually, the contrabass (or "double bass," as it were), evolved from the violone during the late 17th and early 18th centuries.

[whyterabbyt]

That would require all module connections to be digital.

Sorta right, but it technically wouldnt, as you could use analog matrix switch ICs. You'll actually find Eurorack modules with small programmable or button-controlled matrices done this way (and also modules with every other matrix variation including pin matrices, and matrix mixers.)

However to provide the any-output-connects-to-any-input equivalent of cable patching, you basically have to somehow have a matrix that is (numberOfInputs x numberofOutputs) in size via some sort of bus of (numberOfInputs + numberofOutputs) channels.
That gets too large too quickly (same as old-school pin matrices dont get much larger than 32x32 per panel).

So digital is possibly a more pragmatic option, but although it allows bandwidth-saving by only passing 'active' connection data, rather than having to connect every source to every output all the time, obviously it has limits too; bandwidth is finite. (And obviously in a hardware modular the expectation is that control signals can be audio-rate)

[SamDi]

That would require all module connections to be digital. The analog nature of Eurorack would be gone, along with the amazing sound quality.

I'm not seeing an advantage (and plenty of disadvantages) over just using a software modular plugin like we already have.

Why? You could have the module connections analog controlled by digital.

If the switching is even possible and the multiple analog single connectivity could be done it could be quite interesting. A modular with no wires or front panel connectors and just a patch bay. Interesting concept.

Switching is what a transistor basically does. The challenge would be, that for m + n connections (output and input) you would need like m x n matrix elements of switching points, if you want full connectivity. You'd need to develop a dedicated analog IC for that. Interesting idea, but who would invest for building that and primarily who would pay for that? I guess it wouldn't come cheap.

[Gamma-UT]

On the m * n problem, you'd do it like oldskool ProTools hardware handled DSP or aircraft handle communications: a time-sliced bus or ring. Everything gets sliced up into packets and multiplexed onto a small number of rings or buses. At a high enough clock rate, you can multiplex a lot of "digital CV" and probably save a bit of bandwidth using a publish-subscribe protocol where each lump of samples gets sent once and any modules that want to see it just pick them up as they pass through.

It feels like a solution in search of a problem though. You've got to agree on a format and sample rate that's high enough to avoid control-signal aliasing and do it cheaply enough to sling into different for...what? Insanely complicated sequencing patches that might be better off being implemented in Max or Pd and passed through an I/O module into the rack for the sound generation part?

[Albert.VST]

In the past there were numerous manufacturers of small to very large routing systems in all sorts of formats, from analogue audio and video to digital video, AES, MADI and even IP. Up to 2048x2048 and beyond. I’m not sure the analogue crosspoint chips that were used are still manufactured or there is A/D and D/A conversion on input and output modules of modern systems but the technology is there. Ok, maybe expensive. An example module

https://evertz.com/products/EMR-AA

[pdxindy]

That would require all module connections to be digital.

Sorta right, but it technically wouldnt, as you could use analog matrix switch ICs. You'll actually find Eurorack modules with small programmable or button-controlled matrices done this way (and also modules with every other matrix variation including pin matrices, and matrix mixers.)

However to provide the any-output-connects-to-any-input equivalent of cable patching, you basically have to somehow have a matrix that is (numberOfInputs x numberofOutputs) in size via some sort of bus of (numberOfInputs + numberofOutputs) channels.
That gets too large too quickly (same as old-school pin matrices dont get much larger than 32x32 per panel).

Yeah, it's technically possible, but a 120x120 analog matrix would only cover a modest system and anyone who bought it would run right back to patch cables.

[pdxindy]

Switching is what a transistor basically does. The challenge would be, that for m + n connections (output and input) you would need like m x n matrix elements of switching points, if you want full connectivity. You'd need to develop a dedicated analog IC for that. Interesting idea, but who would invest for building that and primarily who would pay for that? I guess it wouldn't come cheap.

I'm supposed to receive my Reliq sometime in the next month (finally). It has a 16x16 analog mixer that can recall connection setups and swap between them glitch free in realtime (either manually or sequenced).

But even a 160x160 analog matrix wouldn't cover all my current modules. And that would actually be less immediate than what we have now as we would look at our case of modules and have no visual indication of what is wired to what.

The advantage of such a system is that it would reveal how great our current method of using patch cables is!